Anda di halaman 1dari 803

HANDBOOK OF

EXACT
SOLUTIONS
for ORDINARY
DIFFERENTIAL
EQUATIONS
SECOND EDITION

HANDBOOK OF

EXACT
SOLUTIONS
for ORDINARY
DIFFERENTIAL
EQUATIONS
SECOND EDITION

Andrei D. Polyanin
Valentin F. Zaitsev

CHAPMAN & HALL/CRC


A CRC Press Company
Boca Raton London New York Washington, D.C.

C2972_frame_DISC Page 1 Thursday, August 29, 2002 1:45 PM

Library of Congress Cataloging-in-Publication Data


Polyanin, A. D. (Andrei Dmitrievich)
Handbook of exact solutions for ordinary differential equations / Andrei D. Polyanin,
Valentin F. Zaitsev.--2nd ed.
p. cm.
Includes bibliographical references and index.
ISBN 1-58488-297-2 (alk. paper)
1. Differential equations--Numerical solutions. I. Zaitsev, V. F. (Valentin F.)
II. Title.
QA372 .P725 2002
515.352--dc21

2002073735

This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with
permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publish
reliable data and information, but the author and the publisher cannot assume responsibility for the validity of all materials
or for the consequences of their use.
Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic or mechanical,
including photocopying, microfilming, and recording, or by any information storage or retrieval system, without prior
permission in writing from the publisher.
The consent of CRC Press LLC does not extend to copying for general distribution, for promotion, for creating new works,
or for resale. Specific permission must be obtained in writing from CRC Press LLC for such copying.
Direct all inquiries to CRC Press LLC, 2000 N.W. Corporate Blvd., Boca Raton, Florida 33431.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for
identification and explanation, without intent to infringe.

Visit the CRC Press Web site at www.crcpress.com


2003 by Chapman & Hall/CRC
No claim to original U.S. Government works
International Standard Book Number 1-58488-297-2
Library of Congress Card Number 2002073735
Printed in the United States of America 1 2 3 4 5 6 7 8 9 0
Printed on acid-free paper

CONTENTS
Authors
Foreword
Notation and Some Remarks
Introduction Some Definitions, Formulas, Methods, and Transformations
0.1. First-Order Differential Equations
0.1.1. General Concepts. The Cauchy Problem. Uniqueness and Existence Theorems
0.1.1-1. Equations solved for the derivative. General solution
0.1.1-2. The Cauchy problem. The uniqueness and existence theorems
0.1.1-3. Equations not solved for the derivative. The existence theorem
0.1.1-4. Singular solutions
0.1.1-5. Point transformations
0.1.2. Equations Solved for the Derivative. Simplest Techniques of Integration
0.1.2-1. Equations with separated or separable variables
0.1.2-2. Equation of the form  =  (  +  )
0.1.2-3. Homogeneous equations and equations reducible to them
0.1.2-4. Generalized homogeneous equations and equations reducible to them
0.1.2-5. Linear equation
0.1.2-6. Bernoulli equation
0.1.2-7. Equation of the form   = +  ( ) (
 )
0.1.2-8. Darboux equation
0.1.3. Exact Differential Equations. Integrating Factor
0.1.3-1. Exact differential equations
0.1.3-2. Integrating factor
0.1.4. Riccati Equation
0.1.4-1. General Riccati equation. Simplest integrable cases
0.1.4-2. Polynomial solutions of the Riccati equation
0.1.4-3. Use of particular solutions to construct the general solution
0.1.4-4. Some transformations
0.1.4-5. Reduction of the Riccati equation to a second-order linear equation
0.1.4-6. Reduction of the Riccati equation to the canonical form
0.1.5. Abel Equations of the First Kind
0.1.5-1. General form of Abel equations of the first kind. Simplest integrable
cases
0.1.5-2. Reduction to the canonical form. Reduction to an Abel equation of the
second kind
0.1.6. Abel Equations of the Second Kind
0.1.6-1. General form of Abel equations of the second kind. Simplest integrable
cases
0.1.6-2. Reduction to the canonical form. Reduction to an Abel equation of the
first kind
0.1.6-3. Use of particular solutions to construct self-transformations
0.1.6-4. Use of particular solutions to construct the general solution
0.1.7. Equations Not Solved for the Derivative
0.1.7-1. The method of integration by differentiation.
0.1.7-2. Equations of the form =  ( )

2003 by Chapman & Hall/CRC

Page v

0.1.7-3. Equations of the form  =  ( )


0.1.7-4. Clairauts equation =   +  (  )
0.1.7-5. Lagranges equation =   ( ) + (  )
0.1.8. Contact Transformations
0.1.8-1. General form of contact transformations
0.1.8-2. A method for the construction of contact transformations
0.1.8-3. Examples of contact transformations linear in the derivative
0.1.8-4. Examples of contact transformations nonlinear in the derivative
0.1.9. Approximate Analytic Methods for Solution of Equations
0.1.9-1. The method of successive approximations (Picard method)
0.1.9-2. The method of Taylor series expansion in the independent variable
0.1.9-3. The method of regular expansion in the small parameter
0.1.10. Numerical Integration of Differential Equations
0.1.10-1. The method of Euler polygonal lines
0.1.10-2. Single-step methods of the second-order approximation
0.1.10-3. RungeKutta method of the fourth-order approximation
0.2. Second-Order Linear Differential Equations
0.2.1. Formulas for the General Solution. Some Transformations
0.2.1-1. Homogeneous linear equations. Various representations of the general
solution
0.2.1-2. Wronskian determinant and Liouvilles formula
0.2.1-3. Reduction to the canonical form
0.2.1-4. Reduction to the Riccati equation
0.2.1-5. Nonhomogeneous linear equations. The existence theorem
0.2.1-6. Nonhomogeneous linear equations. Various representations of the
general solution
0.2.1-7. Reduction to a constant coefficient equation (a special case)
0.2.1-8. KummerLiouville transformation
0.2.2. Representation of Solutions as a Series in the Independent Variable
0.2.2-1. Equation coefficients are representable in the ordinary power series
form
0.2.2-2. Equation coefficients have poles at some point
0.2.3. Asymptotic Solutions
0.2.3-1. Equations not containing  . Leading asymptotic terms
0.2.3-2. Equations not containing  . Two-term asymptotic expansions
0.2.3-3. Equations of special form not containing  
0.2.3-4. Equations not containing  . Equation coefficients are dependent on
0.2.3-5. Equations containing 
0.2.3-6. Equations of the general form
0.2.4. Boundary Value Problems
0.2.4-1. The first, second, third, and mixed boundary value problems
0.2.4-2. Simplification of boundary conditions. Reduction of equation to the
self-adjoint form
0.2.4-3. The Greens function. Boundary value problems for nonhomogeneous
equations
0.2.4-4. Representation of the Greens function in terms of particular solutions
0.2.5. Eigenvalue Problems
0.2.5-1. The SturmLiouville problem
0.2.5-2. General properties of the SturmLiouville problem (1), (2)
0.2.5-3. Problems with boundary conditions of the first kind
0.2.5-4. Problems with boundary conditions of the second kind

2003 by Chapman & Hall/CRC

Page vi

0.2.5-5. Problems with boundary conditions of the third kind


0.2.5-6. Problems with mixed boundary conditions
0.3. Second-Order Nonlinear Differential Equations
0.3.1. Form of the General Solution. Cauchy Problem
0.3.1-1. Equations solved for the derivative. General solution
0.3.1-2. Cauchy problem. The existence and uniqueness theorem
0.3.2. Equations Admitting Reduction of Order
0.3.2-1. Equations not containing explicitly
0.3.2-2. Equations not containing  explicitly (autonomous equations)
0.3.2-3. Equations of the form  (  +  ,  ,   ) = 0
0.3.2-4. Equations of the form  ( ,    ,    ) = 0
0.3.2-5. Homogeneous equations
0.3.2-6. Generalized homogeneous equations
0.3.2-7. Equations invariant under scalingtranslation transformations
0.3.2-8. Exact second-order equations
0.3.2-9. Reduction of quasilinear equations to the normal form
0.3.3. Methods of Regular Series Expansions with Respect to the Independent Variable
or Small Parameter
0.3.3-1. Method of expansion in powers of the independent variable
0.3.3-2. Method of regular (direct) expansion in powers of the small parameter
0.3.3-3. Pade approximants
0.3.4. Perturbation Methods of Mechanics and Physics
0.3.4-1. Preliminary remarks. A summary table of basic methods
0.3.4-2. The method of scaled parameters (LindstedtPoincar e method)
0.3.4-3. Averaging method (Van der PolKrylovBogolyubov scheme)
0.3.4-4. Method of two-scale expansions (ColeKevorkian scheme)
0.3.4-5. Method of matched asymptotic expansions
0.3.5. Galerkin Method and Its Modifications (Projection Methods)
0.3.5-1. General form of an approximate solution
0.3.5-2. Galerkin method
0.3.5-3. The BubnovGalerkin method, the moment method, and the least
squares method
0.3.5-4. Collocation method
0.3.5-5. The method of partitioning the domain
0.3.5-6. The least squared error method
0.3.6. Iteration and Numerical Methods
0.3.6-1. The method of successive approximations (Cauchy problem)
0.3.6-2. The RungeKutta method (Cauchy problem)
0.3.6-3. Shooting method (boundary value problems)
0.3.6-4. Method of accelerated convergence in eigenvalue problems
0.4. Linear Equations of Arbitrary Order
0.4.1. Linear Equations with Constant Coefficients
0.4.1-1. Homogeneous linear equations
0.4.1-2. Nonhomogeneous linear equations
0.4.2. Linear Equations with Variable Coefficients
0.4.2-1. Homogeneous linear equations. Structure of the general solution
0.4.2-2. Utilization of particular solutions for reducing the order of the original
equation
0.4.2-3. Wronskian determinant and Liouville formula

2003 by Chapman & Hall/CRC

Page vii

0.4.2-4. Nonhomogeneous linear equations. Construction of the general


solution
0.4.3. Asymptotic Solutions of Linear Equations
0.4.3-1. Fourth-order linear equations
0.4.3-2. Higher-order linear equations
0.5. Nonlinear Equations of Arbitrary Order
0.5.1. Structure of the General Solution. Cauchy Problem
0.5.1-1. Equations solved for the highest derivative. General solution
0.5.1-2. The Cauchy problem. The existence and uniqueness theorem
0.5.2. Equations Admitting Reduction of Order
0.5.2-1. Equations not containing ,  ,  , (  ) explicitly
0.5.2-2. Equations not containing  explicitly (autonomous equations)
( ) 
=0
0.5.2-3. Equations of the form  +  ,  ,  , 



 )  = 0 and its
0.5.2-4. Equations of the form  ,   ,  ,  , (
generalizations
0.5.2-5. Homogeneous equations
0.5.2-6. Generalized homogeneous equations

0.5.2-7. Equations of the form    , 
 ,  
 ,  , (  )
  = 0
 ) = 0
0.5.2-8. Equations of the form      ,   ,  2   ,  ,   (
0.5.2-9. Other equations
0.5.3. A Method for Construction of Solvable Equations of General Form
0.5.3-1. Description of the method
0.5.3-2. Examples
0.6. Lie Group and Discrete-Group Methods
0.6.1. Lie Group Method. Point Transformations
0.6.1-1. Local one-parameter Lie group of transformations. Invariance
condition
0.6.1-2. Group analysis of second-order equations. Structure of an admissible
operator
0.6.1-3. Utilization of local groups for reducing the order of equations and their
integration
0.6.2. Contact Transformations. Backlund Transformations. Formal Operators.
Factorization Principle
0.6.2-1. Contact transformations
0.6.2-2. Backlund transformations. Formal operators and nonlocal variables
0.6.2-3. Factorization principle
0.6.3. First Integrals (Conservation Laws)
0.6.4. Discrete-Group Method. Point Transformations
0.6.5. Discrete-Group Method. The Method of RF-Pairs
1. First-Order Differential Equations
1.1. Simplest Equations with Arbitrary Functions Integrable in Closed Form
1.1.1. Equations of the Form  =  ( )
1.1.2. Equations of the Form  =  ( )
1.1.3. Separable Equations  =  ( ) ( )
1.1.4. Linear Equation ( ) =  1 ( ) +  0 ( )
1.1.5. Bernoulli Equation ( ) =  1 ( ) +   ( ) 
1.1.6. Homogeneous Equation  =  (
 )

2003 by Chapman & Hall/CRC

Page viii

1.2. Riccati Equation ( )   =  2 ( ) 2 +  1 ( ) +  0 ( )


1.2.1. Preliminary Remarks
1.2.2. Equations Containing Power Functions
1.2.2-1. Equations of the form ( ) =  2 ( ) 2 +  0 ( )
1.2.2-2. Other equations
1.2.3. Equations Containing Exponential Functions
1.2.3-1. Equations with exponential functions
1.2.3-2. Equations with power and exponential functions
1.2.4. Equations Containing Hyperbolic Functions
1.2.4-1. Equations with hyperbolic sine and cosine
1.2.4-2. Equations with hyperbolic tangent and cotangent
1.2.5. Equations Containing Logarithmic Functions
1.2.5-1. Equations of the form ( ) =  2 ( ) 2 +  0 ( )
1.2.5-2. Equations of the form ( ) =  2 ( ) 2 +  1 ( )  +  0 ( )
1.2.6. Equations Containing Trigonometric Functions
1.2.6-1. Equations with sine
1.2.6-2. Equations with cosine
1.2.6-3. Equations with tangent
1.2.6-4. Equations with cotangent
1.2.6-5. Equations containing combinations of trigonometric functions
1.2.7. Equations Containing Inverse Trigonometric Functions
1.2.7-1. Equations containing arcsine
1.2.7-2. Equations containing arccosine
1.2.7-3. Equations containing arctangent
1.2.7-4. Equations containing arccotangent
1.2.8. Equations with Arbitrary Functions
1.2.8-1. Equations containing arbitrary functions (but not containing their
derivatives)
1.2.8-2. Equations containing arbitrary functions and their derivatives
1.2.9. Some Transformations
1.3. Abel Equations of the Second Kind
1.3.1. Equations of the Form ! =  ( )
1.3.1-1. Preliminary remarks. Classification tables
1.3.1-2. Solvable equations and their solutions
1.3.2. Equations of the Form ! =  ( ) + 1
1.3.3. Equations of the Form ! =  1 ( ) +  0 ( )
1.3.3-1. Preliminary remarks
1.3.3-2. Solvable equations and their solutions
1.3.4. Equations of the Form [ 1 ( ) + 0 ( )]  =  2 ( ) 2 +  1 ( ) +  0 ( )
1.3.4-1. Preliminary remarks
1.3.4-2. Solvable equations and their solutions
1.3.5. Some Types of First- and Second-Order Equations Reducible to Abel Equations
of the Second Kind
1.3.5-1. Quasi-homogeneous equations
1.3.5-2. Equations of the theory of chemical reactors and the combustion theory
1.3.5-3. Equations of the theory of nonlinear oscillations
1.3.5-4. Second-order homogeneous equations of various types
1.3.5-5. Second-order equations invariant under some transformations

2003 by Chapman & Hall/CRC

Page ix

1.4. Equations Containing Polynomial Functions of


1.4.1. Abel Equations of the First Kind " =  3 ( ) 3 +  2 ( ) 2 +  1 ( ) +  0 ( )
1.4.1-1. Preliminary remarks
1.4.1-2. Solvable equations and their solutions
1.4.2. Equations of the Form ( # 22 2 + # 12  + # 11  2 + # 0 )  = $ 22 2 + $ 12  + $ 11  2 + $
1.4.2-1. Preliminary remarks. Some transformations
1.4.2-2. Solvable equations and their solutions
1.4.3. Equations of the Form ( # 22 2 + # 12  + # 11  2 + # 2 + # 1  )  =
$ 22 2 + $ 12  + $ 11  2 + $ 2 + $ 1 
1.4.3-1. Preliminary remarks
1.4.3-2. Solvable equations and their solutions
1.4.4. Equations of the Form ( # 22 2 + # 12  + # 11  2 + # 2 + # 1  + # 0 )  =
$ 22 2 + $ 12  + $ 11  2 + $ 2 + $ 1  + $ 0
1.4.4-1. Preliminary remarks. Some transformations
1.4.4-2. Solvable equations and their solutions
1.4.5. Equations of the Form ( # 3 3 + # 2  2 + # 1  2 + # 0  3 +  1 +  0  )  =
$ 3 3 + $ 2 2 + $ 1 2 + $ 0 3 +  1 +  0

142

1.5. Equations of the Form  ( , ) = ( , ) Containing Arbitrary Parameters


1.5.1. Equations Containing Power Functions
1.5.1-1. Equations of the form  =  ( , )
1.5.1-2. Other equations
1.5.2. Equations Containing Exponential Functions
1.5.2-1. Equations with exponential functions
1.5.2-2. Equations with power and exponential functions
1.5.3. Equations Containing Hyperbolic Functions
1.5.4. Equations Containing Logarithmic Functions
1.5.5. Equations Containing Trigonometric Functions
1.5.6. Equations Containing Combinations of Exponential, Hyperbolic, Logarithmic,
and Trigonometric Functions

1.6. Equations of the Form  ( , ,  ) = 0 Containing Arbitrary Parameters


1.6.1. Equations of the Second Degree in  
1.6.1-1. Equations of the form  ( , )( )2 = ( , )
1.6.1-2. Equations of the form  ( , )( )2 = ( , )  + % ( , )
1.6.2. Equations of the Third Degree in 
1.6.2-1. Equations of the form  ( , )( )3 = ( , )  + % ( , )
1.6.2-2. Equations of the form  ( , )( )3 = ( , )(  )2 + % ( , )  + & ( , ) 180
1.6.3. Equations of the Form ( )  =  ( ) + ( )
1.6.3-1. Some transformations
1.6.3-2. Classification tables and exact solutions
1.6.4. Other Equations
1.6.4-1. Equations containing algebraic and power functions with respect to 
1.6.4-2. Equations containing exponential, logarithmic, and other functions with
respect to  
1.7. Equations of the Form  ( , ) = ( , ) Containing Arbitrary Functions
1.7.1. Equations Containing Power Functions
1.7.2. Equations Containing Exponential and Hyperbolic Functions
1.7.3. Equations Containing Logarithmic Functions
1.7.4. Equations Containing Trigonometric Functions
1.7.5. Equations Containing Combinations of Exponential, Logarithmic, and
Trigonometric Functions

2003 by Chapman & Hall/CRC

Page x

1.8. Equations of the Form  ( , ,  ) = 0 Containing Arbitrary Functions


1.8.1. Some Equations
1.8.1-1. Arguments of arbitrary functions depend on  and
1.8.1-2. Argument of arbitrary functions is "
1.8.1-3. Arguments of arbitrary functions are linear with respect to 
1.8.1-4. Arguments of arbitrary functions are nonlinear with respect to 
1.8.2. Some Transformations
2. Second-Order Differential Equations
2.1. Linear Equations
2.1.1. Representation of the General Solution Through a Particular Solution
2.1.2. Equations Containing Power Functions
2.1.2-1. Equations of the form   +  ( ) = 0
2.1.2-2. Equations of the form   +  ( )  + ( ) = 0
2.1.2-3. Equations of the form (  +  )  +  ( )  + ( ) = 0
2.1.2-4. Equations of the form  2   +  ( )  + ( ) = 0
2.1.2-5. Equations of the form (  2 + ' + ( )   +  ( )  + ( ) = 0
2.1.2-6. Equations of the form (  3  3 +  2  2 +  1  +  0 )   +  ( )  + ( ) = 0
2.1.2-7. Equations of the form (  4  4 + ))) +  1  +  0 )   +  ( )  + ( ) = 0
2.1.2-8. Other equations
2.1.3. Equations Containing Exponential Functions
2.1.3-1. Equations with exponential functions
2.1.3-2. Equations with power and exponential functions
2.1.4. Equations Containing Hyperbolic Functions
2.1.4-1. Equations with hyperbolic sine
2.1.4-2. Equations with hyperbolic cosine
2.1.4-3. Equations with hyperbolic tangent
2.1.4-4. Equations with hyperbolic cotangent
2.1.4-5. Equations containing combinations of hyperbolic functions
2.1.5. Equations Containing Logarithmic Functions
2.1.5-1. Equations of the form  ( )  + ( ) = 0
2.1.5-2. Equations of the form  ( )  + ( )  + % ( ) = 0
2.1.6. Equations Containing Trigonometric Functions
2.1.6-1. Equations with sine
2.1.6-2. Equations with cosine
2.1.6-3. Equations with tangent
2.1.6-4. Equations with cotangent
2.1.6-5. Equations containing combinations of trigonometric functions
2.1.7. Equations Containing Inverse Trigonometric Functions
2.1.7-1. Equations with arcsine
2.1.7-2. Equations with arccosine
2.1.7-3. Equations with arctangent
2.1.7-4. Equations with arccotangent
2.1.8. Equations Containing Combinations of Exponential, Logarithmic, Trigonometric,
and Other Functions
2.1.9. Equations with Arbitrary Functions
2.1.9-1. Equations containing arbitrary functions (but not containing their
derivatives)
2.1.9-2. Equations containing arbitrary functions and their derivatives
2.1.10. Some Transformations

2003 by Chapman & Hall/CRC

Page xi

2.2. Autonomous Equations   =  ( ,  )


2.2.1. Equations of the Form    =  ( )
2.2.2. Equations of the Form    +  ( )   + = 0
2.2.2-1. Preliminary remarks
2.2.2-2. Solvable equations and their solutions
2.2.3. Lienard Equations   +  ( )  + ( ) = 0
2.2.3-1. Preliminary remarks
2.2.3-2. Solvable equations and their solutions
2.2.4. Rayleigh Equations   +  (  ) + ( ) = 0
2.2.4-1. Preliminary remarks. Some transformations
2.2.4-2. Solvable equations and their solutions

2.3. EmdenFowler Equation   = #*  +


2.3.1. Exact Solutions
2.3.1-1. Preliminary remarks. Classification table
2.3.1-2. Solvable equations and their solutions
2.3.2. First Integrals (Conservation Laws)
2.3.2-1. First integrals with , = 2
2.3.2-2. First integrals with , = 3
2.3.2-3. First integrals with , = 4
2.3.2-4. First integrals with , = 5
2.3.3. Some Formulas and Transformations
2.4. Equations of the Form   = #
2.4.1. Classification Table
2.4.2. Exact Solutions

  1 +

+#

  2 +

2.5. Generalized EmdenFowler Equation    = #*  + (   )2.5.1. Classification Table


2.5.2. Exact Solutions
2.5.3. Some Formulas and Transformations
2.5.3-1. A particular solution
2.5.3-2. Discrete transformations of the generalized EmdenFowler equation
2.5.3-3. Reduction of the generalized EmdenFowler equation to an Abel
equation
2.6. Equations of the Form   = # 1   1 + 1 (  )- 1 + # 2   2 + 2 (  )- 2
2.6.1. Modified EmdenFowler Equation   = # 1  1  + # 2   +
2.6.1-1. Preliminary remarks. Classification table
2.6.1-2. Solvable equations and their solutions
2.6.2. Equations of the Form   = ( # 1   1 + 1 + # 2   2 + 2 )(  )2.6.2-1. Classification table
2.6.2-2. Solvable equations and their solutions
2.6.3. Equations of the Form   = ./#*  + (  )- + #*  1 + +1 (  )- 1
2.6.3-1. Classification table
2.6.3-2. Solvable equations and their solutions
2.6.4. Other Equations ( 0 1 0 2 )
2.6.4-1. Classification table
2.6.4-2. Solvable equations and their solutions
2.7. Equations of the Form   =  ( ) ( ) % (   )
2.7.1. Equations of the Form   =  ( ) ( )
2.7.2. Equations Containing Power Functions ( %21 const)
2.7.3. Equations Containing Exponential Functions ( %31 const)

2003 by Chapman & Hall/CRC

Page xii

2.7.3-1. Preliminary remarks


2.7.3-2. Solvable equations and their solutions
2.7.4. Equations Containing Hyperbolic Functions ( %31 const)
2.7.5. Equations Containing Trigonometric Functions ( %31 const)
2.7.6. Some Transformations
2.8. Some Nonlinear Equations with Arbitrary Parameters
2.8.1. Equations Containing Power Functions
2.8.1-1. Equations of the form  ( , )  + ( , ) = 0
2.8.1-2. Equations of the form  ( , )  + ( , )  + % ( , ) = 0
2.8.1-3. Equations of the form  ( , )  + ( , )(  )2 + % ( , )  + & ( , ) = 0
2.8.1-4. Other equations
2.8.2. Painleve Transcendents
2.8.2-1. Preliminary remarks. Singular points of solutions
2.8.2-2. First Painleve transcendent
2.8.2-3. Second Painleve transcendent
2.8.2-4. Third Painleve transcendent
2.8.2-5. Fourth Painleve transcendent
2.8.2-6. Fifth Painleve transcendent
2.8.2-7. Sixth Painleve transcendent
2.8.3. Equations Containing Exponential Functions
2.8.3-1. Equations of the form  ( , )  + ( , ) = 0
2.8.3-2. Equations of the form  ( , )  + ( , )  + % ( , ) = 0
2.8.3-3. Equations of the form  ( , )  + ( , )(  )2 + % ( , )  + & ( , ) = 0
2.8.3-4. Other equations
2.8.4. Equations Containing Hyperbolic Functions
2.8.4-1. Equations with hyperbolic sine
2.8.4-2. Equations with hyperbolic cosine
2.8.4-3. Equations with hyperbolic tangent
2.8.4-4. Equations with hyperbolic cotangent
2.8.4-5. Equations containing combinations of hyperbolic functions
2.8.5. Equations Containing Logarithmic Functions
2.8.5-1. Equations of the form  ( , )  + ( , )  + % ( , ) = 0
2.8.5-2. Other equations
2.8.6. Equations Containing Trigonometric Functions
2.8.6-1. Equations with sine
2.8.6-2. Equations with cosine
2.8.6-3. Equations with tangent
2.8.6-4. Equations with cotangent
2.8.6-5. Equations containing combinations of trigonometric functions
2.8.7. Equations Containing the Combinations of Exponential, Hyperbolic, Logarithmic,
and Trigonometric Functions
2.9. Equations Containing Arbitrary Functions
2.9.1. Equations of the Form  ( , )  + 4 ( , ) = 0
2.9.1-1. Arguments of arbitrary functions are algebraic and power functions of
 and
2.9.1-2. Arguments of the arbitrary functions are other functions
2.9.2. Equations of the Form  ( , )  + 4 ( , )  + 5 ( , ) = 0
2.9.2-1. Argument of the arbitrary functions is 
2.9.2-2. Argument of the arbitrary functions is
2.9.2-3. Other arguments of the arbitrary functions

2003 by Chapman & Hall/CRC

Page xiii

2.9.3. Equations of the Form  ( , )  + 6

2.9.4.

2.9.5.
2.9.6.

2.9.7.

=0

4 + ( , )(  ) + = 0 ( 8

= 2, 3, 4)

2.9.3-1. Argument of the arbitrary functions is  .


2.9.3-2. Argument of the arbitrary functions is
2.9.3-3. Other arguments of arbitrary functions
Equations of the Form  ( , ,  )   + 4 ( , ,  ) = 0
2.9.4-1. Arguments of the arbitrary functions depend on  or
2.9.4-2. Arguments of the arbitrary functions depend on  and
2.9.4-3. Arguments of the arbitrary functions depend on  , , and 
Equations Not Solved for Second Derivative
Equations of General Form
2.9.6-1. Equations containing arbitrary functions of two variables
2.9.6-2. Equations containing arbitrary functions of three variables
Some Transformations

3. Third-Order Differential Equations


3.1. Linear Equations
3.1.1. Preliminary Remarks
3.1.2. Equations Containing Power Functions
3.1.2-1. Equations of the form  3 ( )    +  0 ( ) = ( )
3.1.2-2. Equations of the form  3 ( )    +  1 ( )  +  0 ( ) = ( )
3.1.2-3. Equations of the form  3 ( )    +  2 ( )   +  1 ( )  +  0 ( ) = ( )
3.1.3. Equations Containing Exponential Functions
3.1.3-1. Equations with exponential functions
3.1.3-2. Equations with power and exponential functions
3.1.4. Equations Containing Hyperbolic Functions
3.1.4-1. Equations with hyperbolic sine
3.1.4-2. Equations with hyperbolic cosine
3.1.4-3. Equations with hyperbolic sine and cosine
3.1.4-4. Equations with hyperbolic tangent
3.1.4-5. Equations with hyperbolic cotangent
3.1.5. Equations Containing Logarithmic Functions
3.1.5-1. Equations with logarithmic functions
3.1.5-2. Equations with power and logarithmic functions
3.1.6. Equations Containing Trigonometric Functions
3.1.6-1. Equations with sine
3.1.6-2. Equations with cosine
3.1.6-3. Equations with sine and cosine
3.1.6-4. Equations with tangent
3.1.6-5. Equations with cotangent
3.1.7. Equations Containing Inverse Trigonometric Functions
3.1.8. Equations Containing Combinations of Exponential, Logarithmic, Trigonometric,
and Other Functions
3.1.9. Equations Containing Arbitrary Functions
3.1.9-1. Equations of the form  3 ( )    +  1 ( )  +  0 ( ) = ( )
3.1.9-2. Equations of the form  3 ( )    +  2 ( )   +  1 ( )  +  0 ( ) = ( )

3.2. Equations of the Form    = #* 9 : (  ); (   ) <


3.2.1. Classification Table
3.2.2. Equations of the Form    = # :

2003 by Chapman & Hall/CRC

Page xiv

3.2.3. Equations of the Form     = #* 9


3.2.4. Equations with |= | + | > | 0
3.2.5. Some Transformations

3.3. Equations of the Form    =  ( ) (  ) % (   )


3.3.1. Equations Containing Power Functions
3.3.2. Equations Containing Exponential Functions
3.3.3. Other Equations
3.4. Nonlinear Equations with Arbitrary Parameters
3.4.1. Equations Containing Power Functions
3.4.1-1. Equations of the form  ( , )     = ( , )
3.4.1-2. Equations of the form     =  ( , ,   )
3.4.1-3. Equations of the form  ( , ,  )    + ( , ,  )   + % ( , ,  ) = 0
3.4.1-4. Other equations
3.4.2. Equations Containing Exponential Functions
3.4.2-1. Equations of the form    =  ( , ,  )
3.4.2-2. Other equations
3.4.3. Equations Containing Hyperbolic Functions
3.4.3-1. Equations with hyperbolic sine
3.4.3-2. Equations with hyperbolic cosine
3.4.3-3. Equations with hyperbolic tangent
3.4.3-4. Equations with hyperbolic cotangent
3.4.4. Equations Containing Logarithmic Functions
3.4.4-1. Equations of the form    =  ( , ,  )
3.4.4-2. Other equations
3.4.5. Equations Containing Trigonometric Functions
3.4.5-1. Equations with sine
3.4.5-2. Equations with cosine
3.4.5-3. Equations with tangent
3.4.5-4. Equations with cotangent
3.5. Nonlinear Equations Containing Arbitrary Functions
3.5.1. Equations of the Form  ( , )   + 4 ( , ) = 0
3.5.1-1. Arguments of the arbitrary functions are  or
3.5.1-2. Arguments of the arbitrary functions depend on  and
3.5.2. Equations of the Form  ( , ,  )    + 4 ( , ,  ) = 0
3.5.2-1. Arguments of the arbitrary functions depend on  and
3.5.2-2. Arguments of the arbitrary functions depend on  , , and 
3.5.3. Equations of the Form  ( , ,   )     + 4 ( , ,   )    + 5 ( , ,   ) = 0
3.5.3-1. The arbitrary functions depend on  or
3.5.3-2. Arguments of arbitrary functions depend on  and
3.5.3-3. Arguments of arbitrary functions depend on  , , and 
7
3.5.4. Equations of the Form  ( , ,  )    + 4 ( , ,  )(   ) 9 = 0

3.5.4-1. Arbitrary functions depend on  or


3.5.4-2. Arguments of arbitrary functions depend on  , , and 
3.5.5. Other Equations
3.5.5-1. Equations of the form  ( , ,  ,   )    + 4 ( , ,  ,   ) = 0
3.5.5-2. Equations of the form  ( , ,  ,   ,    ) = 0

2003 by Chapman & Hall/CRC

Page xv

xvi

CONTENTS

4. Fourth-Order Differential Equations


4.1. Linear Equations
4.1.1. Preliminary Remarks
4.1.2. Equations Containing Power Functions
4.1.2-1. Equations of the form  4 ( )      +  0 ( ) = ( )
4.1.2-2. Equations of the form  4 ( )      +  1 ( )   +  0 ( ) = ( )
4.1.2-3. Equations of the form  4 ( )     +  2 ( )   +  1 ( )  +  0 ( ) = ( )
4.1.2-4. Other equations
4.1.3. Equations Containing Exponential and Hyperbolic Functions
4.1.3-1. Equations with exponential functions
4.1.3-2. Equations with hyperbolic functions
4.1.4. Equations Containing Logarithmic Functions
4.1.5. Equations Containing Trigonometric Functions
4.1.5-1. Equations with sine and cosine
4.1.5-2. Equations with tangent and cotangent
4.1.6. Equations Containing Arbitrary Functions
4.1.6-1. Equations of the form  4 ( )     +  1 ( )  +  0 ( ) = ( )
4.1.6-2. Equations of the form  4 ( )     +  2 ( )   +  1 ( )  +  0 ( ) = ( )
4.1.6-3. Other equations
4.2. Nonlinear Equations
4.2.1. Equations Containing Power Functions
4.2.1-1. Equations of the form     =  ( , )
4.2.1-2. Equations of the form     =  ( , , 
4.2.1-3. Equations of the form     =  ( , , 
4.2.1-4. Equations of the form     =  ( , , 
4.2.2. Equations Containing Exponential Functions
4.2.2-1. Equations of the form     =  ( , )
4.2.2-2. Other equations
4.2.3. Equations Containing Hyperbolic Functions
4.2.3-1. Equations with hyperbolic sine
4.2.3-2. Equations with hyperbolic cosine
4.2.3-3. Equations with hyperbolic tangent
4.2.3-4. Equations with hyperbolic cotangent
4.2.4. Equations Containing Logarithmic Functions
4.2.4-1. Equations of the form      =  ( , )
4.2.4-2. Other equations
4.2.5. Equations Containing Trigonometric Functions
4.2.5-1. Equations with sine
4.2.5-2. Equations with cosine
4.2.5-3. Equations with tangent
4.2.5-4. Equations with cotangent
4.2.6. Equations Containing Arbitrary Functions
4.2.6-1. Equations of the form      =  ( , )
4.2.6-2. Equations of the form     =  ( , , 
4.2.6-3. Equations of the form     =  ( , , 
4.2.6-4. Equations of the form     =  ( , , 
4.2.6-5. Other equations

)
,   )
,   ,     )

)
,   )
,   ,     )

2003 by Chapman & Hall/CRC

Page xvi

5. Higher-Order Differential Equations


5.1. Linear Equations
5.1.1. Preliminary Remarks
5.1.2. Equations Containing Power Functions
5.1.2-1. Equations of the form   ( ) (  ) +  0 ( ) = ( )
5.1.2-2. Equations of the form   ( ) (  ) +  1 ( )   +  0 ( ) = ( )
5.1.2-3. Other equations
5.1.3. Equations Containing Exponential and Hyperbolic Functions
5.1.3-1. Equations with exponential functions
5.1.3-2. Equations with hyperbolic functions
5.1.4. Equations Containing Logarithmic Functions
5.1.5. Equations Containing Trigonometric Functions
5.1.5-1. Equations with sine and cosine
5.1.5-2. Equations with tangent and cotangent
5.1.6. Equations Containing Arbitrary Functions
5.1.6-1. Equations of the form   ( ) (  ) +  1 ( )  +  0 ( ) = ( )
5.1.6-2. Other equations
5.2. Nonlinear Equations
5.2.1. Equations Containing Power Functions
5.2.1-1. Fifth- and sixth-order equations
5.2.1-2. Equations of the form (  ) =  ( , )
5.2.1-3. Equations of the form (  ) =  ( , ,  ,   )
5.2.1-4. Other equations
5.2.2. Equations Containing Exponential Functions
5.2.2-1. Fifth- and sixth-order equations
5.2.2-2. Equations of the form (  ) =  ( , )
5.2.2-3. Other equations
5.2.3. Equations Containing Hyperbolic Functions
5.2.3-1. Equations with hyperbolic sine
5.2.3-2. Equations with hyperbolic cosine
5.2.3-3. Equations with hyperbolic tangent
5.2.3-4. Equations with hyperbolic cotangent
5.2.4. Equations Containing Logarithmic Functions
5.2.4-1. Equations of the form (  ) =  ( , )
5.2.4-2. Other equations
5.2.5. Equations Containing Trigonometric Functions
5.2.5-1. Equations with sine
5.2.5-2. Equations with cosine
5.2.5-3. Equations with tangent
5.2.5-4. Equations with cotangent
5.2.6. Equations Containing Arbitrary Functions
5.2.6-1. Fifth- and sixth-order equations
5.2.6-2. Equations of the form (  ) =  ( , )
5.2.6-3. Equations of the form (  ) =  ( , ,  )
5.2.6-4. Equations of the form (  ) =  ( , ,  ,   )
5.2.6-5. Equations of the form  ( , ) (  ) + ( , ,  ) (  1) =
 2) 
%   , ,  ,  , (?
 1) 
5.2.6-6. Equations of the form (  ) =    , ,  ,  , (?

( ) 
=0
5.2.6-7. Equations of the general form    , ,  ,  , 

2003 by Chapman & Hall/CRC

Page xvii

Supplements
S.1. Elementary Functions and Their Properties
S.1.1. Trigonometric Functions
S.1.1-1. Simplest relations
S.1.1-2. Relations between trigonometric functions of single argument
S.1.1-3. Reduction formulas
S.1.1-4. Addition and subtraction of trigonometric functions
S.1.1-5. Products of trigonometric functions
S.1.1-6. Powers of trigonometric functions
S.1.1-7. Addition formulas
S.1.1-8. Trigonometric functions of multiple arguments
S.1.1-9. Trigonometric functions of half argument
S.1.1-10. Euler and de Moivre formulas. Relationship with hyperbolic functions
S.1.1-11. Differentiation formulas
S.1.1-12. Expansion into power series
S.1.2. Hyperbolic Functions
S.1.2-1. Definitions
S.1.2-2. Simplest relations
S.1.2-3. Relations between hyperbolic functions of single argument ( 0)
S.1.2-4. Addition formulas
S.1.2-5. Addition and subtraction of hyperbolic functions
S.1.2-6. Products of hyperbolic functions
S.1.2-7. Powers of hyperbolic functions
S.1.2-8. Hyperbolic functions of multiple arguments
S.1.2-9. Relationship with trigonometric functions
S.1.2-10. Differentiation formulas
S.1.2-11. Expansion into power series
S.1.3. Inverse Trigonometric Functions
S.1.3-1. Definitions and some properties
S.1.3-2. Simplest formulas
S.1.3-3. Relations between inverse trigonometric functions
S.1.3-4. Addition and subtraction of inverse trigonometric functions
S.1.3-5. Differentiation formulas
S.1.3-6. Expansion into power series
S.1.4. Inverse Hyperbolic Functions
S.1.4-1. Relationships with logarithmic functions
S.1.4-2. Relations between inverse hyperbolic functions
S.1.4-3. Addition and subtraction of inverse hyperbolic functions
S.1.4-4. Differentiation formulas
S.1.4-5. Expansion into power series
S.2. Special Functions and Their Properties
S.2.1. Some Symbols and Coefficients
S.2.1-1. Factorials
S.2.1-2. Binomial coefficients
S.2.1-3. Pochhammer symbol
S.2.2. Error Functions and Exponential Integral
S.2.2-1. Error function and complementary error function
S.2.2-2. Exponential integral
S.2.2-3. Logarithmic integral

2003 by Chapman & Hall/CRC

Page xviii

S.2.3. Gamma and Beta Functions


S.2.3-1. Gamma function
S.2.3-2. Logarithmic derivative of the gamma function
S.2.3-3. Beta function
S.2.4. Incomplete Gamma and Beta Functions
S.2.4-1. Incomplete gamma function
S.2.4-2. Incomplete beta function
S.2.5. Bessel Functions
S.2.5-1. Definitions and basic formulas
S.2.5-2. Bessel functions for @ = A"BCA 12 , where B = 0, 1, 2, 
S.2.5-3. Bessel functions for @ = A"B , where B = 0, 1, 2, 
S.2.5-4. Wronskians and similar formulas
S.2.5-5. Integral representations
S.2.5-6. Asymptotic expansions
S.2.5-7. Zeros and orthogonality properties of the Bessel functions
S.2.5-8. Hankel functions (Bessel functions of the third kind)
S.2.6. Modified Bessel Functions
S.2.6-1. Definitions. Basic formulas
S.2.6-2. Modified Bessel functions for @ = A"BCA 12 , where B = 0, 1, 2, 
S.2.6-3. Modified Bessel functions for @ = B , where B = 0, 1, 2, 
S.2.6-4. Wronskians and similar formulas
S.2.6-5. Integral representations
S.2.6-6. Asymptotic expansions as EDGF
S.2.7. Degenerate Hypergeometric Functions
S.2.7-1. Definitions. The Kummers series
S.2.7-2. Some transformations and linear relations
S.2.7-3. Differentiation formulas and Wronskian
S.2.7-4. Degenerate hypergeometric functions for B = 0, 1, 2, 
S.2.7-5. Integral representations
S.2.7-6. Asymptotic expansion as | | DGF
S.2.7-7. Whittaker functions
S.2.8. Hypergeometric Functions
S.2.8-1. Definition. The hypergeometric series
S.2.8-2. Basic properties
S.2.8-3. Integral representations
S.2.9. Legendre Functions and Legendre Polynomials
S.2.9-1. Definitions. Basic formulas
S.2.9-2. Trigonometric expansions
S.2.9-3. Some relations
S.2.9-4. Integral representations
S.2.9-5. Legendre polynomials
S.2.9-6. Zeros of the Legendre polynomials and the generating function
S.2.9-7. Associated Legendre functions
S.2.10. Parabolic Cylinder Functions
S.2.10-1. Definitions. Basic formulas
S.2.10-2. Integral representations
S.2.10-3. Asymptotic expansion as | H | DGF
S.2.11. Orthogonal Polynomials
S.2.11-1. Laguerre polynomials and generalized Laguerre polynomials
S.2.11-2. Chebyshev polynomials
S.2.11-3. Hermite polynomials

2003 by Chapman & Hall/CRC

Page xix

S.2.11-4. Gegenbauer polynomials


S.2.11-5. Jacobi polynomials
S.2.12. The Weierstrass Function
S.2.12-1. Definitions
S.2.12-2. Some properties
S.3. Tables of Indefinite Integrals
S.3.1. Integrals Containing Rational Functions
S.3.1-1. Integrals containing  + '
S.3.1-2. Integrals containing  +  and  + 
S.3.1-3. Integrals containing  2 +  2
S.3.1-4. Integrals containing  2  2
S.3.1-5. Integrals containing  3 +  3
S.3.1-6. Integrals containing  3  3
S.3.1-7. Integrals containing  4 AI 4
S.3.2. Integrals Containing Irrational Functions
S.3.2-1. Integrals containing  1 J 2
S.3.2-2. Integrals containing (  + ' )K J 2
S.3.2-3. Integrals containing ( 2 +  2 )1 J 2
S.3.2-4. Integrals containing ( 2  2 )1 J 2
S.3.2-5. Integrals containing (  2  2 )1 J 2
S.3.2-6. Reduction formulas
S.3.3. Integrals Containing Exponential Functions
S.3.4. Integrals Containing Hyperbolic Functions
S.3.4-1. Integrals containing cosh 
S.3.4-2. Integrals containing sinh 
S.3.4-3. Integrals containing tanh  or coth 
S.3.5. Integrals Containing Logarithmic Functions
S.3.6. Integrals Containing Trigonometric Functions
S.3.6-1. Integrals containing cos 
S.3.6-2. Integrals containing sin 
S.3.6-3. Integrals containing sin  and cos 
S.3.6-4. Reduction formulas
S.3.6-5. Integrals containing tan  and cot 
S.3.7. Integrals Containing Inverse Trigonometric Functions
References

2003 by Chapman & Hall/CRC

Page xx

AUTHORS
Andrei D. Polyanin, Ph.D., D.Sc., is a noted scientist of broad
interests who works in various areas of mathematics, mechanics,
and chemical engineering sciences.
A. D. Polyanin graduated from the Department of Mechanics
and Mathematics of the Moscow State University in 1974. He
earned his Ph.D. degree in 1981 and D.Sc. degree in 1986 at the
Institute for Problems in Mechanics of the Russian (former USSR)
Academy of Sciences. Since 1975, A. D. Polyanin has been a
member of the staff of the Institute for Problems in Mechanics of
the Russian Academy of Sciences. He is a member of the Russian
National Committee on Theoretical and Applied Mechanics.
Professor Polyanin has made important contributions to developing new exact and approximate analytical methods of the theory
of differential equations, mathematical physics, integral equations,
engineering mathematics, nonlinear mechanics, theory of heat and
mass transfer, and chemical hydrodynamics. He obtained exact
solutions for several thousand ordinary differential, partial differential, mathematical physics, and integral equations.
Professor Polyanin is an author of 30 books in English, Russian, German, and Bulgarian, as well
as over 120 research papers and three patents. He has written a number of fundamental handbooks,
including A. D. Polyanin and V. F. Zaitsev, Handbook of Exact Solutions for Ordinary Differential Equations (first edition), CRC Press, 1995; A. D. Polyanin and A. V. Manzhirov, Handbook
of Integral Equations, CRC Press, 1998; A. D. Polyanin, Handbook of Linear Partial Differential Equations for Engineers and Scientists, Chapman & Hall/CRC Press, 2002; A. D. Polyanin,
V. F. Zaitsev, and A. Moussiaux, Handbook of First Order Partial Differential Equations, Taylor &
Francis, 2002; and A. D. Polyanin and V. F. Zaitsev, Handbook of Nonlinear Mathematical Physics
Equations, Fizmatlit, 2002. Professor Polyanin is Editor of the book series Differential and Integral
Equations and Their Applications, Taylor & Francis Inc., London.
In 1991, A. D. Polyanin was awarded a Chaplygin Prize of the Russian Academy of Sciences
for his research in mechanics.
Address: Institute for Problems in Mechanics, RAS, 101 Vernadsky Avenue, Building 1, 119526 Moscow, Russia
E-mail: polyanin@ipmnet.ru

Valentin F. Zaitsev, Ph.D., D.Sc., is a noted scientist in the fields


of ordinary differential equations, mathematical physics, and nonlinear mechanics.
V. F. Zaitsev graduated from the Radio Electronics Faculty of
the Leningrad Polytechnical Institute (now St. Petersburg Technical University) in 1969 and received his Ph.D. degree in 1983 at
the Leningrad State University. His Ph.D. thesis was devoted to
the group approach to the study of some classes of ordinary differential equations. In 1992, Professor Zaitsev earned his Doctor
of Sciences degree; his D.Sc. thesis was dedicated to the discretegroup analysis of ordinary differential equations.
In 19711996, V. F. Zaitsev worked in the Research Institute for
Computational Mathematics and Control Processes of the St. Petersburg State University. Since 1996, Professor Zaitsev has been
a member of the staff of the Russian State Pedagogical University
(St. Petersburg).
Professor Zaitsev has made important contributions to new
methods in the theory of ordinary and partial differential equations. He is an author of more than
130 scientific publications, including 18 books and one patent.
Address: Russian State Pedagogical University, 48 Naberezhnaya reki Moiki, 191186 St. Petersburg, Russia
E-mail: zaitsev@osipenko.stu.neva.ru

2003 by Chapman & Hall/CRC

Page xxi

FOREWORD
Exact solutions of differential equations play an important role in the proper understanding of
qualitative features of many phenomena and processes in various areas of natural science. These
solutions can be used to verify the consistencies and estimate errors of various numerical, asymptotic,
and approximate analytical methods.
This book contains nearly 6200 ordinary differential equations and their solutions. A number
of new solutions to nonlinear equations are described. In some sections of this book, asymptotic
solutions to some classes of equations are also given.
When selecting the material, the authors gave preference to the following two types of equations:
Equations that have traditionally attracted the attention of many researchers: those of the
simplest appearance but involving the most difficulties for integration (Abel equations, Emden
Fowler equations, Painleve equations, etc.)
Equations that are encountered in various applications (in the theory of heat and mass transfer,
nonlinear mechanics, elasticity, hydrodynamics, theory of nonlinear oscillations, combustion theory,
chemical engineering science, etc.)
Special attention is paid to equations that depend on arbitrary functions. All other equations
contain one or more arbitrary parameters (in fact, this book deals with whole families of ordinary
differential equations), which can be fixed by a reader at will. In total, the handbook contains many
more equations than any other book currently available (for example, the number of nonlinear
equations of the second and higher order is ten times more than in the well-known E. Kamkes
Handbook on Ordinary Differential Equations).
For the readers convenience, the introductory chapter of the book outlines basic definitions,
useful formulas, and some transformations. In a concise form, it also presents exact, asymptotic,
and approximate analytical methods for solving linear and nonlinear differential equations. Specific
examples of utilization of these methods are considered. Formulations of existence and uniqueness
theorems are also given. Boundary-value problems and eigenvalue problems are described.
The handbook consists of chapters, sections, subsections, and paragraphs. Equations and formulas are numbered separately in each subsection. The equations within subsections and paragraphs
are arranged in increasing order of complexity. The extensive table of contents provides rapid access
to the desired equations.
The main material is followed by some supplements, where basic properties of elementary and
special functions (Bessel, modified Bessel, hypergeometric, Legendre, etc.) are described.
Here are three main distinguishing features of the second edition vs. the first edition:
1200 nonlinear equations with solutions have been added.
An introductory chapter that outlines exact, asymptotic, and approximate analytical methods
for solving ordinary differential equations has been included.
The overwhelming majority of subsections are organized into paragraphs. As a result, the table
of contents has been increased threefold to help the readers get faster access to desired equations.
We would like to express our deep gratitude to Alexei Zhurov for fruitful discussions and
valuable remarks. We are very grateful to Alain Moussiaux who tested a number of solutions, which
allowed us to remove some inaccuracies and misprints.
The authors hope that this book will be helpful for a wide range of scientists, university teachers,
engineers, and students engaged in the fields of mathematics, physics, mechanics, control, chemistry,
and engineering sciences.
Andrei D. Polyanin
Valentin F. Zaitsev

2003 by Chapman & Hall/CRC

Page xxiii

NOTATIONS AND SOME REMARKS


1. Throughout this book, in the original equations the independent variable is denoted by  ,
and the dependent one is denoted by . In the given solutions, the symbols L , L 0 , L 1 , L 2 , 
stand for arbitrary integration constants. Solutions are often represented in parametric form (e.g.,
see Subsections 1.3.1 and 2.3.1).
2. Notation for derivatives:

 = M ,

M

    = M 2 ,

M

     = M 3 ,

M

3. Brief notation for partial derivatives:


  = N ,

N

      = M 4 ;

M


(  ) = M 

M

with

B 5.

2

   = N
,
where  =  ( , ).

N
N N

, which is defined by the recurrence
4. In some cases, we use the operator notation O M
3P
M
relation

 1
( ) =  ( ) M O ( ) M
O ( ) M
(  )S .
2P
RQ
2P
M
M
M
= N .
5. Brief operator notation corresponding to partial derivatives:  = N ,
N
N 

N
N

  = N

2

  = N 2 ,

N

6. In some sections of the book (e.g., see 1.3, 2.32.6, 3.23.3), for the sake of brevity, solutions
are represented as several formulas containing the terms with the signs A and T . Two formulas
are meantone corresponds to the upper sign and the other to the lower sign. For example, the
solution of equation 1.3.1.6 can be written in the parametric form:

 =  

exp(T"U 2 ),

=  

V exp(T"U 2 ) A 2UW ,

where

 = X exp(T U 2 ) U L ,
M

# = T 2 2.

This is equivalent so that the solutions of equation 1.3.1.6 are given by the two formulas:

 =  

exp(U 2 ),

=  

V exp(U 2 ) + 2UW ,

where

 = X exp(U 2 ) U L ,
M

V exp(U 2 ) 2UW ,

where

 = X exp(U 2 ) U L ,
M

# = 2

(for the upper signs) and

 =  

exp(U 2 ),

=  

# = 2

(for the lower signs).

 ( )
, it is often not stated that the assumption  2
7. If a relation contains an expression like
 2
is adopted.
1
  +1 can usually be defined so as to cover the
8. In solutions, expressions like Y  ( ) =
B +1
case B = 1 in accordance with the rule Y 1 ( ) = ln | |. This is accounted for by the fact that such
expressions arise from the integration of the power-law function Y  ( ) = XZ   .
M

9. The order symbol [ is used to compare two functions  =  ( ) and = ( ), where is a


small parameter. So  = [ ( ) means that | 
| is bounded as \D 0, or  and are of the same
order of magnitude as ID 0.

2003 by Chapman & Hall/CRC

Page xxv

10. When referencing a particular equation, a notation like 4.1.2.5 stands for equation 5 in
Subsection 4.1.2.
11. The handbooks by Kamke (1977), Murphy (1960), Zaitsev and Polyanin (1993, 2001),
Polyanin and Zaitsev (1995) were extensively used in compiling this book; references to these
sources are frequently omitted.

^_

12. To highlight portions of the text, the following symbols are used throughout the book:
indicates important information pertaining to a group of equations (Chapters 15);
indicates the literature used in the preparation of the text in subsections, paragraphs, and specific
equations.

2003 by Chapman & Hall/CRC

Page xxvi

Introduction

Some Definitions, Formulas,


Methods, and Transformations
0.1. First-Order Differential Equations
0.1.1. General Concepts. The Cauchy Problem. Uniqueness and
Existence Theorems
0.1.1-1. Equations solved for the derivative. General solution.
A first-order ordinary differential equation* solved for the derivative has the form

  =  ( , ).

(1)

=  ( , )  .
Sometimes it is represented in terms of differentials as
M
M
A solution of a differential equation is a function ( ) that, when substituted into the equation,
turns it into an identity. The general solution of a differential equation is the set of all its solutions.
In some cases, the general solution can be represented as a function = Y ( , L ) that depends on
one arbitrary constant L ; specific values of L define specific solutions of the equation (particular
solutions). In practice, the general solution more frequently appears in implicit form, ` ( , , L ) = 0,
or parametric form,  =  (a , L ), = (a , L ).
Geometrically, the general solution (also called the general integral) of an equation is a family
of curves in the  -plane depending on a single parameter L ; these curves are called integral curves
of the equation. To each particular solution (particular integral) there corresponds a single curve
that passes through a given point in the plane.
For each point ( , ), the equation  =  ( , ) defines a value of   , i.e., the slope of the
integral curve that passes through this point. In other words, the equation generates a field of
directions in the  -plane. From the geometrical point of view, the problem of solving a first-order
differential equation involves finding the curves, the slopes of which at each point coincide with the
direction of the field at this point.
0.1.1-2. The Cauchy problem. The uniqueness and existence theorems.
1 b . The Cauchy problem: find a solution of equation (1) that satisfies the initial condition
=

at

 =  0,

(2)

where 0 and  0 are some numbers.


Geometrical meaning of the Cauchy problem: find an integral curve of equation (1) that passes
through the point ( 0 , 0 ).
Condition (2) is alternatively written ( 0 ) = 0 or |  =  0 = 0 .
* In what follows, we often call an ordinary differential equation a differential equation or even shorter an equation.

2003 by Chapman & Hall/CRC

Page 1

2 b . Existence theorem (Peano). Let the function  ( , ) be continuous in an open domain c of the
 -plane. Then there is at least one integral curve of equation (1) that passes through each point
( 0 , 0 ) dec ; each of these curves can be extended at both ends up to the boundary of any closed
domain c 0 f c such that ( 0 , 0 ) belongs to the interior of c 0 .
3 b . Uniqueness theorem. Let the function  ( , ) be continuous in an open domain c and have in c
a bounded partial derivative with respect to (or the Lipschitz condition holds: |  ( , )  ( , H )|
8 | H |, where 8 is some positive number). Then there is a unique solution of equation (1)
satisfying condition (2).
0.1.1-3. Equations not solved for the derivative. The existence theorem.
A first-order differential equation not solved for the derivative can generally be written as

 ( , ,   ) = 0.

(3)

Existence and uniqueness theorem. There exists a unique solution = ( ) of equation (3)
satisfying the conditions |  =  0 = 0 and  |  =  0 = a 0 , where a 0 is one of the real roots of the equation
 ( 0 , 0 , a 0 ) = 0, if the following conditions hold in a neighborhood of the point ( 0 , 0 , a 0 ):
1. The function  ( , , a ) is continuous in each of the three arguments.
2. The partial derivative hg exists and is nonzero.
3. There is a bounded partial derivative with respect to , |  | 8 .

The solution exists for |  0 |  , where  is a (sufficiently small) positive number.


0.1.1-4. Singular solutions.
1 b . A point ( , ) at which the uniqueness of the solution to equation (3) is violated is called a
singular point. If conditions 1 and 3 of the existence and uniqueness theorem hold, then

 ( , , a ) = 0,

hg ( , , a ) = 0

(4)

simultaneously at each singular point. Relations (4) define a a -discriminant curve in parametric
form. In some cases, the parameter a can be eliminated from (4) to give an equation of this curve
in implicit form, i ( , ) = 0. If a branch = j ( ) of the curve i ( , ) = 0 consists of singular
points and, at the same time, is an integral curve, then this branch is called a singular integral curve
and the function = j ( ) is a singular solution of equation (3).
2 b . The singular solutions can be found by identifying the envelope of the family of integral curves,
` ( , , L ) = 0, of equation (3). The envelope is part of the L -discriminant curve, which is defined
by the equations
` ( , , L ) = 0, `lk ( , , L ) = 0.
The branch of the L -discriminant curve at which
(a) there exist bounded partial derivatives, | `  | < 8
(b) | `  | + | ` | 0
is the envelope.

and | `

 |<8

2,

and

0.1.1-5. Point transformations.


In the general case, a point transformation is defined by

 =  (m , n ),

= 4 (m , n ),

(5)

2003 by Chapman & Hall/CRC

Page 2

where m is the new independent variable, n = n (m ) is the new dependent variable, and 
are some (prescribed or unknown) functions.
The derivative   under the point transformation (5) is calculated by

 =

4po + 4Iqrn
ho + s
 qrn
o

and 4


o ,


where the subscripts m and n denote the corresponding partial derivatives.


Transformation (5) is invertible if tou4Iq sqr4po 0.
Point transformations are used to simplify equations and reduce them to known equations.
Sometimes a point transformation allows the reduction of a nonlinear equation to a linear one.
Examples of point transformations can be found in Subsections 0.1.2 and 0.1.40.1.6.

^_

References for Subsection 0.1.1: G. M. Murphy (1960), G. A. Korn and T. M. Korn (1968), E. Kamke (1977),
A. N. Tikhonov, A. B. Vasileva, and A. G. Sveshnikov (1980), D. Zwillinger (1998).

0.1.2. Equations Solved for the Derivative. Simplest Techniques of


Integration
0.1.2-1. Equations with separated or separable variables.
1 b . An equation with separated variables (a separated equation) has the form

 ( )   = ( ).

= ( )  (the right-hand side depends


Equivalently, the equation can be rewritten as  ( )
M
M
on  alone and the left-hand side on alone). The general solution can be obtained by termwise
integration:
Xv ( ) = Xw ( )  + L ,

where L is an arbitrary constant.

2 b . An equation with separable variables (a separable equation) is generally represented by

 1 ( ) 1( )   =  2 ( ) 2 ( ).

Dividing the equation by  2 ( ) 1 ( ), one obtains a separated equation. Integrating yields:

X
x3y'z|{}~/

 1( )
 2( ) M

=X

2 ( )
 +L .
1 ( ) M

In termwise division of the equation by  2 ( ) 1 ( ), solutions corresponding to

 2 ( ) = 0 can be lost.

0.1.2-2. Equation of the form  =  (  +  ).


The substitution H =  + 
Paragraph 0.1.2-1.

brings the equation to a separable equation, H   =  ( H ) +  ; see

0.1.2-3. Homogeneous equations and equations reducible to them.


1 b . A homogeneous equation remains the same under simultaneous scaling (dilatation) of the
independent and dependent variables in accordance with the rule D , D , where is an
arbitrary constant ( 0). Such equations can be represented in the form

  = *O

3P

The substitution = 
 brings a homogeneous equation to a separable one,   =  ( ) ; see
Paragraph 0.1.2-1.

2003 by Chapman & Hall/CRC

Page 3

2 b . The equations of the form

 1 + 
 2 + 

+( 1
2 +( 2P
can be reduced to a homogeneous equation. To this end, for  1  +  1 , (  2  +  2 ), one should
use the change of variables =   0 , = 0 , where the constants  0 and 0 are determined
by solving the linear algebraic system

  = *O

 1

+
 2 0 + 
0

+(
0+(

1 0

= 0,
= 0.

As a result, one arrives at the following equation for = ( ):


 +  1
 = *O 1
.
 2 +  2 P
Dividing the argument of  by , one obtains a homogeneous equation whose right-hand side
depends on the ratio
only.
For  1  +  1 = , (  2  +  2 ), see the equation of Paragraph 0.1.2-2.
0.1.2-4. Generalized homogeneous equations and equations reducible to them.
1 b . A generalized homogeneous equation (a homogeneous equation in the generalized sense) remains the same under simultaneous scaling of the independent and dependent variables in accordance
with the rule |D , D  , where 0 is an arbitrary constant and , is some number. Such
equations can be represented in the form

 =  

 (   ).

The substitution =   brings a generalized homogeneous equation to a separable equation,


  =  ( ) ,! ; see Paragraph 0.1.2-1.
2 b . The equations of the form

 =  ( 

)
can be reduced to a generalized homogeneous equation. To this end, one should use the change of

variable H =  and set = , .
0.1.2-5. Linear equation.
A first-order linear equation is written as

  +  ( ) = ( ).
The solution is sought in the product form = " , where = ( ) is any function that satisfies
the truncated equation   +  ( ) = 0 [as ( ) one takes the particular solution =  , where
 = e ( )  ]. As a result, one obtains the following separable equation for = ( ): ( )  = ( ).
M
Integrating it yields the general solution:
( ) =  O X  ( )  + L

where

 = Xv ( )  .
M

0.1.2-6. Bernoulli equation.


A Bernoulli equation has the form

  +  ( ) = ( ) ,

 0, 1.
brings it to a linear equation, H   + (1  )  ( )H = (1  ) ( ), which is

The substitution H =
discussed in Paragraph 0.1.2-5. With this in view, one can obtain the general integral:
1

= L  + (1  )  X  ( )  ,
M

where

 = (1  ) Xv ( )  .
M

2003 by Chapman & Hall/CRC

Page 4

0.1.2-7. Equation of the form   =

The substitution = 

Paragraph 0.1.2-1.

+  ( ) (
 ).

brings the equation to a separable equation, 

 =  ( ) ( ); see

0.1.2-8. Darboux equation.


A Darboux equation can be represented as

*O

2P

+  %lO

S   = rO

3P

+ 

3P

%lO

3P

Using the substitution = "H ( ) and taking H to be the independent variable, one obtains a Bernoulli
equation, which is considered in Paragraph 0.1.2-6:

V ( H ) H! ( H )W   =   ( H ) + 

^_

+1

% ( H ).

References for Subsection 0.1.2: G. M. Murphy (1960), E. Kamke (1977), A. N. Tikhonov, A. B. Vasileva, and
A. G. Sveshnikov (1980), A. Moussiaux (1996), D. Zwillinger (1998).

0.1.3. Exact Differential Equations. Integrating Factor


0.1.3-1. Exact differential equations.
An exact differential equation has the form

 ( , )  + ( , )
M
M

= 0,

where


N

= N

The left-hand side of the equation is the total differential of a function of two variables ( , ).
The general integral, ( , ) = L , where the function is determined from the system:

= ,

= .

Integrating the first equation yields = e ( , )  + i ( ) (while integrating, the variable is


M
treated as a parameter). On substituting this expression into the second equation, one identifies the
function i (and hence, ). As a result, the general integral of an exact differential equation can be
represented in the form:

where 

and

X   ( , ) + X  ( 0 , ) = L ,
M
M
0
0

are arbitrary numbers.

0.1.3-2. Integrating factor.


An integrating factor for the equation

 ( , )  + ( , )
M
M

=0


,
*

is a function ( , ) 1 0 such that the left-hand side of the equation, when multiplied by ( , ),
becomes a total differential, and the equation itself becomes an exact differential equation.
An integrating factor satisfies the first-order partial differential equation:

lN
N

3N

= N

which is not generally easier to solve than the original equation.


Table 1 lists some special cases where an integrating factor can be found in explicit form.

^_

References for Subsection 0.1.3: G. M. Murphy (1960), N. M. Matveev (1967), E. Kamke (1977), A. N. Tikhonov,
A. B. Vasileva, and A. G. Sveshnikov (1980), A. Moussiaux (1996), D. Zwillinger (1998).

2003 by Chapman & Hall/CRC

Page 5

TABLE 1
An integrating factor = ( , ) for some types of ordinary differential
= 0, where  =  ( , ) and = ( , )
equations   +

No

Conditions for  and

1  = Y (
2
3
4

), = j (

  =  ,

'


'

Integrating factor

= 

  = 

Remarks

  *1 0;
Y ( H ) and j ( H ) are any functions

 +  is an analytic function
of the complex variable  +

2+ 2

= exp V Y ( )  W
M
V

= exp Y ( ) W
M
V

= exp Y ( H ) HW , H =  +
M
= exp V Y ( H ) HW , H = 
M
= exp V Y ( H ) H W , H = 
M
= exp V 12 Y ( H ) HW , H =  2 +
M
= exp V Y ( )  + j ( ) W
M
M
= exp V Y ( ) W
M

= Y ( )
=Y ( )


' = Y ( + )


' = Y ( )
6

 2 ( )
7
' = Yu  
 +

 ' = Y ( 2 + 2 )
8

9    = Y (  ) j ( ) 

' = Y ( )
10
'!
5

Y ( ) is any function
Y ( ) is any function
Y ( H ) is any function
Y ( H ) is any function
Y ( H ) is any function
2

Y ( H ) is any function
Y ( ) and j ( ) are any functions
= ( , ) is any function
of two variables

0.1.4. Riccati Equation


0.1.4-1. General Riccati equation. Simplest integrable cases.
A Riccati equation has the general form

  =  2 ( )

+  1 ( ) +  0 ( ).

(1)

If  2 0, we have a linear equation (see Paragraph 0.1.2-5), and if  0 0, we have a Bernoulli


equation (see Paragraph 0.1.2-6 for  = 2), whose solutions were given previously. For arbitrary
 2 ,  1 , and  0 , the Riccati equation is not integrable by quadrature.
Listed below are some special cases where the Riccati equation (1) is integrable by quadrature.
1 b . The functions  2 ,  1 , and 

are proportional, i.e.,

  = Y ( )( 

+

+ ( ),

where  ,  , and ( are constants. This equation is a separable equation; see Paragraph 0.1.2-1.
2 b . The Riccati equation is homogeneous:

 = 


2
2

+

+( .

See Paragraph 0.1.2-3.

2003 by Chapman & Hall/CRC

Page 6

3 b . The Riccati equation is generalized homogeneous:

  =  

+ ( 

See Paragraph 0.1.2-4 (with , = B 1). The substitution H =  


equation: "H   = !H 2 + (  + B + 1)H + ( .

+1

brings it to a separable

4 b . The Riccati equation has the form

  =  2 
By the substitution

= +

+ ( 2 + .

 H , the equation is reduced to a separable equation:   + H   = !H 2 + ( .

For other Riccati equations integrable by quadrature, see Section 1.2.


0.1.4-2. Polynomial solutions of the Riccati equation.
Let  2 = 1,  1 ( ), and  0 ( ) be polynomials. If the degree of the polynomial =  12 2(  1)  4  0
is odd, the Riccati equation cannot possess a polynomial solution. If the degree of is even, the
equation involved may possess only the following polynomial solutions:
= 12  

A V
W ,

where V W denotes an integer rational part of the expansion of


(for example, V  2 2 + 3 W =  1).

in decreasing powers of 

0.1.4-3. Use of particular solutions to construct the general solution.


1 b . Given a particular solution
written as:
=

0(

0(

 ) of the Riccati equation (1), the general solution can be

 ) + ` ( ) L X ` ( )  2 ( )  S
Q
M

where

` ( ) = exp /X V 2  2 ( ) 0 ( ) +  1 ( )W t
 .
M
To the particular solution 0 ( ) there corresponds L = F .
2 b . Let 1 = 1 ( ) and 2 = 2 ( ) be two different particular solutions of equation (1). Then the
general solution can be calculated by:
=

To the particular solution

+ ( )
L + ( )
1

1(

X  2( 1 2)  S .
( ) = exp v
Q
M

where

 ), there corresponds L = F ; and to

2(

 ), there corresponds L = 0.

3 b . Let 1 = 1 ( ), 2 = 2 ( ), and 3 = 3 ( ) be three distinct particular solutions of equation (1).


Then the general solution can be found without quadrature:

2
1

1
2

=L .

This means that the Riccati equation has a fundamental system of solutions.

2003 by Chapman & Hall/CRC

Page 7

0.1.4-4. Some transformations.


1 b . The transformation ( Y , j

1,

j 2 , j 3 , and j

are arbitrary functions)


j 4 ( ) + j 3 ( )
 = Y ( ),
=
j 2 ( ) + j 1 ( )
reduces the Riccati equation (1) to a Riccati equation for = ( ).
4

2 b . Let 0 = 0 ( ) be a particular solution of equation (1). Then the substitution


leads to a linear equation for = ( ):

+ 1


 + V 2  ( ) ( ) +  ( )W +  ( ) = 0.
2
0
1
2

For solution of linear equations, see Paragraph 0.1.2-5.


0.1.4-5. Reduction of the Riccati equation to a second-order linear equation.
The substitution

( ) = exp O Xv


M P

reduces the general Riccati equation (1) to a second-order linear equation:

 2    V (  2 )  +  1  2 W  +  0  22 = 0,

which often may be easier to solve than the original Riccati equation.
0.1.4-6. Reduction of the Riccati equation to the canonical form.
1 b . The general Riccati equation (1) can be reduced with the aid of the transformation

 = Y ( ),

1
1
=

 2
2

to the canonical form:


Here, the function i

1
2

1 1
+
,
2  2

 =

where

h ( ) =  ( Y ) Y  ,

+ i ( ).

(2)
(3)

is defined by the formula:

1 2 1  1  2
3  
1  2 
 1 +  1  1 2 +
;
4
2
2  2 4  2
2  2
the prime denotes differentiation with respect to .
Transformation (2) depends on a function Y = Y ( ) that can be arbitrary. For a specific original
Riccati equation, different functions Y in (2) will generate different functions i in equation (3).
[In practice, transformation (2) is most frequently used with Y ( ) = ].

i ( ) =  0 

2 b . In the special case where the original equation has the canonical form

 =

transformation (2) is written as

 = Y ( ),

+  ( ),


1
1 Y

,
Y 
2 ( Y  )2

and the transformed equation (3) is determined by the function


3 Y
i ( ) =  ( Y )(Y  )2 
4 Y


1 Y
.
2 Y 

If the original Riccati equation is integrable by quadrature, one may obtain, specifying different
functions Y , a variety of different integrable equations of the form (3). In Subsection 1.2.9, some
useful transformations are given for specific functions Y .

^_

References for Subsection 0.1.4: G. M. Murphy (1960), E. L. Ince (1964), W. T. Reid (1972), E. Kamke (1977),
W. E. Boyce and R. C. DiPrima (1986), A. D. Polyanin and V. F. Zaitsev (1995).

2003 by Chapman & Hall/CRC

Page 8

0.1.5. Abel Equations of the First Kind


0.1.5-1. General form of Abel equations of the first kind. Simplest integrable cases.
An Abel equation of the first kind has the general form

  =  3 ( )

+  2 ( )

+  1 ( ) +  0 ( ),

 3 ( ) 1 0.

(1)

In the degenerate case  2 ( ) =  0 ( ) = 0, we have a Bernoulli equation (see Paragraph 0.1.2-6


with  = 3). The Abel equation (1) is not integrable for arbitrary   ( ).
Listed below are some special cases where the Abel equation of the first kind is integrable by
quadrature.
1 b . If the functions   ( ) (B = 0, 1, 2, 3) are proportional, i.e.,   ( ) =   ( ), then (1) is a
separable equation (see Paragraph 0.1.2-1).
2 b . The Abel equation is homogeneous:
3

 = 


+

+(

+ .

See Paragraph 0.1.2-3.


3 b . The Abel equation is generalized homogeneous:

  =  2 

+ ' 

+1 3

+  

See Paragraph 0.1.2-4 for , = B 1. The substitution


 2 =  3 +  2 + ( ( + B + 1) + .

= 

+1

.
leads to a separable equation:

4 b . The Abel equation

  =  3  +
can be reduced with the substitution

5 b . Let  0 0,  1 0, and (  3
 2 ) 
leads to a separable equation:  = 
6 b . If

 0=

+ ' 2 

+  2+

M
+

=
H to a separable equation:   + H   = !H 3 + H 2 + ( .
=   2 for some constant  . Then the substitution =  2  31
2 1 3
+ 2 +  ).
2  3 (


 1  2 2  23 1

M 2,
3 3
27  32 3   3
M

  =   ( ),

then the solution of equation (1) is given by:


( ) = O L 2 Xv 3

1 2

M P

3

2
3

where

= exp Xv 1

3

2
2
3

 .

For other solvable Abel equations of the first kind, see Subsection 1.4.1.
0.1.5-2. Reduction to the canonical form. Reduction to an Abel equation of the second kind.
1 b . The transformation
= ( ) ( )

3

2
3

= X  3

 ,

where

( ) = exp X  1

3

2
2
3

 ,

brings equation (1) to the canonical (normal) form:

 = 3 + ` ( ).
2003 by Chapman & Hall/CRC

Page 9

Here, the function ` ( ) is defined parametrically ( is the parameter) by the relations:


1

` =
2 b . Let

0(

 3

 0


 1  2 2  23 1
+
+
M 2 ,
3 3
27  32 3   3
M

= X 3

 .

 ) be a particular solution of equation (1). Then the substitution


=

( )
,
H ( )

( ) = exp X (3 
Q

where

2
3 0

+ 2

2 0

+  1)  S ,

leads to an Abel equation of the second kind:

HH   = (3 

3 0

+  2 ) IH  3

For equations of this type, see Subsection 0.1.6.

^_

References for Subsection 0.1.5: E. Kamke (1977), G. M. Murphy (1960).

0.1.6. Abel Equations of the Second Kind


0.1.6-1. General form of Abel equations of the second kind. Simplest integrable cases.
An Abel equation of the second kind has the general form


[ + ( )]  =  2 ( )

+  1 ( ) +  0 ( ),

( ) 1 0.

(1)

The Abel equation (1) is not integrable for arbitrary   ( ). Given below are some special cases
where the Abel equation of the second kind is integrable by quadrature.
1 b . If ( ) = const and the functions   ( ) (B = 0, 1, 2) are proportional, i.e.,   ( ) =   ( ),
then (1) is a separable equation (see Paragraph 0.1.2-1).
2 b . The Abel equation is homogeneous:


( +  )  =


+

+ ( .

See Paragraph 0.1.2-3.


3 b . The Abel equation is generalized homogeneous:


( +   )   =

+ ' 

+ ( 2 

See Paragraph 0.1.2-4 for , = B . The substitution


 ( + )2 = (  B ) 2 + (  B ) + ( .

  leads to a separable equation:

4 b . The general solution of the Abel equation


( + )  = 

+

+  1  2 2,

  =   ( ), = ( ),

is given by:
= + Lu + X ( 
5 b . If 

+  2  2 )

 ,

where

= exp O X 

= 2  2   , the general solution of the Abel equation (1) has the form:
= *AC

^_

2 X (

+     2 2 )

 + L2S

1 2

where

 .
M P

= exp O X 

 .
M P

For other solvable Abel equations of the second kind, see Section 1.3.
References for Paragraph 0.1.6-1: G. M. Murphy (1960), E. Kamke (1977).

2003 by Chapman & Hall/CRC

Page 10

0.1.6-2. Reduction to the canonical form. Reduction to an Abel equation of the first kind.
1 b . The substitution

= ( + ) ,

= exp O X 

where

 ,
M P

(2)

brings equation (1) to the simpler form:

*  =  ( ) +  ( ),
1
0
where

= (


2  2 +  ) ,


= (

(3)

 1 +  2 2 )

2 b . In turn, equation (3) can be reduced, by the introduction of the new independent variable

H = X  1 ( )  ,
M

(4)

to the canonical form:

*  = ( H ).
Here, the function ( H ) is defined parametrically ( is the parameter) by the relations
 0 ( )
,
 1 ( )

(5)

H = X  1 ( )  .
M

Substitutions (2) and (4), which take the Abel equation to the canonical form, are called canonical.
x3y'z|{}~
The transformation =  , H =  H +  brings (5) to a similar equation, I
 =
1
 (  H +  ). Therefore the function (H ) in the right-hand side of the Abel equation (5) can be
identified with the two-parameter family of functions  1 ( !H +  ).

x3y'z|{}~
Any Abel equations of the second kind related by linear (in ) transformations

 = Y 1 ( ), = Y 2 ( ) + Y 3 ( ) have identical canonical forms (up to the two-parameter family of

functions specified in Remark 1).


3 b . The substitution

+ = 1
leads to an Abel equation of the first kind:

 + (  0  1 +  2 2 )

+ (

2  2 +   )

+  2 = 0.

For equations of this type, see Subsection 0.1.5.

^_

References for Paragraph 0.1.6-2: G. M. Murphy (1960), E. Kamke (1977), V. F. Zaitsev and A. D. Polyanin (1994).

0.1.6-3. Use of particular solutions to construct self-transformations.


1 b . Suppose a particular solution
Then the substitution = 1
(

0)

= 0 ( ) of an Abel equation of the second kind (1) is known.


leads to a similar Abel equation:

 
1
 = 0
0 +

" +

1
0

2
+

2 0


0

(6)

If  0 0, equation (1) has the trivial particular solution 0 = 0. In this case, the change of
variable = 1
 leads to an Abel equation of the form (6) with 0 = 0.
2 b . Given a particular solution

the substitution

5 ( )

2
0( 0

 ) of the Abel equation of the second kind


!   =  ( ) +  ( ),
1
0

0(

where

5 ( ) = exp X

1
0

 ,
M

(7)
(8)

2003 by Chapman & Hall/CRC

Page 11

brings (7) to another, similar Abel equation:

*  = ( ) + ( ).
1
0
Here, the functions

1(

 ) and

(

1 0

0(

 ) are defined by

+ 3  0)5
4
0

(9)

 05

2
6
0

It is not difficult to verify by direct substitution that equation (9) has a particular solution:

( ) = 5 ( ) .
0
2
0 ( )

(10)

The transformation based on the particular solution (10) brings the Abel equation (9) to the original
equation (7) with  1 having the opposite sign.
x3y'z|{}~E
In general, the canonical forms of equations (1) and (6) and also those of equations
(7) and (9) are different. See Paragraph 0.1.6-2.

x3y'z|{}~E

Given , distinct particular solutions  of equation (7), , distinct Abel equations


of the second kind related to (7) by known substitutions of the form (8) can be constructed.

^_

References for Paragraph 0.1.6-3: T. A. Alexeeva, V. F. Zaitsev, and T. B. Shvets (1992), V. F. Zaitsev and A. D. Polyanin
(1994).

0.1.6-4. Use of particular solutions to construct the general solution.


For some Abel equations of the second kind, the general solution can be found if B its distinct
particular solutions  =  ( ), , = 1,  , B , are known.
Below we consider Abel equations of the canonical form

!   = ( ),

(11)

whose general solutions can be represented in the special form:




=1

|  ( )| + = L .

(12)

Here, the particular solutions  =  ( ) correspond to L = 0 (if  > 0) and L = F (if  < 0).

The logarithmization of (12), followed by the differentiation of the resulting expression and
rearrangement, leads to the equation

=1

 1

(    ) (  )  
`
=1
=1

= 0,

(13)

=1


= ( )  . We require that equation (13) be equivalent to the Abel equation (11). To this
where

end, we set:

iu = `l ,

iu

= ( ) `l

and equate the other `l and i3 with zero.

Selecting different values @ = 1, 2,  , B 1, we obtain B 1 systems of differential-algebraic


equations; only one of the systems, corresponding to  0 for all , = 1,  , B and

for , leads to a nondegenerate solution of the form (12). Consider the Abel equations (11)
corresponding to the simplest solutions of the form (12) in more detail.

2003 by Chapman & Hall/CRC

Page 12

1 b . Case B = 2. The system of differential-algebraic equations has the form:


1

where 8

1 1 2

=8

2+

1+

2
2

,
=
0,
1
 =8 ,
2

2 1 2 = 8 (  ),

(14)

is an arbitrary constant. It follows from the second and third equations that
1

where

2
1

2
2

( 8 + ),

2
1

2
2

( 8 + ),

is an arbitrary constant. Introducing the new constants


( + )
( +
)
# = 1 2 1 2 2 8 , $ = 1 2 1 2 2 ,
( 1 2)
( 1 2)

we find from the last relation in (14) that

( ) = #* + $ ,

(15)

which means that for B = 2 the right-hand side of the Abel equation is a linear function of  (see
equation 1.3.1.2).
The particular solutions 1 , 2 , and the corresponding exponents 1 , 2 in the general

integral (12), are expressed in terms of the coefficients # , $ on the right-hand side (15) of the
Abel equation (11) as follows:
1 + 4# + 1
( #* + $ ),
= 2 # + 1 + 4 # + 1,
1
2#
1 + 4# + 1
( #* + $ ),
= 2# .
2 =
2
2# + 1 + 4# + 1
2 b . Case B = 3. Equation (13) with B = 3 leads to the Abel equation (11) with the right-hand side
1

( ) = 29  + # + $I

1 2

(16)

(see equation 1.3.1.3).


The particular solutions
in the general
integral (12) are expressed as:

and the exponents

2
2
3
2#
$
,
=  +  1J 2 +
,
=

3
3

3(2 2 3 # )
where the are roots of the cubic equation

92 3
# 92 $

= 1, 2, 3.

= 0,

3 b . Case B = 4. Equation (13) with B = 4 leads to the Abel equation (11) with the right-hand side

( ) = 163  + #*

J + $I

1 3

5 3

(see equation 1.3.3.61).


The particular solutions and the exponents in (12) are expressed as:
1,2

= 34 u
 AE 3 # + 32 3 $p 1 J 3 + 3 $I

1 3

J ,

3,4

= 34 u
 AE 3 # 32 3 $p 1 J 3 3 $I

1 3

J ,

1,2

= T  2 # 3 $  ,

3,4

= A 4#

+ 3$ .

are algebraic equations of degree B and, in the general case,


4 b . Case B > 4. The equations for
are not soluble in radicals. The right-hand of equation (11) is expressed as
B 1
( ) = 2  + ( ),
where the function ( ) is bounded as EDGF

^_

can be specified in parametric form).

References for Paragraph 0.1.6-4: B. M. Koyalovich (1894), T. A. Alexeeva, V. F. Zaitsev, and T. B. Shvets (1992).

2003 by Chapman & Hall/CRC

Page 13

0.1.7. Equations Not Solved for the Derivative


0.1.7-1. The method of integration by differentiation.
In the general case, a first-order equation not solved for the derivative,

 ( , ,   ) = 0,

(1)

can be rewritten in the equivalent form

a =  .

 ( , , a ) = 0,
We look for a solution in parametric form:  =  (a ),
in (2), the differential of  is given by:

 
M

 + 
M

(2)

= (a ). In accordance with the first relation

+ g

a = 0.

(3)

= a  , we eliminate successively
and  from (3). As a result, we obtain
Using the relation
M
M
M
M
the system of two first-order ordinary differential equations:

M
M


a

g
,

 + a 
M

ag
.

 + a 

(4)

By finding a solution of this system, one thereby obtains a solution of the original equation (1) in
parametric form,  =  (a ), = (a ).
x3y'z|{}~
The application of this method may lead to loss of individual solutions; this issue
should be additionally investigated.
0.1.7-2. Equations of the form

=  (   ).

This equation is a special case of equation (1), with  ( , , a ) =


in Paragraph 0.1.7-1 yields

  (a )
,
=  (a ).
M =

 (a ). The procedure described


(5)

Here, the original equation is used instead of the second equation in system (4); this is valid because
the first equation in (4) does not depend on explicitly.
Integrating the first equation in (5) yields the solution in parametric form:

 =X

  (a )
a +L ,
aM

=  (a ).

0.1.7-3. Equations of the form  =  (   ).


This equation is a special case of equation (1), with  ( , , a ) =   (a ). The procedure described
in Paragraph 0.1.7-1 yields

 =  (a ),
M

= a  (a ).

(6)

Here, the original equation is used instead of the first equation in system (4); this is valid because
the second equation in (4) does not depend on  explicitly.
Integrating the second equation in (5) yields the solution in parametric form:

 =  (a ),

= Xa  (a ) a + L .

2003 by Chapman & Hall/CRC

Page 14

=   +  (  ).

0.1.7-4. Clairauts equation

Clairauts equation is a special case of equation (1), with  ( , , a ) = a  (a ). It can be rewritten
as
= a +  (a ),
a =  .
(7)
and   in

= 

 by their values in accordance with (7), we obtain


[ +   (a )] a = 0.
M

This equation splits into a = 0 and  +  (a ) = 0. The solution of the first equation is obvious:
a = L ; it gives the generalM solution of Clairauts equation,

Substituting

= L3 +  ( L ),

(8)

which is a family of straight lines. The second equation generates a solution in parametric form,

 =   (a ),

= a  (a ) +  (a ),

which is a singular solution and is the envelope of the family of lines (8).
0.1.7-5. Lagranges equation

=   ( ) + (  ).

Lagranges equation is a special case of equation (1), with  ( , , a ) =   (a ) (a ). In the special


case  (a ) a , it coincides with Clairauts equation; see Paragraph 0.1.7-4.
The procedure described in Paragraph 0.1.7-1 yields

M
M


a

  (a )
 (a )
 =
,
 (a ) a
a  (a )

=   (a ) + (a ).

(9)

Here, the original equation is used instead of the second equation in system (4); this is valid because
the first equation in (4) does not depend on explicitly.
The first equation of system (9) is linear. Its general solution has the form  = Y (a ) L + j (a );
the functions Y and j are defined in Paragraph 0.1.2-5. Substituting this solution into the second
equation in (9), we obtain the general solution of Lagranges equation in parametric form:

 = Y (a ) L + j (a ),

= V Y (a ) L + j (a )W  (a ) + (a ).

x3y'z|{}~

With the above method, solutions of the form = a   + (a  ), where the a  are
roots of the equation  (a ) a = 0, may be lost. These solutions can be particular or singular solutions
of Lagranges equation.

^_

References for Subsection 0.1.7: G. M. Murphy (1960), E. Kamke (1977).

0.1.8. Contact Transformations


0.1.8-1. General form of contact transformations.
A contact transformation has the form

 =  (m , n , n o  ),

= 4 (m , n , n o ),

(1)

where the functions  (m , n , ) and 4 (m , n , ) are chosen so that the derivative  does not
depend on n ol  o :


 = o
 o

4po + 4Iqrn
ho + s
 qrn



o + 4Csn olo


o + tsn olo

= 5 (m , n , n o  ).

(2)

2003 by Chapman & Hall/CRC

Page 15

The subscripts m , n , and after  and 4


that t\1 0 and 4C|1 0).
It follows from (2) that the relation

OrN

m
N
N
N

denote the respective partial derivatives (it is assumed

4


O N
r
N
n P

m
N
N
N

+ N

holds; the derivative is calculated by

+ N

4
=0
n P

(3)

4C
,
t

 =

(4)

where 4C
t\1 const.
The application of contact transformations preserves the order of differential equations. The
inverse of a contact transformation can be obtained by solving system (1) and (4) for m , n , n o  .

^_

References for Paragraph 0.1.8-1: D. Zwillinger (1989).

0.1.8-2. A method for the construction of contact transformations.


Suppose the function  =  (m , n , ) in the contact transformation (1) is specified. Then relation (3)
can be viewed as a linear partial differential equation for the second function 4 . The corresponding
characteristic system of ordinary differential equations (see A. D. Polyanin, V. F. Zaitsev, and
A. Moussiaux, 2002),

admits the obvious first integral:

= M

t

 o + usq
h
M

 (m , n , ) = L 1 ,

(5)

where L 1 is an arbitrary constant. It follows that, to obtain the general representation of the function
4 = 4 (m , n , ), one has to deal with the ordinary differential equation

n o = ,

(6)

whose right-hand side is defined in implicit form by (5). Let the first integral of equation (6) has the
form
` (m , n , L 1 ) = L 2 .
Then the general representation of 4

= 4 (m , n , ) in transformation (1) is given by:

4 = i (  , ` ),

where i (  , ` ) is an arbitrary function of two variables, 

=  (m , n , ), and ` = ` (m , n ,  ).

0.1.8-3. Examples of contact transformations linear in the derivative.


Example 1. Legendre transformation (E. Kamke, 1977):

t
= 
=

,
,

u
=  ,
=

 =
t =

(direct transformation);
(inverse transformation).

Example 2. Contact transformation ( 0):

2

+  ,
ln


t
,
1

(direct transformation);
(inverse transformation).

It is apparent from this example that a contact transformation that is linear in the derivative can have a
nonlinear inverse, which is also a contact transformation.

2003 by Chapman & Hall/CRC

Page 16

Example 3. Contact transformation ( 1):

= (

1
+1

3 +1

=
=

1
(
+1

3 ,

)(

3 +1

= +
( + 1)

+1 ,

(direct transformation);
(inverse transformation).

Example 4. Contact transformation:

= ( ) and

where

+  ,

= (  +  ) X

+ (  ' )

+ 

= ( ) are arbitrary functions,

= exp

Xe

Example 5. Contact transformation:

where

= ( ) and

+ 

+ ,

= ( ) are arbitrary functions,

X '
X

= ,

0.1.8-4. Examples of contact transformations nonlinear in the derivative.


Example 1. Contact transformation:

+ ,

)2 +  ,

1
=
2 (
2 2

1
(
2

)2

= 2

(direct transformation);

(inverse transformation).

Example 2. Contact transformation:

1
l

)2

),

= ln

= (

)2


= 2
2

= 2

(direct transformation);
(inverse transformation).



=
t

2(

Example 3. Contact transformation

(direct transformation);
(inverse transformation).

Example 4. Contact transformation:

= ( t )2

cosh

sinh ,

cosh
2

cos

+ sin ,

cos
2

Example 5. Contact transformation:

= ( )2 +

Example 5. Contact transformation (  0):

= ( )2 ,
( )2 1 ,
=
9

= 2 ( r )3 3 ,
2
1
,
=
(  )3
81
3

= 3
1
= 
3

(direct transformation);
(inverse transformation).

2003 by Chapman & Hall/CRC

Page 17

Example 7. Contact transformation (D. Zwillinger, 1989, p. 169):

rt ,

 ,

= 3 (
=


( )2 1

= 2

( ) 1

)2 1,

)2 1,

(direct transformation);
(inverse transformation).

Example 8. Contact transformation (D. Zwillinger, 1989, p. 169):


,
( )2 + 1

,
( )2 + 1

=
=

,  = t

( )2 + 1

, =
( )2 + 1

Example 9. Contact transformation (  0,

= ( ) ,

= ( )
( + 1)

1):
+1

=  ( ) ( + 1) , 
+1
=  ( ) +2 ,
( + 1)
( + 1)

(direct transformation);
(inverse transformation).

= ( + 1)
=

+1

(direct transformation);
(inverse transformation).

0.1.9. Approximate Analytic Methods for Solution of Equations


0.1.9-1. The method of successive approximations (Picard method).
The method of successive approximations consists of two stages. At the first stage, the Cauchy
problem

  =  ( , )
( 0 ) = 0
is reduced to the equivalent integral equation:
( ) =

+X 


0

(equation),
(initial condition)

(1)
(2)

 (a , (a )) a .
M

(3)

Then a solution of equation (3) is sought using the formula of successive approximations:

 +1 ( ) =

0+X 

 (a ,  (a )) a ;
M

B = 0, 1, 2, 

The initial approximation 0 ( ) can be chosen arbitrarily; the simplest way is to take 0 to be a
number. The iterative process converges as BwDF , provided the conditions of the theorems in
Paragraph 0.1.1-2 are satisfied.
0.1.9-2. The method of Taylor series expansion in the independent variable.
A solution of the Cauchy problem (1)(2) can be sought in the form of the Taylor series in powers
of (  0 ):
  ( )
0
(  0 )2 + ))) .
( ) = ( 0 ) +   ( 0 )(  0 ) +
(4)
2!
The first coefficient ( 0 ) in solution (4) is prescribed by the initial condition (2). The values of
the derivatives of ( ) at  =  0 are determined from equation (1) and its derivative equations
(obtained by successive differentiation), taking into account the initial condition (2). In particular,
setting  =  0 in (1) and substituting (2), one obtains the value of the first derivative:

  ( 0 ) =  ( 0 , 0 ).

(5)

2003 by Chapman & Hall/CRC

Page 18

Further, differentiating equation (1) yields

    =   ( , ) +  ( , )   .


(6)

On substituting  =  0 , as well as the initial condition (2) and the first derivative (5), into the
right-hand side of this equation, one calculates the value of the second derivative:

    ( 0 ) =   ( 0 , 0 ) +  ( 0 , 0 )  ( 0 , 0 ).

Likewise, one can determine the subsequent derivatives of at  =  0 .

Solution (4) obtained by this method can normally be used in only some sufficiently small
neighborhood of the point  =  0 .
0.1.9-3. The method of regular expansion in the small parameter.
Consider a general first-order ordinary differential equation with a small parameter :

  =  ( , , ).

(7)

Suppose the function  is representable as a series in powers of :

 ( , , ) =    ( , ).
 =0
ID

(8)

One looks for a solution of the Cauchy problem for equation (7) with the initial condition (2) as
0 in the form of a regular expansion in powers of the small parameter:

=0

 n  ( ).

(9)

Relation (9) is substituted in equation (7) taking into account (8). Then one expands the functions  
into a power series in and matches the coefficients of like powers of to obtain a system of
equations for n  ( ):

n 0 =  0 ( , n 0 ),
n 1 = ( , n 0 ) n 1 +  1 ( , n 0 ),

( , ) = N

(10)
0

(11)

Only the first two equations are written out here. The prime denotes differentiation with respect
to  . The initial conditions for n  can be obtained from (2) taking into account (9):

n 0 ( 0 ) =

0,

n 1 ( 0 ) = 0.

Success in the application of this method is primarily determined by the possibility of constructing a solution of equation (10) for the leading term in the expansion of n 0 . It is significant
that the remaining terms of the expansion, n  with B 1, are governed by linear equations with
homogeneous initial conditions.
x3y'z|{}~
Paragraph 0.3.3-2 gives an example of solving a Cauchy problem by the method
of regular expansion for a second-order equation and also discusses characteristic features of the
method.

x3y'z|{}~

The methods of scaled coordinates, two-scale expansions, and matched asymptotic


expansions are also used to solve problems defined by first-order differential equations with a small
parameter. The basic ideas of these methods are given in Subsection 0.3.4.

^_

References for Subsection 0.1.9: G. M. Murphy (1960), G. A. Korn and T. M. Korn (1968), E. Kamke (1977), D. Zwillinger
(1998).

2003 by Chapman & Hall/CRC

Page 19

0.1.10. Numerical Integration of Differential Equations


0.1.10-1. The method of Euler polygonal lines.
Consider the Cauchy problem for the first-order differential equation

  =  ( , )
with the initial condition ( 0 ) = 0 . Our aim is to construct an approximate solution = ( ) of
this equation on an interval [ 0 , / ].
/  0
. We seek
Let us split the interval [ 0 , / ] into B equal segments of length C =
B
approximate values 1 , 2 ,  ,  of the function ( ) at the partitioning points  1 ,  2 ,  ,   = / .
For a given initial value 0 = ( 0 ) and a sufficiently small C , the values of the unknown
function  = (  ) at the other points   =  0 + , C are calculated successively by the formula:

+1

=  +  (  ,  ) C

(Euler polygonal line),

where , = 0, 1,  , B 1. The Euler method is a single-step method of the first-order approximation
(with respect to the step C ).
0.1.10-2. Single-step methods of the second-order approximation.
Two single-step methods for solving the Cauchy problem in the second-order approximation are
specified by the recurrence formulas:

+1

+1

=  +     + 12 C ,  + 12   C ) C ,
=  + 12 V   +  (  +1 ,  +   C
 )WC
 ,

where   =  (  ,  ); , = 0, 1,  , B 1.
0.1.10-3. RungeKutta method of the fourth-order approximation.
This is one of the widely used methods. The unknown values  are successively found by the
formulas:
 ,
 +1 =  + 16 (  1 + 2  2 + 2  3 +  4 ) C
where

x3y'z|{}~

 1 =  (  ,  ),
 2 =  (  + 12 C ,  + 12  1 C ),
 3 =  (  + 12 C
 ,  + 12  2 C ),
 4 =  (  + C ,  +  3 C ).

All methods described in Subsection 0.1.10 are special cases of the RungeKutta
method (a detailed description of this method can be found in the monographs listed below).

x3y'z|{}~

In practice, calculations are performed on the basis of any of the above recurrence
formulas with two different steps C , 12 C and an arbitrarily chosen small C . Then one compares
the results obtained at common points. If these results coincide within the given order of accuracy,
one assumes that the chosen step C ensures the desired accuracy of calculations. Otherwise, the
step is halved and the calculations are performed with the steps 12 C and 14 C , after which the
results are compared again, etc. (Quite often, one compares the results of calculations with steps
varying by ten or more times.)

^_

References for Subsection 0.1.10: G. A. Korn and T. M. Korn (1968), N. S. Bakhvalov (1973), E. Kamke (1977),
D. Zwillinger (1998).

2003 by Chapman & Hall/CRC

Page 20

0.2. Second-Order Linear Differential Equations


0.2.1. Formulas for the General Solution. Some Transformations
0.2.1-1. Homogeneous linear equations. Various representations of the general solution.
1 b . Consider a second-order homogeneous linear equation in the general form:

 2 ( )    +  1 ( )   +  0 ( ) = 0.

(1)

The trivial solution, = 0, is a particular solution of the homogeneous linear equation.


Let 1 ( ), 2 ( ) be a fundamental system of solutions (nontrivial linearly independent particular
solutions) of equation (1). Then the general solution is given by:
= L 1 1 ( ) + L 2 2 ( ),
(2)
where L 1 and L 2 are arbitrary constants.
2 b . Let 1 = 1 ( ) be any nontrivial particular solution of equation (1). Then its general solution
can be represented as:
=

OL

+L

3 b . Consider the equation

2
1

 ,
M P

where

 =X


1
2

 .

(3)

    +  ( ) = 0,

which is written in the canonical form; see Paragraph 0.2.1-3 for the reduction of equations to this
form. Let 1 ( ) be any nontrivial partial solution of this equation. The general solution can be
constructed by formula (3) with  = 0 or formula (2) in which

[  ( ) 1][ 12 ( 1 )2 ]
1

+
2 ( ) = 1
 2.
2
M
[ 12 + ( 1 )2 ]2
1 + ( 1)
Here, 1 = 1 ( ) and the prime denotes differentiation with respect to  . The last formula is suitable
where 1 vanishes at some points.
0.2.1-2. Wronskian determinant and Liouvilles formula.
The Wronskian determinant (or Wronskian) is defined by:

( ) 2 ( )
( ) = 1
1 ( 2 )  2 ( 1 )  ,
1 ( ) 2 ( )

of equation (1).
where 1 ( ), 2 ( ) is a fundamental system of solutions
Liouvilles formula:
  (a )

1
a .
( ) = ( 0 ) exp X 
0  2 (a ) M
0.2.1-3. Reduction to the canonical form.
1 b . The substitution

1
X
2
brings equation (1) to the canonical (or normal) form:
= ( ) exp

1
2


M

(4)
2

1  1
1  1


.
 2 4  2
2  2 
2 b . The substitution (4) is a special case of the more general transformation ( Y
function)
1
 1 (Y )
Y ,
 = Y ( ),
= ( ) | Y  ( )| exp X
2
 2 (Y ) M
which also brings the original equation to the canonical form.

   +  ( ) = 0,

where

 =

(5)
is an arbitrary

2003 by Chapman & Hall/CRC

Page 21

0.2.1-4. Reduction to the Riccati equation.

The substitution =  
 brings the second-order homogeneous linear equation (1) to the Riccati
equation:
 2 ( )  +  2 ( ) 2 +  1 ( ) +  0 ( ) = 0,
which is discussed in Subsection 0.1.4.
0.2.1-5. Nonhomogeneous linear equations. The existence theorem.
A second-order nonhomogeneous linear equation has the form

 2 ( )    +  1 ( )   +  0 ( ) = ( ).
(6)
Existence and uniqueness theorem. On an open interval  <  <  , let the functions  2 ,  1 ,  0 ,
and be continuous and  2 0. Also let
  ( 0 ) = $
( 0 ) = # ,
be arbitrary initial conditions, where  0 is any point such that  <  0 <  , and # and $ are arbitrary
prescribed numbers. Then a solution of equation (6) exists and is unique. This solutions is defined
for all Ed ( ,  ).
0.2.1-6. Nonhomogeneous linear equations. Various representations of the general solution.
1 b . The general solution of the nonhomogeneous linear equation (6) is the sum of the general solution
of the corresponding homogeneous equation (1) and any particular solution of the nonhomogeneous
equation (6).
2 b . Let 1 = 1 ( ), 2 = 2 ( ) be a fundamental system of solutions of the corresponding homogeneous equation, with 0. Then the general solution of equation (6) can be represented
as:


= L 1 1 + L 2 2 + 2 X 1 M 1 X 2
(7)
M ,
where


1 ( 2)


2 ( 1 )  is the Wronskian determinant.

3 b . Given a nontrivial particular solution 1 = 1 ( ) of the homogeneous equation (with 0), a


second particular solution 2 = 2 ( ) can be calculated from the formula:
2

2
1

 ,

where

 =X


1
2

 ,

=  .

(8)

Then the general solution of equation (6) can be constructed by (7).


Subsections 2.1.22.1.8 present only homogeneous equations; the solutions of the corresponding
nonhomogeneous equations can be obtained using (7) and (8).
4 b . Let 1 and 2 be respective solutions of the nonhomogeneous differential equations L [ 1 ] =
1 ( ) and L [ 2 ] = 2 ( ), which have the same left-hand side but different right-hand sides; where
L [ ] is the left-hand side of equation (6). Then the function = 1 + 2 is a solution of the equation
L [ ] = 1 ( ) + 2 ( ).
0.2.1-7. Reduction to a constant coefficient equation (a special case).
Let 

= 1, 

0, and the condition


1
1
=  = const
| 0| +
M
|  0 | C
|  0|

|  |  leads to a constant coefficient linear equation,


0M
 +   + sign  = 0.
0

be satisfied. Then the substitution = X

2003 by Chapman & Hall/CRC

Page 22

0.2.1-8. KummerLiouville transformation.


The transformation

 = (a ),

= (a ) H + = (a ),

(9)

where (a ), (a ), and = (a ) are arbitrary sufficiently smooth functions (1 0), takes any linear
differential equation for ( ) to a linear equation for H = H (a ). In the special case = 0, a
homogeneous equation is transformed to a homogeneous one.
Special cases of transformation (9) are widely used to simplify second- and higher-order linear
differential equations.

^_

References for Subsection 0.2.1: G. A. Korn and T. M. Korn (1968), E. Kamke (1977), A. D. Polyanin and V. F. Zaitsev
(1995), S. Yu. Dobrokhotov (1998), D. Zwillinger (1998).

0.2.2. Representation of Solutions as a Series in the Independent


Variable
0.2.2-1. Equation coefficients are representable in the ordinary power series form.
Let us consider a homogeneous linear differential equation of the general form

    +  ( )   + ( ) = 0.

(1)

Assume that the functions  ( ) and ( ) are representable, in the vicinity of a point  =  0 ,
in the power series form,

 ( ) = #  (  0 )  ,
 =0

( ) = $  (  0 )  ,
 =0

(2)

on the interval |  0 | < , where stands for the minimum radius of convergence of the two series
in (2). In this case, the point  =  0 is referred to as an ordinary point, and equation (1) possesses
two linearly independent solutions of the form:
1(

 ) =   (  0 )  ,
 =0

2(

 ) =   (  0 )  .
 =0

(3)

The coefficients   and   are determined by substituting the series (2) into equation (1) followed
by extracting the coefficients of like powers of (  0 ).*
0.2.2-2. Equation coefficients have poles at some point.
Assume that the functions  ( ) and ( ) are representable, in the vicinity of a point  =  0 , in the
form

 ( ) = #  (  0 )  , ( ) = $  (  0 )  ,
(4)
 =1
 =2
on the interval |  0 | < . In this case, the point  =  0 is referred to as a regular singular point.
Let 1 and 2 be roots of the quadratic equation

2
1

+ (#

1) + $

= 0.

There are three cases, depending on the values of the exponents of the singularity.
* Prior to that, the terms containing the same powers (

0)

= 0, 1, , should be collected.

2003 by Chapman & Hall/CRC

Page 23

1. If 1
of the form:

and

is not an integer, equation (1) has two linearly independent solutions


1(

 ) = |  0 | 

2 ( ) = |  0 | 

2. If

1+

=1

1+

=1

  (  0 )  S ,
(5)

  (  0 )  S .

= , equation (1) possesses two linearly independent solutions:

 ) = |  0 |  1 +   (  0 )  S ,
Q
 =1

  (  0 )  .
2 (  ) = 1 (  ) ln |   0 | + |   0 | 
 =0
1(

3. If 1 = 2 + , where
solutions of the form:
1(

2(

is a positive integer, equation (1) has two linearly independent

 ) = |  0 | 
 )=,

1(

1+

=1

  (  0 )  S ,

 ) ln |  0 | + |  0 | 

=0

  (  0 )  ,

where , may be equal to zero.


To construct the solution in each of the three cases, the following procedure should be performed:
substitute the above expressions of 1 and 2 into the original equation (1) and equate the coefficients
of (  0 )  and (  0 )  ln |  0 | for different values of B to obtain recurrence relations for the
unknown coefficients. From these recurrence relations the solution sought can be found.

^_

References for Subsection 0.2.2: G. M. Murphy (1960), G. A. Korn and T. M. Korn (1968), E. Kamke (1977), D. Zwillinger
(1989).

0.2.3. Asymptotic Solutions


This subsection presents asymptotic solutions,as hD 0 ( > 0), of some second-order linear ordinary
differential equations containing arbitrary functions (sufficiently smooth), with the independent
variable being real.
0.2.3-1. Equations not containing  . Leading asymptotic terms.
1 b . Consider the equation

     ( ) = 0

(1)

on a closed interval    .
Case 1. With the condition  0, the leading terms of the asymptotic expansions of the
fundamental system of solutions, as ID 0, are given by the formulas:

J exp O 1 X
  ,

M P
1
1 J 4
cos O X
  ,
1 = (  )

M P
1

=

1 4

J exp O 1 X
 

M P
1
1 J 4
sin O X
 
2 = (  )

M P
2

=

1 4

if  > 0,
if  < 0.

2003 by Chapman & Hall/CRC

Page 24

Case 2. Discuss the asymptotic solution of equation (1) in the vicinity of the point  =  0 ,
where function  ( ) vanishes,  ( 0 ) = 0 (such a point is referred to as a transition point). We
assume that the function  can be presented in the form

 ( ) = (

 )j ( ),

In this case, the fundamental solutions, as ID





1
|  ( )|1 J



j ( ) > 0.

0, are described by three different formulas:


X 
|  ( )|  +  S
Q
M
4
0
1


Ai( H )
[ j ( 0 )]1 J 6
0
1
1
exp

 ( )  S
X
Q 
M
2[  ( )]1 J 4

1
1
cos
|  ( )|  +  S
X
Q 0
M
4
|  ( )|1 J 4


sin

where







Bi( H )
[ j ( 0 )]1 J 6

1
1
X
exp
Q 
[  ( )]1 J 4

if  

> ,

if |  0 | > ,
if 

 > ,

if  

> ,

if |  0 | > ,
0

 ( )  S
M

if 

 > ,

where Ai( H ) and Bi( H ) are the Airy functions of the first and second kind, respectively (see equation
2.1.2.2), H = 2 J 3 [j ( 0 )]1 J 3 ( 0  ), and > = [ ( 2 J 3 ).
0.2.3-2. Equations not containing  . Two-term asymptotic expansions.
The two-term asymptotic expansions of the solution of equation (1) with  > 0, as \D
closed interval    , has the form:


1
J
exp
O

  1
X
1
0
M P

1
1 J 4
exp O X 
  1 + X 
2 = 

M P
0
=

1 4

 1    
5 (    )2

S  + [ ( 2 ) ,
X 
3
J
2
32  5 J 2 M
0 Q 8 
 1    
5 (    )2

S  + [ ( 2 ) ,
3J 2
32  5 J 2 M
0 Q 8 

0, on a

(2)

where  0 is an arbitrary number satisfying the inequality   0  .


The asymptotic expansions of the fundamental system of solutions of equation (1) with  < 0
are derived by separating the real and imaginary parts in either formula (2).
0.2.3-3. Equations of special form not containing " .
Consider the equation

     +

 ( ) = 0

(3)

on a closed interval    , where  < 0 and  > 0, under the conditions that is a positive

integer and  ( ) 0. In this case, the leading term of the asymptotic solution, as D 0, in
the vicinity of the point  = 0 is expressed in terms of a simpler model equation, which results
from substituting the function  ( ) in equation (3) by the constant  (0) (the solution of the model
equation is expressed in terms of the Bessel functions of order 1
, see equation 2.1.2.7).

We specify below formulas by which the leading terms of the asymptotic expansions of the
fundamental system of solutions of equation (3) with  <  < 0 and 0 <  <  are related (excluding
a small vicinity of the point  = 0). Three different cases can be extracted.

2003 by Chapman & Hall/CRC

Page 25

1 b . Let

be an even integer, and  ( ) > 0. Then,

[  ( )]1 J 4 exp

 ( )  S

M
1
1
1 J 4
, [  ( )]
exp
 ( )  S
X
Q 0
M

1
[  ( )]1 J 4 exp X
 ( )  S
Q 0 
M
1
1 J 4
, [  ( )]
exp X
 ( )  S
Q 0
M







where  =  ( ), , = sin O

.
where  =  ( ), , = tan O
2 P

be an odd integer. Then,


3 b . Let




|  ( )|1 J 4 cos





if  < 0,
if  > 0,

( )|  +

4 S

|  ( )| 

( )| 

( )|  +

if  < 0,

4 S

4 S

4 S

if  > 0,
if  < 0,
if  > 0,

4 S

if  < 0,

 ( )  S
X
Q 0
M

1
|  ( )|1 J 4 cos X
|  ( )|   S
Q 0 
M
4
1
1 J 4
, [  ( )]
exp X
 ( )  S
Q 0
M

if  > 0,

1
,
2

if  > 0,

be an even integer, and  ( ) < 0. Then,




 |  ( )|1 J 4 cos Q 1 X 0 | 


1 =
1
, 1 |  ( )|1 J 4 cos X
Q 0

1
1 J 4
 |  ( )| cos Q X 0 | 

=

2
1
, |  ( )|1 J 4 cos X
|
Q 0

2 b . Let

if  < 0,

where  =  ( ), , = sin O


2

[  ( )]1 J 4 exp

|  ( )|  +

if  < 0,
if  > 0,

0.2.3-4. Equations not containing  . Equation coefficients are dependent on .


Consider an equation of the form

     ( , ) = 0
(4)
on a closed interval    under the condition that  0. Assume that the following asymptotic

relation holds:

 (  , ) =   (  )  ,
 =0

ID

0.

Then the leading terms of the asymptotic expansions of the fundamental system of solutions of
equation (4) are given by the formulas:
1
J  ) exp 1 X
 ( )  + X
Q
0 M
2

=

1 4
(
0

=

1 4
(
0

1
J  ) exp 1 X
 0 ( )  + X
Q

M
2

 1 ( )
V
 0 (  ) M  S 1 + [ ( )W ,
 1 ( )
V
 0 (  ) M  S 1 + [ ( )W .

2003 by Chapman & Hall/CRC

Page 26

0.2.3-5. Equations containing  .


1 b . Consider an equation of the form

   + ( )   +  ( ) = 0
on a closed interval 0  1. With ( ) > 0, the asymptotic solution of this equation, satisfying
the boundary conditions (0) = L 1 and (1) = L 2 , can be represented in the form:
= (L
where , = exp X

1
0

, L

2 ) exp

(0) W + L

exp X 
Q

 ( )
 S + [ ( ),
( ) M

 ( )
 S .
( ) M

2 b . Now let us take a look at an equation of the form


2

    +  ( )   +  ( ) = 0

(5)

on a closed interval    . Assume

c ( ) [ ( )]2 4  ( ) 0.
Then the leading terms of the asymptotic expansions of the fundamental system of solutions of
equation (5), as pD 0, are expressed by
1

1
1

X
c ( )  X
Q 2

M
2
1

1
X
= | c ( )|1 J 4 exp
c ( )  X
Q 2

M
2
c
= | c ( )|1 J 4 exp

  ( )
 S V 1 + [ ( )W ,
c ( ) M
( )
 S V 1 + [ ( )W .
( ) M

0.2.3-6. Equations of the general form.


The more general equation

    +  ( , )   +  ( , ) = 0

is reducible, with the aid of the substitution

= exp O

1
X  , to an equation of the form (4),
2
M P

2    + (  14 2 12    ) = 0,

to which the asymptotic formulas given above in Paragraph 0.2.3-4 are applicable.

^_

References for Subsection 0.2.3: W. Wasov (1965), F. W. J. Olver (1974), A. H. Nayfeh (1973, 1981), M. V. Fedoryuk
(1993).

0.2.4. Boundary Value Problems


0.2.4-1. The first, second, third, and mixed boundary value problems (

  2 ).

We consider the second-order nonhomogeneous linear differential equation

    +  ( )   + ( ) = % ( ).

(1)

1 b . The first boundary value problem: Find a solution of equation (1) satisfying the boundary
conditions
=  1 at  =  1 ,
=  2 at  =  2 .
(2)
(The values of the unknown are prescribed at two distinct points 

and  2 ).

2003 by Chapman & Hall/CRC

Page 27

2 b . The second boundary value problem: Find a solution of equation (1) satisfying the boundary
conditions
  =  1 at  =  1 ,
  =  2 at  =  2 .
(3)
(The values of the derivative of the unknown are prescribed at two distinct points 

and  2 ).

3 b . The third boundary value problem: Find a solution of equation (1) satisfying the boundary
conditions
  , 1 =  1 at  =  1 ,
(4)
  + , 2 =  2 at  =  2 .
4 b . The third boundary value problem: Find a solution of equation (1) satisfying the boundary
conditions
  =  2 at  =  2 .
=  1 at  =  1 ,
(5)
(The unknown itself is prescribed at one point, and its derivative at another point.)
Conditions (1), (2), (3), and (4) are called homogeneous if  1 =  2 = 0.
0.2.4-2. Simplification of boundary conditions. Reduction of equation to the self-adjoint form.
1 b . Nonhomogeneous boundary conditions can be reduced to homogeneous ones by the change of
variable H = # 2  2 + # 1  + # 0 + (the constants # 2 , # 1 , and # 0 are selected using the method
of undetermined coefficients). In particular, the nonhomogeneous boundary conditions of the first
kind (1) can be reduced to homogeneous boundary conditions by the linear change of variable
 
H = 2 1 (  1 )  1 .
 2 1
2 b . On multiplying by  ( ) = exp X

 ( )  S , one reduces equation (1) to the self-adjoint form:


M

(6)
[ ( )  ]  +  ( ) = & ( ).

Without loss of generality, we further consider equation (6) instead of (1). We assume that the
functions  ,   ,  , and & are continuous on the interval  1   2 , and  is positive.
0.2.4-3. The Greens function. Boundary value problems for nonhomogeneous equations.
The Greens function of the first boundary value problem for equation (6) with homogeneous
boundary conditions (2) is a function of two variables 4 ( , ) that satisfies the following conditions:
1 b . 4 ( , ) is continuous in  for fixed , with 

 

and 

 2.

2 b . 4 ( , ) is a solution of the homogeneous equation (6), with & = 0, for all 


of the point  = .

< <

exclusive

3 b . 4 ( , ) satisfies the homogeneous boundary conditions 4 ( 1 , ) = 4 ( 2 , ) = 0.

4 b . The derivative 4  ( , ) has a jump of 1


 ( ) at the point
 = , that is,
1
.
4  ( , ) 
,  > 4  ( , ) 
,  < =


( )

For the second, third, and mixed boundary value problems, the Greens function is defined
likewise except that in 3 b the homogeneous boundary conditions (3), (4), and (5), with  1 =  2 = 0,
are adopted, respectively.
The solution of the nonhomogeneous equation (6) subject to appropriate homogeneous boundary
conditions is expressed in terms of the Greens function as follows:*
( ) = X 
* The homogeneous boundary value problemwith
solution.

2
1

4 ( , )& ( ) .
M
) = 0 and

= 0is assumed to have only the trivial

2003 by Chapman & Hall/CRC

Page 28

0.2.4-4. Representation of the Greens function in terms of particular solutions.


We consider the first boundary value problem. Let 1 = 1 ( ) and 2 = 2 ( ) be linearly independent
particular solutions of the homogeneous equation (6), with & = 0, that satisfy the conditions

1( 1)

= 0,

2( 2 )

= 0.

(Each of the solutions satisfies one of the homogeneous boundary conditions.)


The Greens function is expressed in terms of solutions of the homogeneous equation as follows:



4 (

( ) =
x3y'z|{}~/

where

1(

( ) 2 ( )
( )
2 ( )
( )

  1( )
, ) = 
1( )
 ( )

for 

 ,

for   2 ,

(7)

 ) 2 ( ) 1 ( ) 2 ( ) is the Wronskian determinant.

Formula (7) can also be used to construct the Greens functions for the second, third,
and mixed boundary value problems. To this end, one should find two linearly independent solutions,
1 = 1 (  ) and 2 = 2 (  ), of the homogeneous equation; the former satisfies the corresponding
homogeneous boundary condition at  =  1 and the latter satisfies the one at  =  2 .

^_

References for Subsection 0.2.4: L. E. Elsgolts (1961), E. Kamke (1977), A. N. Tikhonov, A. B. Vasileva, and
A. G. Sveshnikov (1980).

0.2.5. Eigenvalue Problems


0.2.5-1. The SturmLiouville problem.
Consider the second-order homogeneous linear differential equation

 
[  ( )  ]  + [ ( ) % ( )] = 0

(1)

subject to linear boundary conditions of the general form

 + 
 + 

1
2

= 0 at
= 0 at

 =  1,
 =  2.

(2)

It is assumed that the functions  ,    , , and % are continuous, and  and are positive on an
interval  1   2 . It is also assumed that |  1 | + |  1 | > 0 and |  2 | + |  2 | > 0.
The SturmLiouville problem: Find the values  of the parameter at which problem (1), (2)
has a nontrivial solution. Such  are called eigenvalues and the corresponding solutions  =  ( )
are called eigenfunctions of the SturmLiouville problem (1), (2).
0.2.5-2. General properties of the SturmLiouville problem (1), (2).
1 b . There are infinitely (countably) many eigenvalues. All eigenvalues can be ordered so that
1 < 2 < 3 < ))) . Moreover,  D F as BD F ; hence, there can only be a finite number of
negative eigenvalues. Each eigenvalue has multiplicity 1.
2 b . The eigenfunctions are defined up to a constant factor. Each eigenfunction  ( ) has precisely
B 1 zeros on the open interval ( 1 ,  2 ).
3 b . Any two eigenfunctions  ( ) and + ( ), B , are orthogonal with weight ( ) on the

interval  1   2 :
2
X  ( )  ( ) + ( )  = 0 if B .
1

2003 by Chapman & Hall/CRC

Page 29

4 b . An arbitrary function  ( ) that has a continuous derivative and satisfies the boundary conditions
of the SturmLiouville problem can be decomposed into an absolutely and uniformly convergent
series in the eigenfunctions:

 ( ) =    ( ),
 =1
where the Fourier coefficients   of  ( ) are calculated by
2
2 =  2 ( ) 2 ( )  .
1
  = 2 X  ( )  ( )  ( )  ,
X 


M
M
1
1

5 b . If the conditions

% ( ) 0,

 1  1 0,

 2 2 0

(3)

hold true, there are no negative eigenvalues. If % 0 and  1 =  2 = 0, the least eigenvalue is 1 = 0,
to which there corresponds an eigenfunction 1 = const. In the other cases where conditions (3) are
satisfied, all eigenvalues are positive.
6 b . The following asymptotic formula is valid for eigenvalues as BDGF :

 = 

2 2

+ [ (1),

=X 

( )
 .
 ( ) M

(4)

Paragraphs 0.2.5-3 through 0.2.5-6 will describe special properties of the SturmLiouville
problem which depend on the specific form of the boundary conditions.
x3y'z|{}~
Equation (1) can be reduced to the case where  ( ) 1 and ( ) 1 by the change


of variables

( )
1J 4
 , ( ) = V  (  ) (  )W
=X
( ).
 ( ) M
In this case, the boundary conditions are transformed to boundary conditions of similar form.
x3y'z|{}~
The second-order linear equation

Y 2 ( )    + Y 1 ( )   + [ + Y 0 ( )] = 0
can be represented in the form of equation (1) where  ( ), ( ), and % ( ) are given by:

 ( ) = exp X
Q

Y 1(  )
1
 S , ( ) =
exp X
Y 2(  ) M
Y 2(  )
Q

Y 1(  )
Y ( )
 S , % ( ) = 0 exp X
Y 2(  ) M
Y 2( )
Q

0.2.5-3. Problems with boundary conditions of the first kind (case 

=

Y 1(  )
 S .
Y 2(  ) M

= 0 and 

=

= 1).

Let us note some special properties of the SturmLiouville problem that is the first boundary value
problem for equation (1) with the boundary conditions

 =  1,

= 0 at

= 0 at

 =  2.

(5)

1 b . For BDF , the asymptotic relation (4) can be used to estimate the eigenvalues  . In this
case, the asymptotic formula

 ( ) =


4
2

 (  ) (  )

1 4

sin

X 


1

( )
1
 +[ O
,
 ( ) M
B P

=X 

2
1

( )

 ( ) M

holds true for the eigenfunctions  ( ).

2003 by Chapman & Hall/CRC

Page 30

TABLE 2
Example estimates of the first eigenvalue 1 in SturmLiouville problems
with boundary conditions of the first kind (0) = (1) = 0 obtained
using the RayleighRitz principle [the right-hand side of relation (6)]
Equation

Test function

1, approximate

1, exact

  + (1+  2)2 = 0

H = sin 

15.337

15.0

  + (4  2)2 = 0

H = sin 

135.317

134.837

[(1+  )1  ]  +

=0

H = sin 

7.003

6.772

  )  +

=0

H = sin 

11.9956

11.8985

( 1+ 

H = sin 
 )
H =  (1

  + (1+sin  ) = 0

0.54105

0.55204

0.54032

0.54032

2 b . If % 0, the following upper estimate holds for the least eigenvalue (RayleighRitz principle):

X 

V  ( )( H   )2 + % ( ) H 2 W 
M ,
2
2
X  ( ) H 
M
1

2
1

(6)

where H = H ( ) is any twice differentiable function that satisfies the conditions H ( 1 ) = H ( 2 ) = 0.


The equality in (6) is attained if H = 1 ( ), where 1 ( ) is the eigenfunction corresponding to the
(  1 )
S in (6) to obtain specific
eigenvalue 1 . One can take H = (  1 )( 2  ) or H = sin 
Q  2 1
estimates.
It is significant to note that the left-hand side of (6) usually gives a fairly precise estimate of the
first eigenvalue (see Table 2).
3 b . The extension of the interval [ 1 ,  2 ] leads to decreasing in eigenvalues.
4 b . Let the inequalities
0<

min

 ( ) 

max ,

0<

min

( )

0<%

max ,

% ( ) %

min

max

be satisfied. Then the following bilateral estimates hold:

min
max

(

2 2

)2

min
max

max
min

(

2 2

 1 )2

max
min

5 b . In engineering calculations for eigenvalues, the approximate formula

 = 

2 2

1
+
 2

X 

2
1

% ( )
 ,
( ) M

=X 

2
1

( )

 ( ) M

(7)

may be quite useful. This formula provides an exact result if  ( ) ( ) = const and % ( )
( ) = const
(in particular, for constant equation coefficients,  =  0 , % = % 0 , and = 0 ) and gives a correct
asymptotic behavior of (4) for any  ( ), % ( ), and ( ). In addition, relation (7) gives two correct
leading asymptotic terms as BCD F if  ( ) = const and ( ) = const [and also if  ( ) ( ) = const].

2003 by Chapman & Hall/CRC

Page 31

6 b . Suppose  ( ) = ( ) = 1 and the function % = % ( ) has a continuous derivative. The following


asymptotic relations hold for eigenvalues  and eigenfunctions  ( ) as BDGF :

=



 B 

5 ( 1 ,  2 ) + [O

B (   1 ) 1

(
 ( ) = sin   
B Q
w
2
1
2

 ) 5 ( ,  2 ) + (

where

5 ( , ) =

 ) 5 ( 1 ,  )S cos

B (  1 )
1
+ [O 2 ,
 2 1
B P

1
X % ( )  .
2 
M

(8)

7 b . Let us consider the eigenvalue problem for the equation with a small parameter

    + [ + % ( )] = 0

( pD

0)

subject to the boundary conditions (5) with  1 = 0 and  2 = 1. We assume that % ( ) = % ( ).


This problem has the following eigenvalues and eigenfunctions:

 = 2B


#  +

 ( ) = 2 sin( B/ )

  B
2
2

# 2 

  B

# 

+ [ ( 3 ),

#  = 2 X

1
0

% ( ) sin( B/ ) sin( ,! )  ;



 M

sin( ,! ) + [ ( 2 ).

Here, the summation is carried out over , from 1 to F . The next term in the expansion of
be found in Nayfeh (1973).
0.2.5-4. Problems with boundary conditions of the second kind (case 

1 = 2 =1

 can

and  1 =  2 = 0).

Let us note some special properties of the SturmLiouville problem that is the second boundary
value problem for equation (1) with the boundary conditions

  = 0 at  =  1 ,

  = 0 at  =  2 .

1 b . If % > 0, the upper estimate (6) is valid for the least eigenvalue, with H = H ( ) being any
twice-differentiable function that satisfies the conditions H   ( 1 ) = H   ( 2 ) = 0. The equality in (6) is
attained if H = 1 ( ), where 1 ( ) is the eigenfunction corresponding to the eigenvalue 1 .
2 b . Suppose  ( ) = ( ) = 1 and the function % = % ( ) has a continuous derivative. The following
asymptotic relations hold for eigenvalues  and eigenfunctions  ( ) as BDGF :
1
(B 1)
1
+
5 ( 1 ,  2 ) + [ O 2 ,
B P
2 1
 (B 1)
1
(B 1)(  1 )
+
( 1  ) 5 ( ,  2 )
 ( ) = cos 
 2 1
 ((B B 1)1)(Q   1 )
1
+ [O 2 ,
+ ( 2  ) 5 ( 1 ,  )S sin 
 2 1
B P

= 



where 5 ( , ) is given by (8).

2003 by Chapman & Hall/CRC

Page 32

0.2.5-5. Problems with boundary conditions of the third kind (case 

=

= 1 and  1 

0).

We consider the third boundary value problem for equation (1) subject to condition (2) with
 1 =  2 = 1. We assume that  ( ) = ( ) = 1 and the function % = % ( ) has a continuous derivative.
The following asymptotic formulas hold for eigenvalues  eigenfunctions  ( ) as BDGF :

1
(B 1)
V 5 ( 1 ,  2 )  1 +  2 W + [O 1 ,
+
(B 1)
B 2P
2 1

1
(B 1)(  1 )
+
( 1  ) V 5 ( ,  2 ) +  2 W
 ( ) = cos 
 2 1
(B 1)

(B 1)(  1 )
1
+ ( 2  ) V 5 ( 1 ,  )  1 W! sin 
+ [O 2 ,
 2 1
B P

 = 


where 5 ( , ) is defined by (8).


0.2.5-6. Problems with mixed boundary conditions (case 

=

= 1 and 

=

= 0).

Let us note some special properties of the SturmLiouville problem that is the a mixed boundary
value problem for equation (1) with the boundary conditions

  = 0 at  =  1 ,

= 0 at

 =  2.

1 b . If % 0, the upper estimate (6) is valid for the least eigenvalue, with H = H ( ) being any
twice-differentiable function that satisfies the conditions H   ( 1 ) = 0 and H ( 2 ) = 0. The equality
in (6) is attained if H = 1 ( ), where 1 ( ) is the eigenfunction corresponding to the eigenvalue 1 .
2 b . Suppose  ( ) = ( ) = 1 and the function % = % ( ) has a continuous derivative. The following
asymptotic relations hold for eigenvalues  and eigenfunctions  ( ) as BDGF :

2
(2B 1)
1
+
5 ( 1 ,  2 ) + [O 2 ,


)
(2
B

1)
B
P
2
1

2
(2B 1)(  1 )
+
( 1  ) 5 ( ,  2 )
 ( ) = cos  2(  )
(2B 1) Q
2
1
 (2B 1)(  1 )
1
+ [O 2 ,
+ ( 2  ) 5 ( 1 ,  )S sin 
2( 2  1 )
B P

 = 
2(

where 5 ( , ) is defined by (8).

^_

References for Subsection 0.2.5: L. Collatz (1963), E. Kamke (1977), A. G. Kostyuchenko and I. S. Sargsyan (1979),
V. A. Marchenko (1986), B. M. Levitan and I. S. Sargsyan (1988), V. A. Vinokurov and V. A. Sadovnichii (2000),
A. D. Polyanin (2002).

0.3. Second-Order Nonlinear Differential Equations


0.3.1. Form of the General Solution. Cauchy Problem
0.3.1-1. Equations solved for the derivative. General solution.
A second-order ordinary differential equation solved for the highest derivative has the form

    =  ( , ,   ).

(1)

The general solution of this equation depends on two arbitrary constants, L 1 and L 2 . In some cases,
the general solution can be written in explicit form, = Y ( , L 1 , L 2 ), but more often implicit or
parametric forms of the general solution are encountered.

2003 by Chapman & Hall/CRC

Page 33

0.3.1-2. Cauchy problem. The existence and uniqueness theorem.


1 b . Cauchy problem: Find a solution of equation (1) satisfying the initial conditions
( 0 ) =

0,

  ( 0 ) =

(At a point  =  0 , the value of the unknown function,

0,

1.

and its derivative,

(2)
1,

are prescribed.)

2 b . Existence and uniqueness theorem. Let  ( , , H ) be a continuous function in all its arguments

in a neighborhood of a point ( 0 , 0 , 1 ) and let  have bounded partial derivatives  and  in



this neighborhood, or the Lipschitz condition is satisfied: |  ( , , H )  ( , , H )| #  | |+| H H |  ,
where # is some positive number. Then a solution of equation (1) satisfying the initial conditions (2)
exists and is unique.

^_

References for Subsection 0.3.1: G. A. Korn and T. M. Korn (1968), I. G. Petrovskii (1970), E. Kamke (1977),
A. N. Tikhonov, A. B. Vasileva, and A. G. Sveshnikov (1980).

0.3.2. Equations Admitting Reduction of Order


0.3.2-1. Equations not containing

explicitly.

In the general case, an equation that does not contain

implicitly has the form

 ( ,   ,    ) = 0.

(1)

Such equations remain unchanged under an arbitrary translation of the dependent variable:
+ const . The substitution  = H ( ),   = H   ( ), brings (1) to a first-order equation:
D
 ( , H , H   ) = 0.
0.3.2-2. Equations not containing  explicitly (autonomous equations).
In the general case, an equation that does not contain  implicitly has the form

 ( ,   ,    ) = 0.

(2)

Such equations remain unchanged under an arbitrary translation of the independent variable:
vD  + const. Using the substitution  = ( ), where plays the role of the independent
variable, and taking into account the relations "  = 2 = 2  = 2 , one can reduce (2) to a


first-order equation:  ( , , *u ) = 0.
 
x3y'z|{}~
The equation  =  ( +  2 + ' + ( ) is reduced by the change of variable
2
= +  + ' + ( to an autonomous equation,   =  ( ) + 2  .
0.3.2-3. Equations of the form  (  +  ,  ,   ) = 0.
Such equations are invariant under simultaneous translations of the independent and dependent
!( , where ( is an arbitrary constant.
variables in accordance with the rule ED  + ( , D
For  = 0, see equation (1). For  0, the substitution  =  +  leads to equation (2):
 (  , 2 
 , 2   ) = 0.
0.3.2-4. Equations of the form  ( ,   ,   ) = 0.
The substitution ( ) = 



leads to a first-order equation:  ( , , u


 ) = 0.

2003 by Chapman & Hall/CRC

Page 34

0.3.2-5. Homogeneous equations.


1 b . The equations homogeneous in the independent variable remain unchanged under scaling of the
independent variable, ED  , where is an arbitrary nonzero number. In the general case, such
equations can be written in the form

 ( ,   , 

The substitution H ( ) = 

    ) = 0.

(3)

  leads to a first-order equation:  ( , H , HH  H ) = 0.




2 b . The equations homogeneous in the dependent variable remain unchanged under scaling of the
D , where is an arbitrary nonzero number. In the general case, such
variable sought,
equations can be written in the form

 ( ,   
,    
) = 0.

(4)

The substitution H ( ) =  
 leads to a first-order equation:  ( , H , H   + H 2 ) = 0.
3 b . The equations homogeneous in both variables are invariant under simultaneous scaling (dilatation) of the independent and dependent variables,  DG and DG , where is an arbitrary

nonzero number. In the general case, such equations can be written in the form
The transformation a = ln | |,
 ( , 2g  + , 2g   g + 2g  ) = 0.

= 


 (
 ,   ,     ) = 0.

(5)

leads to an autonomous equation (see Paragraph 0.3.2-2):

0.3.2-6. Generalized homogeneous equations.


1 b . The generalized homogeneous equations remain unchanged under simultaneous scaling of the
independent and dependent variables in accordance with the rule D and D  , where
is an arbitrary nonzero number and , is some number. Such equations can be written in the form

 (  ,  1    ,  2     ) = 0.
(6)
The transformation a = ln  , =   leads to an autonomous equation (see Paragraph 0.3.2-2):
  , g  + , , g   g + (2 , 1) g  + , ( , 1)  = 0.
2 b . The most general form of representation of generalized homogeneous equations is as follows:

(  + ,   
 ,  2   
 ) = 0.
(7)
+



, = 
brings this equation to the first-order equation
The transformation H =  
H , , H ( + B )  + 2  = 0.

x3y'z|{}~
For
0, equation (7) is equivalent to equation (6) in which , = B
. To

= 0 there correspond equations (3) and (4) homogeneous in


the particular values B = 0 and

the independent and dependent variables, respectively. For B = 0, we have an equation

homogeneous in both variables (5).


0.3.2-7. Equations invariant under scalingtranslation transformations.
1 b . The equations of the form

 ( 

, 

 

 ,       ) = 0

(8)

remain unchanged under simultaneous translation and scaling of variables, D  + and D ,



brings (8) to an
where =  9  and is an arbitrary number. The substitution = 


2



autonomous equation (see Paragraph 0.3.2-2):  ( ,  ,  2  + 2 ) = 0.

2003 by Chapman & Hall/CRC

Page 35

2 b . The equation

   

,  ,    
) = 0

 ( 

(9)

is invariant under the simultaneous translation and scaling of variables, D  + and



where = exp( s
B ) and is an arbitrary number. The transformation H = !  ,

2
brings (9) to a first-order equation:   H , , H (B
+ ) +
= 0.
3 b . The equation

 (    ,    , 

DG ,
= 


    ) = 0

(10)
+ ,

is invariant under the simultaneous scaling and translation of variables, DG and D
where = exp(h
B ) and is an arbitrary number. The transformation H =    ,
brings (10) to a first-order equation:   H , , H ( + B )2  = 0.

=  

0.3.2-8. Exact second-order equations.


The second-order equation

 ( , ,   ,    ) = 0

(11)

is said to be exact if it is the total differential of some function,  = Y  , where Y = Y ( , ,   ). If


equation (11) is exact, then we have a first-order equation for Y :
Y ( , ,   ) = L ,
(12)
where L is an arbitrary constant.
If equation (11) is exact, then  ( , ,  ,   ) must have the form:
 ( , ,   ,    ) =  ( , ,   )    + ( , ,   ).
(13)
Here,  and are expressed in terms of Y by the formulas:
Y
Y
Y 
 .
 ( , ,   ) = N  , ( , ,   ) = N + N
(14)


N
N
N
By differentiating (14) with respect to  , , and  =   , we eliminate the variable Y from the
two formulas in (14). As a result, we have the following test relations for  and :
  + 2    +  2   =  K +   K  ,
(15)
  K +    K + 2   = K'K .
Here, the subscripts denote the corresponding partial derivatives.
If conditions (15) hold, then equation (11) with  of (13) is exact. In this case, we can integrate
the first equation in (14) with respect to  =  to determine Y = Y ( , ,   ):

Y = X  ( , ,  )

+ j ( , ),

(16)

where j ( , ) is an arbitrary function of integration. This function is determined by substituting


(16) into the second equation in (14).
Example. The left-hand side of the equation

+ (

)2 + 2

can be represented in the form (13), where = and = 2 + 2


satisfied. Hence, equation (17) is exact. Using (16), we obtain

=0

(17)

2.

It is easy to verify that conditions (15) are

+ ( , ).





(18)

Substituting this expression into the second equation in (14) and taking into account the relation = 2 + 2 + 2 ,
. Since
= ( , ), we have 2 =
and 2 = . Integrating yields
we find that 2 + 2 = +

2
= + const. Substituting this expression into (18) and taking into account relation (12), we find a first integral of
equation (17):
+ 2 = 1 , where = .
(19)
= 2 , we arrive at the first-order linear equation + 2
= 2 1 , which is easy to integrate. Thus, we find the
Setting
solution of the original equation in the form:







=2



exp

2 X

exp



exp

2003 by Chapman & Hall/CRC

Page 36

0.3.2-9. Reduction of quasilinear equations to the normal form.


We consider the equation

    +  ( )   + ( ) = ` ( , )
(20)
with linear left-hand side and nonlinear right-hand side. Let 1 ( ) and 2 ( ) form a fundamental
system of solutions of the truncated linear equation corresponding to ` 0. The transformation
=

 )
,
(
1  )
2(

1(

(21)

 )

brings equation (20) to the normal form:

  = i ( , ),

where

i ( , ) =

 )
` , 1 ( )  .
 )

3
1(
2(

( ) = 1 2 2 1 is the Wronskian of the truncated equation; and the variable  must be


Here,
expressed in terms of using the first relation in (21).
Transformation (21) is convenient for the simplification and classification of equations having
the form (20) with ` ( , ) = % ( )  , thus reducing the number of functions from three to one:
{  , , % }  {0, 0, % 1}.

^_

References for Subsection 0.3.2: G. M. Murphy (1960), E. Kamke (1977), V. F. Zaitsev and A. D. Polyanin (1993, 2001),
A. D. Polyanin and V. F. Zaitsev (1995), D. Zwillinger (1998).

0.3.3. Methods of Regular Series Expansions with Respect to the


Independent Variable or Small Parameter
0.3.3-1. Method of expansion in powers of the independent variable.
A solution of the Cauchy problem

    =  ( , ,   ),
  ( 0 ) =
( 0 ) = 0 ,

(1)
(2)

can be sought in the form of a Taylor series in powers of the difference (  0 ), specifically:

   ( )
    ( )
0
0
(   0 )2 +
(  0 )3 + ))) .
( ) = ( 0 ) +   ( 0 )(  0 ) +
2!
3!

(3)

    ( 0 ) =  ( 0 , 0 , 1 ).

(4)

The first two coefficients ( 0 ) and  ( 0 ) in solution (3) are defined by the initial conditions (2).
The values of the subsequent derivatives of at the point  =  0 are determined from equation (1)
and its derivative equations (obtained by successive differentiation of the equation) taking into
account the initial conditions (2). In particular, setting  =  0 in (1) and substituting (2), we obtain
the value of the second derivative:

Further, differentiating (1) yields

     =   ( , ,   ) +  ( , ,   )   +  ( , ,   )    .
(5)

 
"

On substituting  =  0 , the initial conditions (2), and the expression of  ( 0 ) of (4) into the

right-hand side of equation (5), we calculate the value of the third derivative:

     ( 0 ) =   ( 0 , 0 , 1 ) +  ( 0 , 0 , 1 ) 1 +  ( 0 , 0 , 1 )  ( 0 , 0 , 1 ).

 
The subsequent derivatives of the unknown are determined likewise.
The thus obtained solution (3) can only be used in a small neighborhood of the point  = 

0.

2003 by Chapman & Hall/CRC

Page 37

0.3.3-2. Method of regular (direct) expansion in powers of the small parameter.


We consider an equation of general form with a parameter :

    +  ( , ,   , ) = 0.

(6)

We assume that the function  can be represented as a series in powers of :

 ( , ,   , ) =    ( , ,   ).
 =0

ID

(7)

Solutions of the Cauchy problem and various boundary value problems for equation (6) with
0 are sought in the form of a power series expansion:

  ( ).

=0

(8)

One substitutes expression (8) into equation (6) taking into account (7). Then the functions   are
expanded into power series with respect to the small parameter, and the coefficient of like powers
of are collected. Equating the resulting expressions (the coefficient of like powers of ) to zero,
one arrives at a system of equations for  :

  +  ( , ,  ) = 0,
0
0 0
  +  ( , ,  )  + 4 ( , ,  ) +  ( , ,  ) = 0,
0 0 1
0 0 1
1
0 0
1
0

 = N
N

 ,

4 = N

(9)
0

(10)

Here, only the first two equations are written out. The prime denotes differentiation with respect
to  . To obtain the initial (or boundary) conditions for  , the expansion (8) is taken into account.
The success in the application of this method is primarily determined by the possibility of
constructing a solution of equation (9), for the leading term 0 . It is significant to note that the other
terms  with B 1 are governed by linear equations with homogeneous initial conditions.
Example. The Duffing equation with initial conditions

+ + 

= 0,

(0) = ,

(0) = 0,

describes the motion of a cubic oscillator, i.e., oscillations of a point mass on a nonlinear spring. Here, is the deviation of
the point mass from the equilibrium and is dimensionless time.
0, an approximate solution of the problem is sought in the form of the asymptotic expansion (8). We
For
substitute (8) into the equation and initial conditions and expand in powers of . On equating the coefficients of like powers
of the small parameter to zero, we obtain the following problems for 0 and 1 :

!

The solution of the problem for

0 +
1 +

= 0,

= 03 ,

is given by:

= ,

= 0,

0 = 0;
1 = 0.

cos .

Substituting this expression into the equation for 1 and taking into account the identity cos3 =
obtain
1 + 1 = 14 3 (cos 3 + 3 cos ),
1 = 0, 1 = 0.
Integrating yields
1 = 38 3 sin + 321 3 (cos 3 3 cos ).
Thus the two-term solution of the original problem is given by:

2$#

cos + '

38 sin +

1
(cos 3
32

3 cos ) +

1
4

cos 3 +

3
4

cos , we

" ( 2 ).

The term sin causes 1 0


as
. For this reason, the solution obtained is unsuitable at
 1 .
1; this results from the condition of applicability of the expansion, 0
large times. It can only be used for
This circumstance is typical of the method of regular series expansions with respect to the small parameter; in other
words, the expansion becomes unsuitable at large values of the independent variable. This method is also inapplicable if
the expansion (8) begins with negative powers of . Methods that allow avoiding the above difficulties are discussed in
Subsection 0.3.4.
2  Growing terms as
, like sin , that narrow down the domain of applicability of asymptotic
expansions are called secular.

,+

^_

 )(

% '&

'&

* 

-&

References for Paragraph 0.3.3-2: A. H. Nayfeh (1973, 1981).

2003 by Chapman & Hall/CRC

Page 38

0.3.3-3. Pade approximants.


Suppose the , + 1 leading coefficients in the Taylor series expansion of a differential equation about
the point  = 0 are obtained by the method presented in Paragraph 0.3.3-1, so that

+1 (

 ) =  0 +  1  + ))) +     .

(11)

The partial sum (11) pretty well approximates the solution at small  but is poor for intermediate
and large values of  , since the series can be slowly convergent or even divergent. This is also
related to the fact that  DGF as EDGF , while the exact solution can well be bounded.
cases, instead of the expansion (11), it is reasonable to consider a Pad e approximant
.0/ (In ),many
which is the ratio of two polynomials of degree and 8 , specifically,

. /
6

( ) =

+ # 1  + ))) + #
1 + $ 1  + )) ) + $

/ /

+8

where

=, .

(12)

are selected so that the , + 1 leading terms in the


The coefficients # 1 ,  , # / and $ 1 ,  , $
Taylor series expansion of (12) coincide with 6 the respective terms of the expansion (11). In other
words, the expansions (11) and (12) must be asymptotically equivalent as D 0.
In practice, one usually takes = 8 (the diagonal sequence). It often turns out that formula (12)
pretty well approximates the exact solution on the entire range of  (for sufficiently large ).
Example 1. Consider the following Cauchy problem for a second-order nonlinear equation:



+ 3;

The Taylor series expansion of the solution about


|

=1+

(0) =

(0) = 1.

(13)

= 0 has the form (see Paragraph 0.3.3-1):


+

12131 .

(14)

This geometric series is convergent only for | < 1.


The diagonal sequence of Pade approximants corresponding to series (14) is:

1
1(

)=

1
,
1

2
2(

)=

1
,
1

3
3(

1
.
1

)=

(15)

1
is the exact solution of the Cauchy problem (13). Hence, in
1
this case, the diagonal sequence of Pade approximants recovers the exact solution from only a few terms in the Taylor series.
It is not difficult to verify that the function

)=

Example 2. Consider the Cauchy problem for a second-order nonlinear equation:

= 2 ;

(0) = 0,

(0) = 1.

(16)

Following the method presented in Paragraph 0.3.3-1, we obtain the Taylor series expansion of the solution to problem (16) in the form:
17 7
( ) = + 13 3 + 152 5 + 315
+
.
(17)

13121

) = tan

. Hence it has singularities at


The exact solution of problem (16) is given by:
any finite segment of the Taylor series (17) does not have any singularities.
With series (17), we construct the diagonal sequence of Pade approximants:

2
2(

)=

3
3

,
2

3
3(

)=

3(2

15)
,
2 5)

4
4(

)=

1
(2
2

+ 1) . However,

5 (21 2 2 )
.
4 45 2 + 105

(18)

These Pade approximants have singularities (at the points where the denominators vanish):

)7
)7
)7

1.732
1.581
1.571 and

7

for

for

4
for
6.522

2
2(
3
3(
4
4(

),

),
).

2 and
It is apparent that the Pade approximants are attempting to recover the singularities of the exact solution at =
= 3 2. The Pade approximant 4 ( ) gives an accurate numerical approximation of the exact solution on the interval
4
2).
| | 2 (the error is everywhere less than 1%, except for a very small neighborhood of the point =

^_

6%

6 %

6 %

References for Paragraph 0.3.3-3: G. A. Baker (Jr.) and P. GravesMorris (1981), D. Zwillinger (1989, pp. 450453).

2003 by Chapman & Hall/CRC

Page 39

0.3.4. Perturbation Methods of Mechanics and Physics


0.3.4-1. Preliminary remarks. A summary table of basic methods.
Perturbation methods are widely used in nonlinear mechanics and theoretical physics for solving
problems that are described by differential equations with a small parameter . The primary purpose
of these methods is to obtain an approximate solution that would be equally suitable at all (small,
intermediate, and large) values of the independent variable as ID 0.
Equations with a small parameter can be classified according to the following:
(i) the order of the equation remains the same at = 0;
(ii) the order of the equation reduces at = 0.
For the first type of equations, solutions of related problems* are sufficiently smooth (little varying
as decreases). The second type of equation is said to be degenerate at = 0, or singularly
perturbed. In related problems, thin boundary layers usually arise whose thickness is significantly
dependent on ; such boundary layers are characterized by high gradients of the unknown.
All perturbation methods have a limited domain of applicability; the possibility of using one or
another method depends on the type of equations or problems involved. The most commonly used
methods are summarized in Table 3 (the method of regular series expansions is set out in Paragraph
0.3.3-2). In subsequent paragraphs, additional remarks and specific examples are given for some
of the methods. In practice, one usually confines oneself to few leading terms of the asymptotic
expansion.
In many problems of nonlinear mechanics and theoretical physics, the independent variable is
dimensionless time a . Therefore, in this subsection we use the conventional a , 0 a < F instead
of  .
0.3.4-2. The method of scaled parameters (LindstedtPoincar e method).
We illustrate the characteristic features of the method of scaled parameters with a specific example
(the transformation of the independent variable we use here as well as the form of the expansion are
specified in the first row of Table 3).
Example 1. Consider the Duffing equation:

89 8 +

 3 = 0.
(1)
On performing the change of variable : = ; (1 + 2< 1 + 12131 ), we have
=2 = + (1 + 2< 1 + 12131 )2 ( +  3 ) = 0.
(2)
The solution is sought in the series form = 0 ( ; ) +  1 ( ; ) + 13131 . Substituting it into equation (2) and matching the
coefficients of like powers of  , we arrive at the following system of equation for two leading terms of the series:
0 +
1 +

= 0,

where the prime denotes differentiation with respect to .


The general solution of equation (3) is given by:

where

<

and

+ 

3
0

<

(3)

1 0,

(4)

cos( + ),

(5)

are constants of integration. Taking into account (5) and rearranging terms, we reduce equation (4) to

1 +

= 14

cos 3( + ) 2

3
8

<

cos( + ).

(6)

the particular solution of equation (6) contains a secular term proportional to cos( + ). In this case,
For 1
the condition of applicability of the expansion 1 0 = (1) (see the first row and the last column of Table 3) cannot be
satisfied at sufficiently large . For this condition to be met, one should set
38

2,

"

<

= 38 2 .

(7)

* Further on, we assume that the initial and/or boundary conditions are independent of the parameter .

2003 by Chapman & Hall/CRC

Page 40

TABLE 3
Perturbation methods of nonlinear mechanics and theoretical physics
(the third column gives B leading asymptotic terms with respect to the small parameter )
Method
name

Examples of problems
solved by the method

Method
of scaled
parameters
(0 < )

One looks for periodic


solutions of the equation
+ 02 = ( , );
see also Paragraph 0.3.4-2

Method
of strained
coordinates
(0 < )

Cauchy problem:
= ( , , ); ( 0) = 0
( is of a special form);
see also the problem in the
method of scaled parameters

> ?
> ?

Averaging
method
(0 < )

> ?

@ 98 8 A @ BDC @ @ 8

@8 C > @ B @ >
C

Form of the solution


sought

@ (> ) = E F 1 B @ (G ),

=0F 1
> = GIH 1+ E B A 2J
=1
@ (> ) = E F 1 B @ (G ),

F =0
1

Additional
conditions and remarks

Unknowns:

and ;
= (1);

secular terms are


eliminated through
selection of the constants

+1

K@

@
M
@ = L and
K
M = L (1),
M
(1)
+1 K

Unknowns:

+1

> = G + E B NM (G )
=1
@ = P (> ) cos M (> ),
problem:
Unknowns: P and M ;
the amplitude P and phase M
@ Cauchy
2X
89 8 + A 02@ = BDC (@ , @ 8 ),
1
C
s = 2 X!Y 0 sin M[Z]\^M ,
are governed
@ (0) = @ 0, @ 8 (0) = @ 1;
Q Q 8 by the equations
= RS CUT (P ),
C c = 21X Y 02X cos M[Z]\^M ,
for more general problems,
N
Q
V
Z = C (P cos M ,P
A 0 sin M )
see Paragraph 0.3.4-3, Item 2 O
Q 8 = A 0 RS CUW (P )

@ = P cos M + E F 1 B @ (P , M ),

One looks for periodic


=1
Unknowns: @ , _ , ` ;
P
and M are determined
@ are 2a -periodic
solutions of the equation
functions of M ;
by the equations
@ 89 8 + A 02@ = BDC (@ , @ 8 );
Q Q 8 F 2_

= E B
Cauchy problem for this
the @ are assumed
(P ),
=1
and other equations
not
to contain cos M

QNQ V 8
F
= A + E B ` (P )
0

Krylov
Bogolyubov
Mitropolskii
method
(0 < )

> ?

Cauchy problem:
Method
+ 02 = ( , ),
of two-scale
(0) = 0, (0) = 1;
expansions
for boundary value problems,
(0 < )
see Paragraph 0.3.4-4, Item 2

> ?

Multiple
scales
method
(0 < )

> ?

Method of
matched
asymptotic
expansions
(0 )

p q

Method of
composite
expansions
(0 )

p q

@ 98 8 A @ DB C @ @ 8
@
@ @8 @

=1

@ = E F 1 B @ (b , c

=0

), where

1
b = UB > , c = > H 1+ E F B A J ,
O Q Q 8 = Bd + f 1+ B 2A =22 + gNgNg ) d
de
dih
One looks for periodic
F= E 1 B @ , where
@
solutions of the equation

@ 89 8 + A 02@ = BDC (@ , @ 8 );
@ = @ (j 0, =0j 1, kNkNk , j ), j = B >
Q
Cauchy problem for this
Q 8 = d + Bmd + gUgNE g + B E d
and other equations
dil dil
diln
Outer
expansion:
Boundary value problem:
B @ @ + C (p @ , @ )@@ = r (@
p s , @ ), @ =E F 1 (B )@ (p ), L (B ) p q .
(0) = 0, (q ) =
=0 t
Inner expansion ( G = p K B ):
( C assumed positive);
for other problems, see
u@ =E F 1u (B )u@ (G ), 0 p L (B )
Paragraph 0.3.4-5, Item 2 O

=0 t
@ = y (p , B )+ y u (G , B ),
Boundary value problem:
F 1 (B )y (p ),
B @ @ + C (p @ , @ )@@ = r (
@ p s , @ ),
y
=E

(0) = 0, (q ) =
u F 1 u =0 t u
( C assumed positive);
y = E (B )y (G ), G = p B .
boundary value problems
=0 t u
for other equations
Here, y v 0 as G[vx?
0

Unknowns:

;
and
=
(1);

secular terms are


eliminated through
selection of

@
+1 K

Unknowns: ;
+1 = (1);
for = 1, this method
is equivalent to
the averaging method

K@

Unknowns:

u @@

@
+1 K u @
+1 K

@ , u@ , , u
(1),
t t
== LL (1);

the procedure of
matching expansions is used:
(
0) = (
)

u @ [G vx?
u
u
Unknowns: y , y , ,
y (q ,u B ) = @
s , t t
y (0, B )+ y (0, B ) = @ 0;
@ pwv

two forms of representation


of the equation
(in terms of and )
are used to obtain solutions

2003 by Chapman & Hall/CRC

Page 41

In this case, the solution of equation (6) is given by:

1
32

cos 3( + ) .

(8)

Subsequent terms of the expansion can be found likewise.


With (5), (7), and (8), we obtain a solution of the Duffing equation in the form:

<: + ) + 321 ' 3 cos 3(<: + ) + " ( 2),


< = 1 38 ' 2 + " ( 2) 1 = 1 + 38 ' 2 + " ( 2).

cos(

0.3.4-3. Averaging method (Van der PolKrylovBogolyubov scheme).


1 b . The averaging method involved two stages. First, the second-order nonlinear equation


g g +

2
0

=  ( , g  )

(9)

is reduced with the transformation


g = 0  sin Y ,

=  cos Y ,

 =  (a ),

where

Y = Y (a ),

to an equivalent system of two first-order differential equations:


 g =  (  cos Y , 0  sin Y ) sin Y ,
0

Y g =


 (  cos Y , 0  sin Y ) cos Y .
0

(10)

The right-hand sides of equations (10) are periodic in Y , with the amplitude  being a slow function
of time a . The amplitude and the oscillation character are changing little during the time the phase Y
changes by 2 .

At the second stage, the right-hand sides of equations (10) are being averaged with respect to Y .
This procedure results in an approximate system of equations:


 g =  s (  ),
0
where

 s ( ) =

1
X
2

2
0

sin Ye (  cos Y ,

 sin Y ) Y ,
M

Y g =

 c ( ) =


 (  ),
0 c
1
X
2

2
0

cos Ye (  cos Y ,

(11)

 sin Y ) Y .
M

System (11) is substantially simpler than the original system (10)the first equation in (11), for
the oscillation amplitude  , is a separable equation and, hence, can readily be integrated; then the
second equation in (11), which is linear in Y , can also be integrated.
Note that the KrylovBogolyubovMitropolskii method (see the fourth row in Table 3) generalizes the above approach and allows obtaining subsequent asymptotic terms as ID 0.
2 b . Below we outline the general scheme of the averaging method. We consider the second-order
nonlinear equation with a small parameter:




g g + (a , , g ) =  (a , , g ).

(12)

Equation (12) should first be transformed to the equivalent system of equations


g = ,

g + (a , , ) =  (a , , ).

(13)

Suppose the general solution of the truncated system (13), with = 0, is known:
0

= Y (a , L

1,

L 2 ),

= j (a , L

1,

L 2 ),

(14)

2003 by Chapman & Hall/CRC

Page 42

where L 1 and L 2 are constants of integration. Taking advantage of the method of variation of
constants, we pass from the variables , in (13) to Lagranges variables  1 ,  2 according to the
formulas
= Y (a ,  1 ,  2 ), = j (a ,  1 ,  2 ),
(15)
where Y and j are the same functions that define the general solution of the truncated system (14).
Transformation (15) allows the reduction of system (13) to the standard form:

 1 =  1 (a ,  1 ,  2 ),

 2 =  2 (a ,  1 ,  2 ).

(16)

Here, the prime denotes differentiation with respect to a and

Y
Y 2  (a , Y , j )
Y 1  (a , Y , j )
,  2=
; Y  = N
,

Y 2j 1 Y 1j 2
Y 2j 1 Y 1 j 2
N 
Y = Y (a ,  1 ,  2 ), j = j (a ,  1 ,  2 ).

j
j  = N ,

N 

It is significant to note that system (16) is equivalent to the original equation (12). The unknowns
and  2 are slow functions of time.
As a result of averaging, system (16) is replaced by a simpler, approximate autonomous system
of equations:
 1 =  1 ( 1 ,  2 ),  2 =  2 ( 1 ,  2 ),
(17)

where

{ 1 }|

 ( 1 ,  2 ) =

 ( 1 ,  2 ) = lim

x3y'z|{}~

  (a ,  1 ,  2 ) a ,
M

{ 1 }|

-periodic function of a ;

  (a ,  1 ,  2 ) a , if   is not periodic in a .
M

if   is a

The averaging method is applicable to equations (9) and (12) with nonsmooth
right-hand sides.

x3y'z|{}~

The averaging method has rigorous mathematical substantiation. There is also a


procedure that allows finding subsequent asymptotic terms. For this procedure, e.g., see the books
by Bogolyubov and Mitropolskii (1974), Zhuravlev and Klimov (1988), and Arnold, Kozlov, and
Neishtadt (1993).
0.3.4-4. Method of two-scale expansions (ColeKevorkian scheme).
1 b . We illustrate the characteristic features of the method of two-scale expansions with a specific
example. Thereafter we outline possible generalizations and modifications of the method.
Example 2. Consider the Van der Pol equation:

89 8

+ = (1 2 )  .

(18)

The solution is sought in the form (see the fifth row in Table 3):

~ ,  ) +  1(~ ,  ) +  2 2 (~ ,  ) + 13131 ,
(19)
~ = : ,  = 1 +  2 < 2 + 13131 : .
On substituting (19) into (18) and on matching the coefficients of like powers of  , we obtain the following system for two

leading terms:
2
(20)
 20 + 0 = 0,

0(

1
2

= 2

~ 

+ (1 02 )

(21)

2003 by Chapman & Hall/CRC

Page 43

The general solution of equation (20) is given by:

(~ ) cos  + (~ ) sin  .

(22)

and
on the slow variable is not being established at this stage.
The dependence of
We substitute (22) into the right-hand side of equation (21) and perform elementary manipulations to obtain

2
1
2
2
1
 2 + 1 = 2 se + 4 (4 )
+ 14 ( 3 3 2 ) cos 3  + 14 (

cos + 2

1
4

(4

) sin

2 ) sin 3 .
(23)
The solution of this equation must not contain unbounded terms as & ; otherwise the necessary condition 1 % 0 = " (1)
is not satisfied. Therefore the coefficients of cos  and sin  must be set equal to zero:
2 + 14 (4 2 2 ) = 0,
e
(24)
e 14 (4 2 2 ) = 0.
2
Equations (24) serve to determine = ( ~ ) and = ( ~ ). We multiply the first equation in (24) by and the second
by and add them together to obtain
' 18 (4 2 ) = 0,
where 2 = 2 + 2 .
(25)
e
3

The integration by separation of variables yields:

2
2 = 2 + (44 0 2 ) e
0

(26)

where 0 is the initial oscillation amplitude.


and
in terms of the amplitude and phase , we have = cos and = sin . Substituting
On expressing
we find that = 0 or = 0 = const.

Therefore
these expressions into either of the two equations in (24) and using (25),


the leading asymptotic term can be represented as:

where

~ =  :

and

= ( ) cos( +

0 ),

 = : , and the function (~ ) is determined by (26).

2 b . The method of two-scale expansions can also be used for solving boundary value problems
where the small parameter appears together with the highest derivative as a factor (such problems
for 0   are indicated in the seventh row of Table 3 and in Paragraph 0.3.4-5). In the case
where a boundary layer arises near the point  = 0 (and its thickness has an order of magnitude
of ), the solution is sought in the form:

, ) + 1 ( , ) + 2 2 ( , ) + ))) ,
=  , = 1 V 0 ( ) +  1 ( ) + 2 2 ( ) + ))) W ,
=

0(

where the functions  =  ( , ) and  =  ( ) are to be determined. The derivative with respect
to  is calculated in accordance with the rule:

M


= N

+  N

= N

 0 +  1 + 2 2 + )))  N .

Additional conditions are imposed on the asymptotic terms in the domain under consideration;
namely,  +1
  = [ (1) and  +1
 = [ (1) for , = 0, 1,  , and 0 ( ) D as D 0.
x3y'z|{}~/
The two-scale method is also used to solve problems that arise in mechanics and
physics and are described by partial differential equations.
0.3.4-5. Method of matched asymptotic expansions.
1 b . We illustrate the characteristic features of the method of matched asymptotic expansions with a
specific example (the form of the expansions is specified in the seventh row of Table 3). Thereafter
we outline possible generalizations and modifications of the method.

2003 by Chapman & Hall/CRC

Page 44

Example 3. Consider the linear boundary value problem

+ + ( ) = 0,
(0) = , (1) = ,



&

(27)
(28)

where 0 < (0) < .


At = 0 equation (27) degenerates; the solution of the resulting first-order equation

+ ( ) = 0

(29)

cannot meet the two boundary conditions (28) simultaneously. It can be shown that the condition at = 0 has to be omitted
in this case (a boundary layer arises near this point).
The leading asymptotic term of the outer expansion, = 0 ( ) + ( ), is determined by equation (29). The solution
of (29) that satisfies the second boundary condition in (28) is given by:

" 

0 (

) = exp

~ ~ .

( )

(30)

We seek the leading term of the inner expansion, in the boundary layer adjacent to the left boundary, in the following
form (see the seventh row and third column in Table 3):

%,

(31)

 1 , we obtain
 +  = 0,
(32)
0
0
where the prime denotes differentiation with respect to ; . The solution of equation (32) that satisfies the first boundary
condition in (28) is given by:
0 =  + s = .
(33)
The constant of integration  is determined from the condition of matching the leading terms of the outer and inner
expansions:

0 (
0) = 0 ( ;'& ).
(34)

where

= 0 (; ) + " ( ), ;

is the extended variable. Substituting (31) into (27) and extracting the coefficient of

Substituting (30) and (33) into condition (34) yields

where

1
0

(35)

Taking into account relations (30), (31), (33), and (35), we represent the approximate solution in the form:


1
exp X (~ ) ~

for 0
for

" ( )

" ( ),
1.

(36)

It is apparent that inside the thin boundary layer, whose thickness is proportional to , the solution rapidly changes by a
= .
finite value,
To determine the function on the entire interval [0, 1] using formula (36), one has to switch at some intermediate
point = 0 from one part of the solution to the other. Such switching is not convenient and, in practice, one often resorts
to a composite solution instead of using the double formula (36). In similar cases, a composite solution is defined as:

0 (

) + 0( )

For  (

= lim

0 (

=)

) = lim

0 (; ).

and hence the composite solution becomes:


1

(~ ) ~ .
R + exp X

In the problem under consideration, we have

0 (

1, this solution transforms to the outer solution


) and for 0
an approximate representation of the unknown over the entire domain.

 , to the inner solution, thus providing

2 b . We now consider an equation of the general form

   =  ( , ,   )
subject to boundary conditions (28).
For the leading term of the outer expansion

0(

(37)

 ) + ))) , we have the equation:

 ( , 0 , 0 ) = 0.
2003 by Chapman & Hall/CRC

Page 45

In the general case, when using the method of matched asymptotic expansions, the position of
the boundary layer and the form of the inner (extended) variable have to be determined in the course
of the solution of the problem.
First we assume that the boundary layer is located near the left boundary. In (37), we make a
change of variable H = 
> ( ) and rewrite the equation as
1
'  = > 2 
 O>H , ,  .

> P

(38)

The function > = > ( ) is selected so that the right-hand side of equation (38) has a nonzero limit

value as ID 0, provided that H , , and  are of the order of 1.


Example 4. For

( , ,

) =

+ , where 0

=2 =

< 1, the substitution

1+
; = + 2 .



In order that the right-hand side of this equation has a nonzero limit value as

!

%N ( ) brings equation (37) to

0, one has to set

1+ % 

= 1 or

%  = const, where const is any positive number. It follows that =  .


leading
asymptotic term of the inner expansion in the boundary layer, = 0 ( ; ) + 1131 , is determined by the equation

+ The

; 0 = 0, where the prime denotes differentiation with respect to ; .


0
1
1+

1+

If the position of the boundary layer is selected incorrectly, the outer amd inner expansions
cannot be matched. In this situation, one should consider the case where an arbitrary boundary layer
is located on the right (this case is reduced to the previous one with the change of variable  = 1 H ).
In example 4 above, the boundary layer is on the left if , > 0 and on the right if , < 0.
There is a procedure for matching subsequent asymptotic terms of the expansion (see the seventh
row and last column in Table 3). In its general form, this procedure can be represented as:
inner expansion of the outer expansion ( -expansion for D 0)

= outer expansion of the inner expansion ( -expansion for HCDGF ).

x3y'z|{}~

The method of matched asymptotic expansions can also be applied to construct


periodic solutions of singularly perturbed equations (e.g., in the problem of relaxation oscillations
of the Van der Pol oscillator).

x3y'z|{}~

Two boundary layers can arise in some problems (e.g., in cases where the right-hand
side of equation (37) does not explicitly depend on " ).

x3y'z|{}~E

The method of matched asymptotic expansions is also used for solving equations
(in semi-infinite domains) that do not degenerate at = 0. In such cases, there are no boundary
layers; the original variable is used in the inner domain, and an extended coordinate is introduced in
the outer domain.

x3y'z|{}~E

The method of matched asymptotic expansions is successfully applied for the


solution of various problems in mathematical physics that are described by partial differential
equations; in particular, it plays an important role in the theory of heat and mass transfer and in
hydrodynamics.

^_

References for Subsection 0.3.4: M. Van Dyke (1964), A. Blaquiere (1966), G. D. Cole (1968), G. E. O. Giacaglia (1972),
A. H. Nayfeh (1973, 1981), N. N. Bogolyubov and Yu. A. Mitropolskii (1974) J. Kevorkian and J. D. Cole (1981, 1996),
P. A. Lagerstrom (1988), V. Ph. Zhuravlev and D. M. Klimov (1988), V. I. Arnold, V. V. Kozlov, and A. I. Neishtadt (1993).

0.3.5. Galerkin Method and Its Modifications (Projection Methods)


0.3.5-1. General form of an approximate solution.
Consider a boundary value problem for the equation

[ ]  ( ) = 0

(1)

2003 by Chapman & Hall/CRC

Page 46

with linear homogeneous boundary conditions at the points  =  1 and  =  2 ( 1   2 ). Here,


is a linear or nonlinear differential operator of the second order (or a higher order operator);

= ( ) is the unknown function and  =  ( ) is a given function. It is assumed that [0] = 0.


Let us choose a sequence of linearly independent functions

Y = Y  ( )

(B = 1, 2,  , )

(2)

satisfying the same boundary conditions as = ( ). According to all methods that will be considered
below, an approximate solution of equation (1) is sought as a linear combination

=1

#  Y  ( ),

(3)

with the unknown coefficients #  to be found in the process of solving the problem.
The finite sum (3) is called an approximation function. The remainder term / obtained after
the finite sum has been substituted into the equation (1) has the form

= [

] .

If the remainder / is identically equal to zero, then the function


equation (1). In general, / 1 0.

(4)
is the exact solution of

0.3.5-2. Galerkin method.


In order to find the coefficients #  in (3), consider another sequence of linearly independent
functions
j = j  ( )
( , = 1, 2,  , ).
(5)

Let us multiply both sides of (4) by j  and integrate the resulting relation over the region
= {   }, in which we seek the solution of equation (1). Next, we equate the cor1
2
responding integrals to zero (for the exact solutions, these integrals are equal to zero). Thus, we
obtain the following system of linear algebraic equations for the unknown coefficients #  :

X 

2
1

j 

/
M

 =0

( , = 1, 2,  , ).

(6)

Relations (6) mean that the approximation function (3) satisfies equation (1) on the
 2 average
(i.e., in the integral sense) with weights j  . Introducing the scalar product , % = X  %  of
1

arbitrary functions and % , we can consider equations (6) as the condition of orthogonality of the
remainder / to all weight functions j  .
The Galerkin method can be applied not only to boundary value problems but also to eigenand seeks eigenfunctions  , together with
value problems (in the latter case, one takes  =
eigenvalues  ).
Mathematical justification of the Galerkin method for specific boundary value problems can be
found in the literature listed at the end of Subsection 0.3.5. Below we describe some other methods,
which are in fact special cases of the Galerkin method.
x3y'z|{}~/
As a rule, one takes suitable sequences of polynomials or trigonometric functions as
Y  ( ) in the approximation function (3).
0.3.5-3. The BubnovGalerkin method, the moment method, and the least squares method.

1 b . The sequences of functions (2) and (5) in the Galerkin method can be chosen arbitrarily. In the
case of equal functions,
Y  ( ) = j  ( )
( , = 1, 2,  , ),
(7)
the method is often called the BubnovGalerkin method.

2003 by Chapman & Hall/CRC

Page 47

2 b . The moment method is the Galerkin method with the weight functions (5) being powers of  ,

j  =  .

(8)

3 b . Sometimes, the functions j  are expressed in terms of Y  by the relations

j  = [Y  ]

( , = 1, 2,  ),

is the differential operator of equation (1). This version of the Galerkin method is called
where
the least squares method.
0.3.5-4. Collocation method.
In the collocation method, one chooses a sequence of points   , , = 1,  , , and imposes the
condition that the remainder (4) be zero at these points,

 = 
( , = 1,  , ).
(9)
When solving a specific problem, the points   , at which the remainder / is set equal to zero,
are regarded as most significant. The number of collocation points is taken equal to the number
= 0 at

of the terms of the series (3). This allows one to obtain a complete system of algebraic equations for
the unknown coefficients #  (for linear boundary value problems, this algebraic system is linear).
Note that the collocation method is a special case of the Galerkin method with the sequence (5)
consisting of the Dirac delta functions:

j  = > (   ).
In the collocation method, there is no need to calculate integrals, and this essentially simplifies
the procedure of solving nonlinear problems (although usually this method yields less accurate
results than other modifications of the Galerkin method).
0.3.5-5. The method of partitioning the domain.

The domain = { 1   2 } is split into subdomains:


In this method, the weight functions are chosen as follows:

 = {  1    2 }, , = 1,  , .

1 for d  ,

0 for   .

j  ( ) =

The subdomains  are chosen according to the specific properties of the problem under consider
ation and can generally be arbitrary (the union of all subdomains  may differ from the domain
, and some and + may overlap).

0.3.5-6. The least squared error method.
Sometimes, in order to find the coefficients #  of the approximation function (3), one uses the least
squared error method based on the minimization of the functional:

` =X 

2
1

D

min .

(10)

For given functions Y  in (3), the integral ` is a quadratic polynomial with respect to the coefficients #  . In this case, the necessary conditions of minimum in (10) have the form:

# 

=0

(B = 1,  , ).

is a system of linear algebraic equations for the coefficients #  .


^This
_

References for Subsection 0.3.5: L. V. Kantorovich and V. I. Krylov (1962), M. A. Krasnoselskii, G. M. Vainikko,
P. P. Zabreiko et al. (1969), S. G. Mikhlin (1970), B. A. Finlayson (1972), D. Zwillinger (1989).

2003 by Chapman & Hall/CRC

Page 48

0.3.6. Iteration and Numerical Methods


0.3.6-1. The method of successive approximations (Cauchy problem).
The method of successive approximations is implemented in two steps. First, the Cauchy problem

    =  ( , ,   )
  ( 0 ) = 0
( 0 ) = 0 ,

(equation),
(boundary conditions)

(1)
(2)

is reduced to an equivalent system of integral equations by the introduction of the new variable
( ) =  . These integral equations have the form


( ) = 0 + X    a , (a ), (a )  a ,
M
0

( ) =

+X 

(a ) a .
M

(3)

Then the solution of system (3) is sought by means of successive approximations defined by the
following recurrence formulas:


+1 (  ) = 0 + X 

  a ,  (a ),  (a )  a ,
M

As the initial approximation, one can take

^_

 +1 ( ) =
0(

 )=

0,

0+X 

0 ( ) = 0 .

 (a ) a ;
M

B = 0, 1, 2, 

References for Paragraph 0.3.6-1: G. A. Korn and T. M. Korn (1968), N. S. Bakhvalov (1973), E. Kamke (1977).

0.3.6-2. The RungeKutta method (Cauchy problem).


For the numerical integration of the Cauchy problem (1)(2), one often uses the RungeKutta
method.
Let C be sufficiently small. We introduce the following notation:




 = (  ),
 =  (  ),   =  (  ,  ,  );
The desired values  and  are successively found by the formulas:

2
 +1 =  +  C + 16 (  1 +  2 +  3 )( C ) ,

 1
 +1 =  + 6 (  1 + 2  2 + 2  3 +  4 ) C ,
  =

+ , C ,

, = 0, 1, 2, 

where

 1 =     ,  ,   ,
 2 =     + 12 C ,  + 12  C ,  + 12  1 C  ,
 3 = l  + 12 C ,  + 12  C + 14  1 ( C )2 ,  + 12  2 C  ,
 4 =     + C ,  +  C + 12  2 ( C )2 ,  +  3 C  .
In practice, the step C is determined in the same way as for first-order equations (see Remark 2
in Paragraph 0.1.10-3).

^_

References for Paragraph 0.3.6-2: G. A. Korn and T. M. Korn (1968), N. S. Bakhvalov (1973), E. Kamke (1977),
D. Zwillinger (1989).

0.3.6-3. Shooting method (boundary value problems).


In order to solve the boundary value problem for equation (1) with the boundary conditions
( 1 ) =

1,

( 2 ) =

2,

(4)

one considers an auxiliary Cauchy problem for equation (1) with the initial conditions
( 1 ) =

1,

  ( 1 ) =  .

(5)

2003 by Chapman & Hall/CRC

Page 49

(The solution of this Cauchy problem can be obtained by the RungeKutta method or some other
numerical method.) The parameter  is chosen so that the value of the solution = ( ,  ) at the
point  =  2 coincides with the value required by the second boundary condition in (4):
( 2 ,  ) =

2.

In a similar way one constructs the solution of the boundary value problem with mixed boundary
conditions
  ( 2 ) + , ( 2 ) = 2 .
( 1 ) = 1 ,
(6)
In this case, one also considers the auxiliary Cauchy problem (1), (5). The parameter  is chosen so
that the solution = ( ,  ) satisfies the second boundary condition in (6) at the point  =  2 .

^_

References for Paragraph 0.3.6-3: S. K. Godunov and V. S. Ryabenkii (1973), N. N. Kalitkin (1978).

0.3.6-4. Method of accelerated convergence in eigenvalue problems.


Consider the SturmLiouville problem for the second-order nonhomogeneous linear equation
[  ( )   ]  + [ ( ) % ( )] = 0

(7)

with linear homogeneous boundary conditions of the first kind


(0) = (1) = 0.

It is assumed that the functions  ,    , , % are continuous and  > 0, > 0.

(8)

First, using the RayleighRitz principle, one finds an upper estimate for the first eigenvalue 01
[this value is determined by the right-hand side of relation (6) from Paragraph 0.2.5-3]. Then, one
solves numerically the Cauchy problem for the auxiliary equation
[  ( )   ]  + [ 01 ( ) % ( )] = 0

with the boundary conditions

(0) = 0,

  (0) = 1.

(9)
(10)

satisfies the condition ( 0 ,


where  0 < 1. The criterion of closeness
The function ( ,
the form of the inequality |1  0 | > ,
of the exact and approximate solutions, 1 and
where > is a sufficiently small given constant. If this inequality does not hold, one constructs a
refinement for the approximate eigenvalue on the basis of the formula:
0
1)

0
1 ) = 0,
0
1 , has

1
1

0
1

0  (1)

[  (1)]2

0 = 1  0,

(11)

2
= X ( ) 2 ( )  . Then the value 11 is substituted for 01 in the Cauchy problem
where
M
0
(9)(10). As a result, a new solution and a new point  1 are found; and one has to check whether
the criterion |1  1 | > holds. If this inequality is violated, one refines the approximate eigenvalue
by means of the formula:

2
1

1
1

1  (1)

[  (1)]2

1 = 1  1,

(12)

and repeat the above procedure.


x3y'z|{}~
Formulas of the type (11) are obtained by a perturbation method based on a transformation of the independent variable  (see Paragraph 0.3.4-1). If   > 1, the functions  , ,
and % are smoothly extended to the interval (1, ], where   .

x3y'z|{}~

The algorithm described above has the property of accelerated convergence

2
4
to 108
+1  , which ensures that the relative error of the approximate solution becomes 10
after two or three iterations for 0 0.1. This method is quite effective for high-precision calcula-

tions, is fail-safe, and guarantees against accumulation of roundoff errors.

2003 by Chapman & Hall/CRC

Page 50

x3y'z|{}~E

In a similar way, one can compute subsequent eigenvalues + ,

that end, a suitable initial approximation 0+ should be chosen).

= 2, 3,  (to

x3y'z|{}~E

A similar computation scheme can also be used in the case of boundary conditions
of the second and the third kinds, periodic boundary conditions, etc. (see the references below).
Example 1. The eigenvalue problem for the equation

+ (1 +

2 2

=0

= 16 2 1.
with the boundary conditions (8) admits an exact analytic solution and has eigenvalues 1 = 15, 2 = 63, ,
= sin( ) yields the approximate
According the RayleighRitz principle, formula (6) of Paragraph 0.2.5-3 for
value 01 = 15.33728. The solution of the Cauchy problem (9)(10) with ( ) = 1, ( ) = (1 + 2 )2 , ( ) = 0 yields
0 = 0.983848, 1 0 = 0.016152, 2 = 0.024585, ( 0 ) = 0.70622822.
The first iteration for the first eigenvalue is determined by (11) and results in the value 11 = 14.99245 with the relative
1 = 5 104 .
error
1
2 < 106 .
The second iteration results in 21 = 14.999986 with the relative error
1

Example 2. Consider the eigenvalue problem for the equation


( 1+

+  = 0

with the boundary conditions (8).


The RayleighRitz principle yields 01 = 11.995576. The next two iterations result in the values
2 = 11.898458. For the relative error we have
2 < 105 .
1
1

^ _

1
1

= 11.898578 and

References for Paragraph 0.3.6-4: L. D. Akulenko and S. V. Nesterov (1996, 1997).

For more details about finite-difference methods and other numerical methods, see, for instance,
the books by Lambert (1973), Keller (1976), and Zwillinger (1998).

0.4. Linear Equations of Arbitrary Order


0.4.1. Linear Equations with Constant Coefficients
0.4.1-1. Homogeneous linear equations.
An B th-order homogeneous linear equation with constant coefficients has the general form

 + 
( )



( 1)

+ ))) + 

 + 

= 0.

(1)

The general solution of this equation is determined by the roots of the characteristic equation:

( ) = 0,

where

( ) =  +  

+ ))) +  1 +  0 .

(2)

The following cases are possible:


1 b . All roots 1 , 2 ,  ,  of the characteristic equation (2) are real and distinct. Then the general
solution of the homogeneous linear differential equation (1) has the form:
=L

exp( 1  ) + L

exp( 2  ) + ))) + L  exp(   ).

equal real roots 1 = 2 = ))) = +


2 b . There are

distinct. In this case, the general solution is given by:

B ), and the other roots are real and

= exp( 1  )( L 1 + L 2  + ))) + L +  + 1 )
+ L + +1 exp( + +1  ) + L + +2 exp( + +2  ) + ))) + L  exp(   ).
equal complex conjugate roots = \AC (2 B ), and the other roots are real
3 b . There are

and distinct. In this case, the general solution is:

 + ))) + # +  + 1 )
+ exp(  ) sin( )( $ 1 + $ 2  + ))) + $ +  + 1 )
+ L 2 + +1 exp( 2 + +1  ) + L 2 + +2 exp( 2 + +2  ) + ))) + L  exp(   ),
where # 1 ,  , # + , $ 1 ,   , $ + , L 2 + +1 ,  , L  are arbitrary constants.
= exp(  ) cos( )( #

+#

2003 by Chapman & Hall/CRC

Page 51

4 b . In the general case, where there are & different roots 1 , 2 ,  ,  of multiplicities
,
,  , , respectively, the right-hand side of the characteristic equation (2) can be rep 1 2

resented as the product

where

1+

2+

))) +

( ) = ( 1)

+ 1 ( )+
2

+ ,
 (  )

= B . The general solution of the original equation is given by the formula:

=1

exp(   )( L 

,0

+ L  ,1  + ))) + L  , +

 +

),

where L  , are arbitrary constants.


If the characteristic equation (2) has complex conjugate roots, then in the above solution, one
should extract the real part on the basis of the relation exp( eAI ) =  9 (cos eAI sin ).
0.4.1-2. Nonhomogeneous linear equations.
An B th-order nonhomogeneous linear equation with constant coefficients has the general form

(  ) +  

( 

1)

+ ))) + 

 + 

=  ( ).

(3)

The general solution of this equation is the sum of the general solution of the corresponding
homogeneous equation (see Paragraph 0.4.1-1) and any particular solution of the nonhomogeneous
equation.
If all the roots 1 , 2 ,  ,  of the characteristic equation (2) are different, equation (3) has
the general solution:
=

=1


 
Ll  
+
.   
( " )
=1 

 ( ) 



M

(for complex roots, the real part should be taken).


Table 4 lists the forms of particular solutions corresponding to some special forms of functions
on the right-hand side of the linear nonhomogeneous equation.
In the general case, a particular solution can be constructed on the basis of the formulas from
Paragraph 0.4.2-3.

^_

References for Subsection 0.4.1: G. A. Korn and T. M. Korn (1968), E. Kamke (1977), A. N. Tikhonov, A. B. Vasileva,
and A. G. Sveshnikov (1980), D. Zwillinger (1989).

0.4.2. Linear Equations with Variable Coefficients


0.4.2-1. Homogeneous linear equations. Structure of the general solution.
The general solution of the B th-order homogeneous linear differential equation

  ( ) (  ) +  

1 (

 ) ( 

1)

+ ))) +  1 ( )   +  0 ( ) = 0

(1)

has the form:


=L

1 1(

 )+L

2 2(

 ) + ))) + L   ( ).

(2)

Here, 1 ( ), 2 ( ),  ,  ( ) is a fundamental system of solutions (the  are linearly independent


particular solutions,  1 0); L 1 , L 2 ,  , L  are arbitrary constants.

2003 by Chapman & Hall/CRC

Page 52

TABLE 4
Forms of particular solutions of the constant coefficient nonhomogeneous linear equation
(  )
+   1 (  1) + ))) +  1  +  0 =  ( ) that correspond to some special forms of the function  ( )
Form of the
function  ( )

.
.

Zero is not a root of the


characteristic equation (i.e.,  0 0)

+ ( )

+ ( )  9

Form of a particular

solution = ( )

Roots of the characteristic equation


 +   1  1 + ))) +  1 +  0 = 0

Zero is a root of the


characteristic equation (multiplicity & )

characteristic equation

( is a real constant)

is a root of the

characteristic equation (multiplicity & )

is not a root of the

+ ( )

+ ( )  9

+ ( )  9




( ) cos 
+ I ( ) sin 

characteristic equation

+ ( ) cos 
+  ( ) sin 

.


is not a root of the

+ ( )

.
 [ ( ) cos 
characteristic equation (multiplicity & )
+ I ( ) sin  ]
.
+ is not a root of the
[ ( ) cos  
.
characteristic equation
+ I ( ) sin  ] 9
[ + ( ) cos  
.
9
+  ( ) sin  ]
+ is a root of the
 [ ( ) cos  
characteristic equation (multiplicity & )
+ I ( ) sin  ] 9
.
. .
Notation:
+ and  are polynomials of degrees and B with given coefficients; + , ,

and I are polynomials of degrees and @ whose coefficients are determined by substituting

the particular solution into the basic equation; @ = max( , B ); and and are real numbers,

2 = 1.
is a root of the

0.4.2-2. Utilization of particular solutions for reducing the order of the original equation.
1 b . Let

1(

 ) be a nontrivial particular solution of equation (1). The substitution

 ) XZH ( ) 
M
results in a linear equation of order B 1 for the function H ( ).
2 b . Let 1 = 1 ( ) and 2 = 2 ( ) be two nontrivial linearly independent solutions of equation (1).
=

The substitution

1(

results in a linear equation of order B 2 for ( ).

linearly independent solutions 1 ( ), 2 ( ),  , + ( ) of equation (1) are


3 b . Suppose that

known. Then one can reduce the order of the equation to B by successive application of the

following procedure. The substitution = + ( ) X H ( )  leads to an equation of order B 1 for


M
the function H ( ) with known linearly independent solutions:

H 1= O

,
+ P 

H 2= O

,
+ P 

 ,

H +

= O

+
+

.
P 

2003 by Chapman & Hall/CRC

Page 53

The substitution H = H + 1 ( ) X ( )  yields an equation of order B 2. Repeating this procedure


M
times, we arrive at a homogeneous linear equation of order B .

0.4.2-3. Wronskian determinant and Liouville formula.


The Wronskian determinant (or simply, Wronskian) is the function defined as:
1 ( )
 ( )
1
( ) =
)))
(  1)

1 ( )

)))
)))
)))
)))

 ( )
 ( )

,
)))
(  1)
( )


(3)

1 (  ),  ,  (  ) is a fundamental system of solutions of the homogeneous equation (1);


+3
+ ) ( ) = M
 , = 1,  , B 1; , = 1,  , B .


 +
M
The following
Liouville formula holds:

where
(

( ) =

( 0 ) exp X 

 
 1 (a ) a .

0
 (a ) M

0.4.2-4. Nonhomogeneous linear equations. Construction of the general solution.


1 b . The general nonhomogeneous B th-order linear differential equation has the form

  ( ) (  ) +  



( 1)

+ ))) +  1 ( )   +  0 ( ) = ( ).

(4)

The general solution of the nonhomogeneous equation (4) can be represented as the sum of its
particular solution and the general solution of the corresponding homogeneous equation (1).
2 b . Let 1 ( ),  ,  ( ) be a fundamental system of solutions of the homogeneous equation (1),

and let ( ) be the Wronskian determinant (3). Then the general solution of the nonhomogeneous
linear equation (4) can be represented as:
=

Ll ( ) +
=1

( ) X
=1

( ) 
M ,
  ( ) ( )

where ( ) is the determinant obtained by replacing the @ th column of the matrix (3) by the
column vector with the elements 0, 0,  , 0, .

^_

References for Subsection 0.4.2: G. A. Korn and T. M. Korn (1968), E. Kamke (1977), A. N. Tikhonov, A. B. Vasileva,
and A. G. Sveshnikov (1980), D. Zwillinger (1989).

0.4.3. Asymptotic Solutions of Linear Equations


This subsection presents asymptotic solutions, as D 0 ( > 0), of some higher-order linear ordinary
differential equations containing arbitrary functions (sufficiently smooth), with the independent
variable being real.
0.4.3-1. Fourth-order linear equations.
1 b . Consider the equation

       ( ) = 0

2003 by Chapman & Hall/CRC

Page 54

on a closed interval    . With the condition  > 0, the leading terms of the asymptotic
expansions of the fundamental system of solutions, as pD 0, are given by the formulas:
1
1
3 J 8
exp [  ( )]1 J 4  , 2 = [  ( )]3 J 8 exp
[  ( )]1 J 4  ,
1 = [  (  )]

= [  ( )]3 J 8 cos

[  ( )]1 J 4  ,

M

= [  ( )]3 J 8 sin

[  ( )]1 J 4  .

M

2 b . Now consider the biquadratic equation:


4      2 2 ( )     ( ) = 0.
(1)
Introduce the notation:
c ( ) = [ ( )]2 +  ( ).
In the range where the conditions  ( ) 0 and c ( ) 0 are satisfied, the leading terms of the
asymptotic expansions of the fundamental system of solutions of equation (1) are described by the
formulas:
1
1
[ ( )] 
1 J 2
1 J 4
 ( )  
 ; , = 1, 2, 3, 4,
exp

 = [  ( )] [ c ( )]

M
2
c ( ) M
where
1 ( ) = ( ) + c ( ),
2 ( ) = ( ) + c ( ),

3 ( ) =

4 ( ) = ( ) c ( ).

( ) c ( ),

0.4.3-2. Higher-order linear equations.


1 b . Consider an equation of the form

 (  )  ( ) = 0
on a closed interval    . Assume that  0. Then the leading terms of the asymptotic
expansions of the fundamental system of solutions, as pD 0, are given by:
1
+
1 + 1
V  ( )W   V 1 + [ ( )W ,
+ = V  ( )W 2 2  exp


M

where 1 , 2 ,  ,  are roots of the equation = 1:
2

+ = cos O /
+ sin O /
,
B P
B P

= 1, 2,  , B .

2 b . Now consider an equation of the form


 (  ) +  1   1 ( ) (  1) + ))) +  1 ( )   +  0 ( ) = 0
(2)
on a closed interval    . Let + = + ( ) ( = 1, 2,  , B ) be the roots of the characteristic

equation:
. ( , )  +  ( )  1 + ))) +  ( ) +  ( ) = 0.
 1
1
0
Let all the roots of the characteristic equation be different on the interval    , i.e., the
, , are satisfied, which is equivalent to the fulfillment of the
conditions + ( )  ( ),
. ( , + ) 0. Then
the leading terms of the asymptotic expansions of the fundamental
conditions

system of solutions of equation (2), as ID 0, are given by:
.  , ( ) 
1
1
+


 ,
+ = exp + ( )  [ + ( )] . 

M
2
, + ( )  M




where
.

 1 + (B 1)   1 ( )  2 + ))) + 2 " 2 ( ) +  1 ( ),
 ( , ) N = B
. ( , ) N N 2 . = B (B 1)  2 + (B 1)(B 2)  ( )  3 + ))) + 6 " ( ) + 2  ( ).
3
2
 1

2
^_
N
References for Subsection 0.4.3: W. Wasov (1965), M. V. Fedoryuk (1993).

2003 by Chapman & Hall/CRC

Page 55

0.5. Nonlinear Equations of Arbitrary Order


0.5.1. Structure of the General Solution. Cauchy Problem
0.5.1-1. Equations solved for the highest derivative. General solution.
An B th-order differential equation solved for the highest derivative has the form

  =  ( , ,   ,  , ( 
( )

1)

).

(1)

The general solution of this equation depends on B arbitrary constants L 1 ,  , L  . In some
cases, the general solution can be written in explicit form as = Y ( , L 1 ,  , L  ).
0.5.1-2. The Cauchy problem. The existence and uniqueness theorem.
1 b . The Cauchy problem: find a solution of equation (1) with the initial conditions
( 0 ) =

0,

  ( 0 ) =

(1)
0 ,

 ,

( 

1)

( 0 ) =

( 1)
.
0

(2)

(At a point  0 , the values of the unknown function ( ) and all its derivatives of orders B 1 are
prescribed.)
2 b . The existence and uniqueness theorem. Suppose the function  ( , , H 1,  , H  1 ) is continuous in all its arguments in a neighborhood of the point ( 0 , 0 , 0(1) ,  , 0(  1) ) and has bounded
derivatives with respect to , H 1 ,  , H  1 in this neighborhood. Then a solution of equation (1)
satisfying the initial conditions (2) exists and is unique.

^_

References for Subsection 0.5.1: G. A. Korn and T. M. Korn (1968), I. G. Petrovskii (1970), E. Kamke (1977),
A. N. Tikhonov, A. B. Vasileva, and A. G. Sveshnikov (1980).

0.5.2. Equations Admitting Reduction of Order


0.5.2-1. Equations not containing ,  ,  , (  ) explicitly.
An equation that does not explicitly contain the unknown function and its derivatives up to order ,
inclusive can generally be written as

   , ( 

+1)

 ) = 0
,  , (

(1 , + 1 < B ).

(1)

+const (the
Such equations are invariant under arbitrary translations of the unknown function, D
form of such equations is also preserved under the transformation ( ) = +     + ))) +  1  +  0 ,
where the  + are arbitrary constants). The substitution H ( ) = (  +1) reduces (1) to an equation
whose order is by , + 1 smaller than that of the original equation,  , H , H   ,  , H (   1)  = 0.
0.5.2-2. Equations not containing  explicitly (autonomous equations).
An equation that does not explicitly contain  has in the general form

 )  = 0.
  ,   ,  , (

(2)

Such equations are invariant under arbitrary translations of the independent variable, CD  + const .
The substitution   = ( ) (where plays the role of the independent variable) reduces by one
the order of an autonomous equation. Higher derivatives can be expressed in terms of and its
derivatives with respect to the new independent variable, "  = *2 ,    = 2 2  + (2 )2 , 



2003 by Chapman & Hall/CRC

Page 56

( ) 
0.5.2-3. Equations of the form    +  ,  ,  , 
= 0.

Such equations are invariant under simultaneous translations of the independent variable and the
!( , where ( is an arbitrary constant.
unknown function, ED + ( and D
For  = 0, see equation (1). For  0, the substitution ( ) = +(
 ) leads to an autonomous
equation of the form (2).

 )  = 0 and its generalizations.


0.5.2-4. Equations of the form  ,   ,   ,  , (

The substitution ( ) =    reduces the order of this equation by one.


This equation is a special case of the equation

   ,    , ( +

The substitution ( ) =   

+1)

,  , (  )  = 0,

where

= 1, 2,  , B 1.

(3)

reduces by one the order of equation (3).

0.5.2-5. Homogeneous equations.


1 b . Equations homogeneous in the independent variable are invariant under scaling of the independent variable, |D , where is an arbitrary constant ( 0). In general, such equations can be
written in the form
 )  = 0.
  ,    ,  2    ,  ,   (
The substitution H ( ) = 

  reduces by one the order of this equation.

2 b . Equations homogeneous in the unknown function are invariant under scaling of the unknown
function, D , where is an arbitrary constant ( 0). Such equations can be written in the
general form
  ,  
 ,   
 ,  , (  )
  = 0.

The substitution H ( ) =  
 reduces by one the order of this equation.

3 b . Equations homogeneous in both variables are invariant under simultaneous scaling (dilatation)
of the independent and dependent variables, CD  and D , where is an arbitrary constant
( 0). Such equations can be written in the general form

 )  = 0.
  
 ,   ,     ,  ,   1 (
The transformation a = ln | |, = 
 leads to an autonomous equation considered in Paragraph 0.5.2-2.

0.5.2-6. Generalized homogeneous equations.


1 b . Generalized homogeneous equations (equations homogeneous in the generalized sense) are
invariant under simultaneous scaling of the independent variable and the unknown function, uD 
and D  , where 0 is an arbitrary constant and , is a given number. Such equations can
be written in the general form
( ) 
    ,  1    ,  ,    
= 0.

=  
leads to an autonomous equation considered in ParaThe transformation a = ln  ,

graph 0.5.2-2.

2 b . The most general form of generalized homogeneous equations is

   + ,   
 ,  ,   (  )
  = 0.
The transformation H =   + , =  
 reduces the order of this equation by one.
2003 by Chapman & Hall/CRC

Page 57

   
  

,  ,  ,  , (  ) 
 = 0.

0.5.2-7. Equations of the form  

Such equations are invariant under simultaneous translation and scaling of variables, eD  + and

DG , where = exp(s
B ), and is an arbitrary constant. The transformation H = !  ,

= 
 leads to an equation of order B 1.
0.5.2-8. Equations of the form       ,   ,  2   ,  ,  

   = 0.

( )

Such equations are invariant under simultaneous scaling and translation of variables,  D  and
+ , where = exp(h
B ), and is an arbitrary constant. The transformation H =    ,
D
=   leads to an equation of order B 1.
0.5.2-9. Other equations.
Consider the nonlinear differential equation

   , L1 [ ],  , L  [ ]  = 0,

(4)

forms,
where the L [ ] are linear homogeneous differential

L [ ]=
Let

0(

=0

Y (+ ) ( ) ( + ) ,

= 1,  , , .

 ) be a common particular
solution of the linear equations:
( = 1,  , , ).

L [ 0] = 0
Then the substitution

= Y ( ) V ( )    ( ) W
0
0

(5)

with an arbitrary function Y ( ) reduces by one the order of equation (4).


Example 1. Consider the third-order equation

= (

2 ).

It can be represented in the form (4) with

= 2,

, , )=

The linear equations L [ ] = 0 are

( ),

= 0,

L2 [ ] =

2 = 0.

2 .

2 . Therefore, the substitution


= 2
These equations have a common particular solution

with ( ) = 1 ) leads to an autonomous second-order equation of the form 2.9.1.1: = ( ).

Example 2. The th-order equation

( E

can be represented in the form (4) with

= 2,

The linear equations


have a common particular solution,
leads to an ( 1)st-order equation.

^_

, , )=

L1 [ ]

0=

0=

L1 [ ] =

= ( ,

( , ),

( E

= 0,

(see formula (5)

)+

L1 [ ] =

( E

L2 [ ]

. Therefore, the substitution

L2 [ ] =

=0
=

(see formula (5) with

( ) =

References for Subsection 0.5.2: G. M. Murphy (1960), G. A. Korn and T. M. Korn (1968), E. Kamke (1977), V. F. Zaitsev
and A. D. Polyanin (1993, 2001), A. D. Polyanin and V. F. Zaitsev (1995), D. Zwillinger (1998).

2003 by Chapman & Hall/CRC

Page 58

0.5.3. A Method for Construction of Solvable Equations of General


Form
0.5.3-1. Description of the method.
Consider a function

=  ( , L

L 2 ,  , L 

1,

+1 )

(1)

depending on B + 1 free parameters L  . Differentiating relation (1) B times, we obtain the following
sequence of equations:

  =   (  ) ( , L 1 , L 2 ,  , L 
( )

, = 1, 2,  , B .

+1 ),

(2)

Treating relations (1), (2) as an algebraic (transcendental) system of equations for the parameters
L 1 , L 2 ,  , L  +1 and solving this system, we obtain

L  = Y    , ,   ,  , (  )  ,

, = 1, 2,  , B + 1.

(3)

Consider a general B th-order equation of the form

 ( Y 1 , Y 2 ,  , Y 

+1 )

= 0,

(4)

 )  are the functions


where  is an arbitrary function of (B +1) variables and Y  = Y    , ,  ,  , (
from (3). Equation (4) is satisfied by the function (1), where the (B + 1) arbitrary parameters
L 1 , L 2 ,  , L  +1 are related by a single constraint:
 ( L 1 , L 2 ,  , L 

+1 )

= 0.

x3y'z|{}~

Equation (4) may also have singular solutions depending on a smaller number of
arbitrary constants. In order to examine these solutions, one should differentiate equation (4); see
Example 1.

x3y'z|{}~

Instead of (4), one can consider a more general equation

 (j 1 , j 2 ,  , j 

+1 )

= 0,

j  = j  ( Y 1 , Y 2 ,  , Y 

where

+1 ).

x3y'z|{}~E
The original expression (1) can be specified in an implicit form.
x3y'z|{}~E
The original expression (1) can be written as an th-order differential equation

( < B ) with B + 1 free parameters L  . The solution of the B th-order differential equation
obtained in this way can be expressed in terms of the solution of an th-order differential equation

(see Example 4).


0.5.3-2. Examples.
Example 1. Consider the function

By differentiation we obtain
Let us solve equations (5)(6) for the parameters

and

2.

(5)

(6)

2 . We have

+ .

Using the above method, we construct an equation in accordance with (4):

 , 

This equation admits a solution of the form (5) with constants

+
1

= 0.

and

(7)
related by the constraint

 1, 

2)

= 0.

2003 by Chapman & Hall/CRC

Page 59

, we get

Singular solution. Differentiating equation (7) with respect to

+  )(
( 

) = 0,

(8)

where the subscripts and indicate the respective partial derivatives of the function = ( , ). Equating the first factor
in (8) to zero, we obtain solution (5). Equating the second factor to zero, we obtain an expression which, combined with
equation (7), yields a singular solution in parametric form:

One should eliminate

( , ) = 0,

= 0,

where

:, =:

+ .

from these expressions.

Example 2. Consider the function

Differentiating this function twice, we get

Solving (9)(10) for the parameters


1
2

=2
=2

2,

3.

(9)
(10)

1.

, we find that

1
2

2 

These relations lead to a second-order equation of general form:

1
2

 , 

 ,

1
2

2  

= 0,

 1 ,  2 , and  3 related by the constraint (  1 ,  2 , 


Example 3. In Example 2, one can choose the functions  of the form (see Remark 2)

 1 = 2 1 ,  2 = 2 ,  3 = 4 1 3 22 ,

which has a solution of the type (9) with the three constants

where

1
2

Its solution is given by (9) with three constants

 1, 

1
2

2 , and

Example 4. Consider the autonomous equation

= 0.

. As a result, we obtain the differential equation:

, 2
3

3)

)2

= 0.

related by a single constraint

(2 

 

1, 2, 4 1

2
3)

2.

= 0.

(11)

Its solution can be represented in implicit form (see 2.4.2.1 and 2.9.1.1). Differentiating (11), we obtain

D 1 1  .
Let us solve equations (11)(12) for the parameters  1 and  2 :
 1 = +1 ,  2 = +
Taking  1 = 1 and  2 = 2 (see Remark 2), we obtain the equation:

C

+1

(12)

= 0.

This equation is satisfied by the solutions of a second-order autonomous equation of the form (11), where the constants
and 2 are related by the constraint ( 1 , 2 ) = 0.

D D

0.6. Lie Group and Discrete-Group Methods


0.6.1. Lie Group Method. Point Transformations
0.6.1-1. Local one-parameter Lie group of transformations. Invariance condition.
Here, we examine transformations of the ordinary differential equation

  =    , ,  , ( 
( )

1)

.

(1)

2003 by Chapman & Hall/CRC

Page 60

Consider the set of transformations


T =

 |

 = Y ( , , ),
= j ( , , ),

=0

=0

= ,
= ,

(2)

where Y , j are smooth functions of their arguments and is a real parameter. The set T is called
a continuous one-parameter Lie group of point transformations if, for any 1 and 2 , the following
relation holds: T 1 T 1 = T 1 + 2 , i.e., consecutive application of two transformations of the form (1)
with parameters 1 and 2 is equivalent to a single transformation of the same form with parameter
1 + 2.
In what follows, we consider local continuous one-parameter Lie groups of point transformations
(briefly called point groups) corresponding to an infinitesimal transformation (2) for D 0. Taylors
expansion of  and in (2) with respect to the parameter about = 0 yields:


v + ( , ) ,

+ (  , ) ,

(3)

where

Y ( , , )
j ( , , )
, ( , ) = N

=0 .



=0

N
N

to the curve described by the transformed points


At each point ( , ), the vector ( , ) is tangent
( , ).
( , ) = N

The first-order linear differential operator

X = ( , ) N

+ ( , ) N

(4)

corresponding to the infinitesimal transformation (3), is called the infinitesimal operator (or infinitesimal generator) of the group.
By definition, the universal invariant (briefly, invariant) of the group (2) and the operator (4) is
a function 0 ( , ), satisfying the condition 0 ( , ) = 0 ( , ). Taylors expansion with respect to
the small parameter yields the following linear partial differential equation for 0 :
X

= ( , ) N

+ ( , ) N

= 0.

Equation (1) will be treated as a relation for B +2 variables  , ,   ,  , ( 


constraints



( +1)

= M

( )

(5)
)

with the differential

(6)

The space of these B + 2 variables is called the space of B th prolongation; and in order to work
with differential equations, one has to define the action of operator (4) on the new variables
 ,  , (  ) , taking into account the differential constraints (6). For example, let us calculate the
infinitesimal transformation of the first derivative. We have

c  ( +  )
=
 c  ( +  )

 + (  +  )

1 + (  +   )


( c  =  + 
+ ))) is the operator of total derivative). Expanding the right-hand side into a power
N
N 
series with respect to the parameter and preserving
 the first-order terms, we obtain
where


1

=  + (

  + 1 (  , ,   ) ,

 

 2
 
 )  ( ) = c ( ) c ( ).

The action of the group on higher-order


 derivatives
 is determined by the recurrence formula:

+1

= c  (  ) ( 

+1)

c  ( ).

2003 by Chapman & Hall/CRC

Page 61

To a prolonged group there corresponds a prolonged operator:


X = ( , ) N

+ ( , ) N





( ) 
   , , ,  , 

=1

N ( ) .


(7)

The ordinary differential equation (1) admits the group (2) if


X V (  )    , ,   ,  , ( 

1)

W

= 0.
 ( n ) =

(8)

Relation (8) is called the invariance condition.


x3y'z|{}~/
The invariant 0 , which is a solution of equation (5), also satisfies the equation
X 0 = 0.

0.6.1-2. Group analysis of second-order equations. Structure of an admissible operator.


For second-order nonlinear equations

    =  ( , ,   ),

(9)

the invariance condition (8) is written in the form

 + (2     )   + (   2   )(   )2  (   )3



= (2  + 3  )  +   +  + [  + (  )  (  )2 ]  ,







where  =  ( , ,  ). This condition is in fact a second-order partial differential equation for
two unknown functions ( , ) and ( , ). Since the unknown functions do not depend on the
derivative   , this equation can be represented (after  has been expanded in a power series with
respect to   , unless it is already a polynomial) in the form

=0

`  (   )  = 0,

(10)

with the `  independent of   . In order to ensure that condition (10) holds identically, one should
set `  = 0, , = 0, 1,  Thus, the invariance condition for a second-order equation can be split
and represented as a system of equations (whose number can generally be infinite).

Example 1. If = ( , ), i.e., the right-hand side of equation (9) does not depend on
equation can be split and represented as the system:

~2N = 0,
N 2~  = 0,
2   ~ 3 ( , ) ~2 = 0,
 + ( 2~ ) ( , ) (

~  (

, then the determining

, )

, ) = 0.

From the first two equations we find that


where ( ), ( ),
equations, we get

), and
3

= ( ) + ( ),

( )

+ ( ) + ( ),

( ) are arbitrary functions. Substituting these expressions into the third and the fourth
+ 2 3 ( , ) = 0,

+ ( 2 )

( + )

+ + )

= 0.

(11)

In what follows, it is assumed that the function ( , ) is nonlinear with respect to the second argument. Then from the
is an arbitrary constant. The second equation in (11)
first equation in (11), we find that = 0 and = 12 + , where
becomes
1
+ + 32 12 + + = 0.
(12)
2

Equation (12) allows us to solve two different problems.

2003 by Chapman & Hall/CRC

Page 62

1 . If the function ( , ) is given, then, splitting equation (12) with respect to powers of (the unknown functions
and are independent of ), we obtain a new system, from which , , and can be found; i.e., we ultimately obtain an

admissible
operator.

2 . Assuming that the functions , and the constant are known but arbitrary, one can regard relation (12) as an equation

this equation, we obtain a class of equations admitting a point operator.


for the unknown function ( , ). Solving

1
2

E D , i.e., we are dealing with the EmdenFowler equation. Then equation (12) becomes
+ + 32 5 1 12 + + 1 = 0.
E
E
E

( ,

Example 2. Let

)=

This relation must be satisfied identically by any function = ( ), and therefore, the coefficients of different powers of
must be equal to zero. As a result, we obtain a new system whose structure essentially depends on the value of .

( ,

1 . It was assumed above that


Then the system has the form:

) is nonlinear in its second argument, and therefore,

0 and

1. Let

2.

= 0,
= 0,

It follows that

= 0 and ( ) =
(

+ 3)

= 0,

1
(3
2

! = 0.

+ 0 , and the last equation of the system can be written in the form

(1

1
(
2

+ 2 + 3)

1)

+ (

+ !

= 0.

(13)

To ensure relation (13), we equate all coefficients of this quadratic trinomial to zero to obtain
(

+ 3)

1
(
2

= 0,

+ 2 + 3)

+ (

5! 0 = 0.

1) = 0,

(14)

Analysis of system (14) yields solutions of the determining system corresponding to three different operators:

X1 = (
X2 =

2 . Let

X3 =

1)

( + 2)

if
if

5
5

if

and
= 0,
+

are arbitrary,

+ 3 = 0.

= 2. Then equation (12) becomes

Equating the term

( ) =

1
2

and the coefficient of


1

5
2

in parentheses to zero, we get

0,

4 0 +3
4 1 +4
+
+
( ) =

( + 2)( + 3)( + 4) ( + 1)( + 2)( + 3)

5 E 1 2 = 0.

+ !

5 E 5

5 E 5

+ 0,

The expression in square brackets (the coefficient of 2 ) can be split with respect to powers of
system which, to within nonzero coefficients, has the form:
(7
(7
(
(2

5!

1, 2, 3, 4.

and we obtain an algebraic

+ 20) 1 = 0,

+ 15) 0 = 0,

+ 5) 2 = 0,
+ 5) 1 + 2 = 0,
0 = 0.

The last three equations coincide with the corresponding equations of system (14), whose solutions are already known.
The first two equations yield two cases of prolongation of the admissible group:

X1 = 343 8 7

X2 = 343 6 7

+ 4 49 1 7 3 

+ 3 49 1 7 + 4 

if
if

= 20
,
7
= 15
.
7

Page 63

0.6.1-3. Utilization of local groups for reducing the order of equations and their integration.
Suppose that an ordinary differential equation (1) admits an infinitesimal operator X of the form
(4). Then the order of the equation can be reduced by one. Below we describe two methods for
reducing the order of an equation.
1 b . The first method. The transformation

a =  ( , ),

= ( , ),

(15)

with  and ( *1 0) being arbitrary particular solutions of the first-order linear partial differential
equations

( , ) N

+ ( , ) N

=, ,
(16)

( , ) N + ( , ) N

N
N

= 0,

reduces equation (1) to an autonomous equation (the constant , 0 can be chosen arbitrarily). The
function = ( , ) is a universal invariant of the operator X.
Suppose that the general solution of the characteristic equation

has the form

M
( , )

( , )

( , ) = L ,

where L is an arbitrary constant. Then the general solutions of equations (16) are given by (see
A. D. Polyanin, V. F. Zaitsev, and A. Moussiaux, 2002):

 = ,X

+ i 1 ( ),
( , )
= i 2 ( ), = ( , ),

where i 1 ( ) and i 2 ( ) are arbitrary functions,   , ( , )  ( , ), and


regarded as a parameter.

Example 3. The EmdenFowler equation


Example 2):
X = ( , )

Equations (16) for

Solving (15) for

= 49

+ ( , )

~ (

where

1 7

15 7 2

3 7

6 7 ,  (

6
49

, we obtain the transformation

: 7,

: 3

= 49

2 7

, ) = 147

1 7

+ 12.

:,

6
49

which reduces the original equation to the autonomous equation

89 8

admits the operator (cf. the operator X2 in Item 2 of

, ) = 343

admit the particular solutions

and

in the integral is

which can easily be integrated by quadrature.

2 b . The second method. Suppose that we know two invariants of the admissible operator X:

0 = 0 (
1 = 1 (

, )
, ,  )

(universal invariant),
(first differential invariant*).

(17)
(18)

* By definition, an th-order differential invariant of the operator X is a function


=
= 0 with the operator X defined by (7).
satisfying the linear partial differential equation X

, , , , ( )

2003 by Chapman & Hall/CRC

Page 64

Then the second differential invariant can be found by differentiation,

1,

, ,   ,    ) = M

2 (

(19)

where + = ( c  + )  . Using (18)(19), let us eliminate the derivatives   and    from the
M
M
original equation and take into account relation (17). Thus we obtain the first-order equation:

1
M
M

Example 4. The EmdenFowler equation


whose first prolongation has the form:
X=
1

This operator admits the invariants:

6 3 (see 2.3.1.3,
the special case
+  + (  )  .


0= %

= 4 ( 0 , 1 ).

1 = 

= 3) admits an operator

(20)

which form an integral basis of the first-order linear partial differential equation

+ (

= 0.

Using (19) and (20), we find the second invariant:

2= 1

(21)

Let us express the unknown function and its derivatives from (20)(21) to obtain



where

Substituting these expressions into the original equation, we see that the variable
form
h = 3 ,

0,

1.

is canceled and the equation takes the



i.e., it becomes a first-order separable equation.

^_

References for Subsection 0.6.1: G. W. Bluman and J. D. Cole (1974), L. V. Ovsiannikov (1982), J. M. Hill (1982),
P. J. Olver (1986), G. W. Bluman and S. Kumei (1989), H. Stephani (1989), N. H. Ibragimov (1994).

0.6.2. Contact Transformations. Backlund


Transformations.
Formal Operators. Factorization Principle
0.6.2-1. Contact transformations.
The set of transformations
T =



 = Y ( , ,  , ),
= j ( , ,  , ),
  = ( , ,  , ),

 |

=0 = 
| =0 =
  | =
=0

,
,



(1)

(here, Y , j , are smooth functions of their arguments and is a real parameter) is called a
continuous one-parameter Lie group of tangential transformations (or simply, a tangential or contact
group) if T 1 T 2 = T 1 + 2 , i.e., if successive application of transformations (1) with parameters 1
and 2 is equivalent to the same transformation with parameter 1 + 2 . The transformed derivative
  depends only on the first derivative   and does not depend on the second derivative. Thus, the
functions Y and j in (1) cannot be arbitrary but are related by (see Paragraph 0.1.8-1):

Y
j
j
Y
Y
j
O N +  N
N  ON +   N
N 
= 0,



P

P
N
N
N
N
N
N
2003 by Chapman & Hall/CRC

Page 65

where the function

is defined by

= N 


Y
N  .


Proceeding as in Paragraph 0.6.1-1, we consider the Taylor expansions of  , , and /  in (1) with
We have
respect to the parameter about = 0, preserving only the first-order terms.



 + ( , ,   ) ,

where

Y ( , ,   ,
( , ,   ) = N

N

+ (  , ,   ) ,




 ) = N j ( , ,  ,

,

(

,
,
=0

On the other hand,

c  ( +  )
=
 c  ( +  )

 M

  + (  , ,   ) ,

=0


( , ,  ) = N

 + (  +  +


1+
 (  +   +  

( , ,   , )

  )
.
   )

=0

(2)


Expanding (2) with respect to and requiring that be independent of   , we find that three the

functions , , and are expressed in terms of a single function ( , ,  ) as follows:


=N
N



 N



= N

+  N

(3)

To an infinitesimal tangential transformation (1) there corresponds the infinitesimal operator:


X = ( , ,   ) N + ( , ,   ) N

N
N

+ ( , ,   ) N 


(4)

whose coordinates satisfy relations (3).


The action of the group on higher
 derivatives
 is determined by the recurrence formula:




+1

= c  (  ) ( 

+1)

c  ( ),

where 1 = . The invariance condition and the algorithm of finding tangential operators (4) admitted
by ordinary differential equations are similar as those for point operators. The only difference is that
the coordinates of the tangential operator depend on the first derivative; therefore, the determining
equation can be split and reduced to a system only in the case of equations whose order is greater or
equal to three.
x3y'z|{}~/
There are no tangential transformations of finite order , > 1 other than prolonged
point transformations and contact transformations [these transformations are described by formulas
similar to (1) and, in addition to  ,   , contain higher derivatives of up to order , inclusive].
0.6.2-2. Backlund transformations. Formal operators and nonlocal variables.
1 b . If the coordinates of the infinitesimal operator are allowed to depend on the derivatives of
arbitrary (up to infinity) orders, we obtain LieBacklund groups (of tangential transformations of
infinite order). However, on the manifold determined by an ordinary differential equation, all higher
derivatives are expressed through finitely many lower derivatives, as dictated by the structure of
the equation itself and the differential relations obtained from the equation. The substitution of
the right-hand side of equation (1) into an infinite series with derivatives usually results in very
cumbersome formulas hardly suitable for practical calculations. For this reason, the LieB a cklund
groups are widely used only for the investigation of partial differential equations, whereas in the
case of ordinary differential equations, a more effective approach is that based on the canonical form
of an operator and the notion of a formal operator.

2003 by Chapman & Hall/CRC

Page 66

2 b . The canonical form X is defined by the relation


X = X (  , ) c  = [ (  , ) ( , )  ] N ,
N

where X = ( , )  + ( , ) is the infinitesimal operator of the group [see formula (4) in


Paragraph 0.6.1-1], and c  is the operator of total derivative. The operators X and X are
equivalent in the sense that if one of them is admissible for the equation, then the other is also
admissible (the operator of total derivative is admissible for any ordinary differential equation). The
function 0 ( , )  is an invariant of any operator in canonical form.
 order derivatives

The action of the group on higher
for an operator in canonical form is determined

by the simple recurrence formula  +1 = c  (  ). The order of an equation that admits an operator
in canonical form can be reduced on the basis of the algorithm described in Paragraph 0.6.1-3 (see
Item 2 b , the second method).
3 b . By definition, a formal operator is an infinitesimal operator of the form

=`

,
N 

(5)

where the function ` depends on  , ,  ,  , (  ) (with , smaller than the order of the equation
under investigation) and auxiliary variables whose definition involves the symbol of indefinite

integral, for instance,
X ( , ,   ) 

(the integration is with respect to the variable  which is involved both explicitly and implicitly,
through the dependence of on  ). Such auxiliary variables are called nonlocal, in contrast to the
coordinates of the prolonged space defined pointwise. The nonlocal variables depend on derivatives
of arbitrarily high order, for instance,

 + +1 ( + )
 .
 = (1) +
( + 1)!
M
+ =0

This formula is obtained by successive integration by parts of its left-hand side. Thus, a nonlocal
variable can be represented as an infinite formal series; and this allows us to express the coordinates
of the LieBacklund operator in concise form.
A formal operator is a far-reaching generalization of an operator in canonical form. The function
0 ( , )  is an invariant of the formal operator (5) for any ` .
When solving the direct problem, one usually prescribes the nonlocal operator in the general


form
 + 2S
or X = O 1 X
 + 2 ,
(6)
X = 1 exp O X

N 

M P

P/N 

and then, in order to find an admissible operator, one uses a search algorithm similar to that



described in Paragraph 0.6.1-2. The coordinates of the prolonged operator are found by the formulas

 = c  (  1 ), where 0 = ` . In contrast to the method of finding a point operator, in the present
case, there are three unknown functions ( 1 , 2 , ); and the splitting procedure to obtain a system
can be realized with respect to all independent variables, in particular, the nonlocal variables.
Suppose that the differential equation

  =   , ,   ,  ,
can be written in new variables  = 0 , H = 1 ( , ,  ), H
( )

 
?
  , H    ,  , H ( 
( 1)

invariants of an admissible operator of the form (5). Then the coordinate `


the equation

N
N

1`

+ N

1c


(7)
, where 0 and 1 are
of this operator satisfies

1)

 [ ` ] = 0,

2003 by Chapman & Hall/CRC

Page 67

which is an analogue of a linear ordinary differential equation for a function of several variables,
since it involves the total derivative of the unknown function (exact differential equation). Its solution
has the form:

` = exp O X
N

N   ,
 M P

(8)

where the integral is taken with respect to  involved explicitly and implicitly (through the dependence of ,  ,  on  ), which means that this representation of an operator through a nonlocal
variable is most universal. The function (8) generates a nonlocal exponential operator of the form (5)
[the class of nonlocal exponential operators is specified by the first expression in (6) with 2 0].
Example 1. The equation

admits two LieBacklund operators:


X1 = ( , ,  )

X2 =

=0

=0

 (

 , 

) + (

 , 

where , , are arbitrary functions of their variables. The first operator is trivial (the operator of total derivative is
admissible for any differential equation), while the second operator determines the maximal group of contact transformations
admitted by the equation under consideration.

A LieBacklund operator admitted by an ordinary differential equation can by found by three


methods:
(i) in the form of an infinite formal series;
(ii) by passing to an equivalent system of ordinary first-order differential equations:

 =

2,

 =

3,

 ,


 =  ( , 1 , 2 ,  ,  ),

and finding an admissible point group;


(iii) by its representation as a formal operator whose coordinates depend on nonlocal variables
(the general form of the operator is chosen by the investigator).
In all cases, the search algorithm amounts to solving the determining system which is constructed by
a procedure similar to that of Subsection 0.6.1. From the standpoint of simplicity and the possibility
of integrating equations, the third method seems to be the most effective if one takes into account
that an equation admitting an operator can be written in terms of new variablesinvariants of the
admissible operatoras a new ordinary differential equation whose order is by one less than that of
the original equation.
0.6.2-3. Factorization principle.
The use of formal operators allows us to formulate universal principles for reducing the order of
an equation, independently of the specific structure of the operator (it can be a point operator, a
tangential or nonlocal operator, or a LieBacklund operator).
Theorem 1. An arbitrary B th-order differential equation (7) can be factorized to a system of
special form
H (  1) = 4  , H , H   ,  , H (  2)  ,
(9)
H = 5 ( , ,   ),
if and only if equation (7) admits the nonlocal exponential operator:
X = exp O X

5 
 N .
5   M P

(10)

2003 by Chapman & Hall/CRC

Page 68

The function 5 ( , ,   ) is the first differential invariant of the operator (10). Therefore, having
found an admissible operator (10) of the form
X=`

` exp X ( , ,   )  S ,
Q
M

N ,

(11)

we can calculate 5 by solving the first-order linear partial differential equation 5


 + p5   = 0.
The function ( , ,   ) is found as a solution of the determining system obtained by splitting
the invariance condition for operator (11):

 1)  W
X V (  )    , ,   ,  , (?

= 0,
 ( n ) =

where
X=




=0

` 

`  =c  ` 

N ( ) ,


1 ,

c  = N +  N

N
N

=` ,


+  
N

N  + ))) .


Theorem 1 generalizes the classical Lie algorithm, which is restricted to the case of unconditional
solvability of the second equation of system (9). On the other hand, the introduction of the factor
system (9) allows for two more cases, since the first equation is independent of . These cases are
the following:
1. The first equation of system (9) allows for the reduction of the order or is solvable.
2. The first equation of system (9) has some special properties, for instance, admits a fundamental
system of solutions.
Example 2. The equation

= ( ) + ( )

[ ( )]2

(12)

) and
) is the only equation of the form (its uniqueness is to within a KummerLiouville
for arbitrary functions
equivalence transformation; see Paragraph 0.2.1-8)

( ,

admitting the nonlocal exponential operator:

 +   ,  =  ( , ), = ( , ).

+
X

The second prolongation of the operator X has the form:


  + ( +  +  )  
X = exp X

+  + 2  +  + 2  + (2   + 2  +   ) + N ( )2 + 


  .


X = exp

X}

Applying this operator to the equation


invariance condition in the form:

+2

= exp

( ,

) and replacing all instances of

 +  + 2  +    

+ (2



+2

by

( ,

), we obtain the

 +   ) +  N ( )2 = 0.

Splitting this relation with respect to powers of the independent variable


equations for the functions , , and :

N = 0,
  + 2  +   = 0,
 + 2  +  + 2  +   

, we obtain the following system of three

= 0.

From the first two equations it follows that


= ( ) + ( ),

+ 2 + ( )
,
( + )2

2003 by Chapman & Hall/CRC

Page 69

where = ( ), = ( ), and = ( ) are arbitrary functions. The third equation can be treated as a first-order linear
differential equation for the unknown function = ( , ):

Substituting the above expressions of

( ,

) = ( + ) ( ) +

[ 

and

2( )2 ] ( 




1
 (

+2

 +  +  2 ).

into this relation and integrating the result, we obtain


2

[ 

3( )2 ]

i 2 )

+ 2  (
2 3 ( + )

where ( ) is an arbitrary function.


The differential invariant of the operator X satisfies the linear partial differential equation

;


(obtained after the division by exp(


equation

where

;
+ (  +  +  )

)). Substituting the above


=



2 

+ (3

( 2 )2
,
4 3 ( + )3

=0

and

into this equation, we pass to the characteristic

+  ) + + ,
( + )

. Integrating this equation, we find the differential invariant:

 + 2 i .
2 ( + ) 2 2 ( + )2
, one can find and, taking into account the known structure of the function
=

Having calculated the derivative ;


( , ), one obtains the factorization of the original equation:
; + i; 2 + ( % ); = ,
( + )  = ( + )2 ; + 2 (  )( + ) 12 2 ( 2 i ).
v , combined with the corresponding transformation of the independent
An equivalence transformation of the form + 
variable and changed notation, yields:
; + ; 2 = ( ),
(13)
= ; + ( ) 1 .
The first equation of system (13) is the Riccati equation. Its general solution can be represented in terms of a fundamental
system of solutions of the truncated linear equation:

= ( ) ,

(14)

which coincides with (12) for 0. The second equation of system (13) is a Bernoulli equation. It can be integrated by
quadrature for an arbitrary function = ( , ). Therefore, the general solution of equation (12) can be expressed in terms
of a fundamental system of solutions of the linear equation (14). Note that in the general case, equation (12) admits no point
groups.

; ;

If an operator admitted by equation (1) has no differential invariants of the first-order, then it is
possible to apply the general factorization principle.
Theorem 2. An arbitrary B th-order differential equation (1) can be factorized to the system of
special structure
H (   ) = 4   , H , H   ,  , H (   1)  ,
(15)
 ) ,
H = 5   , ,   ,  , (Z
( ) 0,

 )  is a lowerprovided that equation (7) admits a formal operator (5) for which 5  , ,   ,  , (Z
order differential invariant on the manifold given by (7). The coordinate ` of this operator satisfies
the linear equation with total derivatives:
`EN
N

+ c  [ ` ] N  + ))) + c (  ) [ ` ] N ( 

= 0.

(16)

Equation (16) plays a crucial role in both the direct and inverse problems. It can be regarded
as an equation for the determination of the coordinate of the canonical operator (if one knows the

2003 by Chapman & Hall/CRC

Page 70

invariant H ). It can also be regarded as an equation for the determination of an invariant (if one
knows the coordinate ` ). In the latter case, this is a first-order partial differential equation.
Example 3. The third-order nonlinear equation

admits two operators


X1 =

,

X2 =

( )

=0

(17)

,

(18)

which can be found with the help of the direct algorithm, if the structure of the operator is specified by the second expression
in (6). The first operator, X1 , is the usual point operator of scaling (the original equation is homogeneous) and provides the
usual reduction of order of equation (17) by one. The second operator, X2 , is nonlocal.
Let us construct differential invariants of the operator X2 . To this end, we should solve the equations:

1 + [ ] 1 = 0, = X 2 ,


2 + [ ] 2 + 2 [ ] 2 = 0.

in (19) becomes
After differentiation with respect to , the first equation

2 1 + 1 + X 2 1 = 0.

(19)

Let us show that this equation admits no solutions depending only on , , , and 1 0, i.e., there are no
first-order differential invariants. The nonlocal expression 2 depends on derivatives of arbitrarily high orders and

can be regarded as an independent quantity. Therefore, the first equation


(19) can be split and we obtain the system:

1 +  1

= 0,

= 0.

It follows that 1 = 0.
Let us find a second-order differential invariant. After differentiation with respect to
becomes

2
1

+ X 2
+
X 2

2
, we establish that
Splitting this equation with respect to the nonlocal variable
equation, the nonlocal variable is canceled,

Hence, we find that

2=;

+ 

, the second equation in (19)



2 = 0.
2 % = 0. In the remaining

= 0.

, and equation (17) is factorized to the system:

; + ; 2 = ( ),
 D; = 0.

^_

References for Subsection 0.6.2: R. L. Anderson and N. H. Ibragimov (1979), O. N. Pavlovskii and G. N. Yakovenko
(1982), N. H. Ibragimov (1985), P. J. Olver (1986), V. F. Zaitsev (2001).

0.6.3. First Integrals (Conservation Laws)

. = .   , ,  ,  , ( 
A function
differential equation

1)

.
if the total derivative of the function
if

.
c  [ ]8

 is called a first integral (conservation law) of the ordinary

 1) 
  =    , ,  , (?
( )

(1)

along the trajectories of equation (1) is zero or, equivalently,

 1)  V (  )   , ,  , (?


 1)  W = 0,
 , ,   ,  , (?

(2)

where 8 is an integrating factor. From this definition it is clear that 8 = D(n 1) .



The algorithm of finding a first integral is similar to that of finding an admissible
operator. It
is necessary to prescribe the desired structure of the first integral (or the integrating factor) and

2003 by Chapman & Hall/CRC

Page 71

substitute it into the determining equation (2). Subsequent splitting with respect to lower derivatives
(assumed to be independent variables) leads to the determining system.
x3y'z|{}~
An arbitrary function of first integrals is also a first integral of the same equation.
Therefore, having found a first integral depending on (" )  , one has to make sure that it is nontrivial;
i.e., it cannot be represented as the product of first integrals depending on lower powers of the
derivative.

x3y'z|{}~

If the equation has , functionally independent first integrals, then its order can be
reduced by , by successively excluding higher derivatives (see Example 3).
For second-order equations
   =  ( , ,   ),
(3)
the determining equation (2) can be written in the form

+  N


N

+  ( , ,   ) N



= 0.

(4)

for the given equation), as well as the


In this case, one can solve the direct problem (find
inverse problem (find possible  for the given structure of the first integral).
Example 1. Let us find all equations of the form

( ,

(5)

admitting a first integral that is quadratic with respect to the first derivative:

( , )(  )2 + ( , )  +

( , ).

Then the left-hand side of the determining equation (4) is a cubic polynomial with respect to
with respect to powers of yields the system of four equations:

The solution of this system for

( ,

where

= 0,
+
= 0,
+
+ 2
+ = 0.

^
w

. The procedure of splitting

= 0,

is given by:

3 2 (; ) + 21 2
; = 1 2 + 21 X 3 2 ,
)=

(; ), = ( ), and =

4 = ( )2 (

1 2

2

( ) are arbitrary functions. The first integral has the form:

1
4

Example 2. Consider the equation

)2

1
2

1 2

1
4

2X

(; ) ; .

Let us find its first integral, which is a cubic polynomial with respect to the first derivative:

( , )(  )3 + ( , )(  )2 +

( , )  +

( , ).

In this case, the left-hand side of the determining equation (4) is a fourth-order polynomial in
system consists of five equations:
= 0,
+ = 0,

^ 
+ w + 3 1 2
+  + 2 1 2
+ 1 2 = 0.

, and hence the determining

= 0,
= 0,

2003 by Chapman & Hall/CRC

Page 72

Solving this system, we obtain the first integral in the form:

= (  )3 6

Example 3. The equation

1 2

admits three first integrals:

4
4

 

Equating these expressions to independent constants


equation (see 4.2.1.1).

^_

3 2 + 2

+ 4

5 3

2 3

 )2 + 32 2 3 ,
3
 + 12 ,
2
1 24
3
( )2 .
1 + 2 
2

1
(
2

 1,  2, 

and eliminating

and

, we obtain a first-order

References for Subsection 0.6.3: P. J. Olver (1986), A. D. Polyanin and V. F. Zaitsev (1995).

0.6.4. Discrete-Group Method. Point Transformations


Consider transformations of the class of ordinary differential equations

  =    , ,  , ( 
( )

1)

, a ,

(1)

whose elements are uniquely defined by a vector of essential parameters a.


Any set of invertible transformations

 =  (a , ),

= (a , )

( g

  g 0),

(2)

mapping each equation of class (1) into some (other) equation of the same class

(g  ) =  a , ,  , (g 

1)

, b ,

(3)

and containing the identical transformation is called a discrete point group of transformations
admitted by the class (1). Transformation (2) maps any solution of equation (1) to a solution of
equation (3). Therefore, knowing the discrete group of transformations for some class of equations
and having a set of solvable equations of this class, one can construct new solvable cases.
Point transformations (2) can be found by a direct method namely, if one substitutes an
arbitrary transformation of the form (2) into equation (1) and imposes condition (3), one arrives
at a determining equation containing partial derivatives up to order B of the unknown functions 
and and having variable coefficients depending on  , ,   ,  , (  1) . Since the functions 
and do not depend on the derivatives, the determining equation can be split with respect to the
independent variables   ,  , (  1) , and we obtain an overdetermined system which is nonlinear,
in contrast to that obtained by the Lie method (see Subsection 0.6.1).
Example 1. For second-order equations

( , , 

, a),

(4)

the substitution of (2) into (4) yields


(

8 U

8  ) 89 8 + ( Ui Ui )(8 )3 + ( 8 Ui 8 i + 2 U 8 2U 8 )(8 )2


+ (  898 U 898 + 2 8 U 8 2 8  8 ) 8 + 8 898 8 898 = ( 8 + 8 )3

, ,

8 + U
8
8 + 
8

, a . (5)

Let us require that the transformed equation (5) belong to the class (4), i.e.,

89 8

: 8

( , , , b).

(6)

Condition (6) imposed on the determining equation (5), i.e., the replacement of 89 8 by the right-hand side of equation (6),
leads us to the relation
( 8 U 8  ) ( : , , 8 , b) + ( Ui Ui )( 8 )3 + ( 8 Ui 8 i + 2 U 8 2U 8 )( 8 )2
8 + U 8 , a , (7)
+ (  898 U 898 + 2 8 U 8 2 8  8 ) 8 + 8 898 8 898 = ( 8 + 
8 )3 , , 8
+  8

2003 by Chapman & Hall/CRC

Page 73

8 . Expanding the function into a series in powers of 8 , we can represent


4
, , [ ], [ ] ( 8 ) = 0,
(8)

which contains the independent variable


(7) in the form

=0

where the symbols [ ] and [ ] indicate dependence on the functions , and their partial derivatives involved in (7). The
sum in (8) is finite if is a polynomial with respect to the third variable [for a polynomial of degree 4, both sides of
the equation must be first multiplied by ( +
) 3 ]. Condition (8) is satisfied if the following equations hold:

8  8
4 E= 0,

= 0, 1, 2,

Example 2. Consider a special case of equation (4) with the right-hand side independent of the derivative


Relation (7) has the form:
(

8 U

8 

( ,

, a).

(9)

Ui Ui )(8 )3 + ( 8 Ui 8 i + 2 U 8 2U 8 )(8 )2


+ (  898 U 898 + 2 8 U 8 2 8  8 ) 8 + 8 898 8 898 = ( 8 + 8 )3

) ( , , b) + (

( , , a).

In this case, the sum (8) is finite and the determining system has the form:

8   8 U

It can be shown that for


Consider the case

Since

Ui Ui = 3 ( , , a),


8 U i 8 i + 2 U 8 2U 8 = 3 8 2 ( , , a),
 898 U 898 + 2 8 U 8 2 8  8 = 3 8 2  ( , , a),
8 898 8 898 + ( 8 U 8  ) (: , , b) = 8 3 ( , , a).

(10)

0, solving system (10) is equivalent to solving the original equation (9).


= 0. In this case, the first equation of the system holds identically and the system becomes

8 U i = 0,
U 898 2 8 U 8 = 0,
8 898 8 898 + 8 U

( : , , b) = 8 3

(11)
( , , a).

0, the first two equations yield

( , )=

(: ) + (: ), 8 =  [ (: )]2 .

(12)

Substituting (12) into the last equation of system (11) and splitting the resulting relation with respect to powers of the
independent variable , we obtain a new system of (ordinary) differential equations. Solving this system, we find the
and , and finally, the desired discrete group of transformations. In order to give calculation details,
unknown functions
one has to know the specific structure of the function ( , ), for in the general case it was only shown that any discrete
point group of transformations of equation (9) for = 0 consists of KummerLiouville transformations (12).

Example 3. Consider the generalized EmdenFowler equation:

E
Here, a = { 5 , , } is the vector of essential parameters, and


by scaling the independent variable and the unknown function).

(  ) .

(13)

is an unessential parameter (it can be made equal to unity

1 . First, we note that equation (13) admits a discrete group of transformations determined by the hodograph transformation,
i.e., by passing to the inverse function:

= ,

= ,

where

= (: ).

(14)

,
,
This transformation is a consequence of the invariance of equation (13) with respect to the transformation
3 ,
(note that the hodograph transformation changes the sign of the unessential parameter ). Denoting
the transformation (14) by , let us schematically represent its action on the parameters of the equation as follows:

$ { , 5 , 3 }
5 , }
yields the original equation.
Double application of the transformation
{ ,

transformation

(15)

2003 by Chapman & Hall/CRC

Page 74

2 . For

= 0, equation (13) is of the class (8), and the last equation of system (11) becomes

8 h8 + 2 : i^ =  2 ( + ) E ,
(16)
where , , and are the parameters of the transformed equation 89 8 = : , and
(: ) =  X [ (: )]2 : .

Let , 0, 1, 2. Then relation (16) is possible only if ( : ) 0. Splitting with respect to powers of leads us to
the system:
 89 8 2( 8 )2 = 0,
(17)
: =  2 +3 E .
By integration we find that = : 1 , = : 1 (to within unessential coefficients). Thus, we arrive at the transformation
= : 1 , = : 1 , where = (: ).
(18)
Denoting the transformation (18) by , let us schematically represent its action on the parameters of the equation:
{5 3, , 0}
transformation .
(19)
{ 5 , , 0} ,

Double application of the transformation yields the original equation.
3 O . Let = 0 and = 2. Then, = 2 and the splitting procedure for equation (16) yields the system of three equations:
 89 8 2( 8 )2 = 2 2 6 t E ,
h 89 8 2 8 h 8 =  2 5 2 E ,
: =  2 5 E .
[

89 8 2(8 )2 ] + h89 8

Its solution gives us the transformation

= ,
{ , 2, 0}

= :  + : T
,
 { , 2, 0}

where we use the notation:

= (8

+ 40 + 49)1 2 ,

1
,
2

transformation of the variables, = ( );


transformation of the vector of essential parameters;

Example 4. Likewise, for the class of equations

= ( ) ( ) (


we find two transformations:

^_

: { , , }
: { ,

, 1}

0
,


% )}
(: ), , 1}

{ , , (
{:

1
[ (2 + 5) 5],
2

)3 (1

= ( + 2),

( + 2)( + 3)

transformation of the variables; see (14);


transformation of the variables; see (18).

References for Subsection 0.6.4: V. F. Zaitsev and A. D. Polyanin (1994), A. D. Polyanin and V. F. Zaitsev (1995).

0.6.5. Discrete-Group Method. The Method of RF-Pairs


The direct method (see Subsection 0.6.4) is unsuitable for finding nonpoint transformations of
second-order equations (i.e., transformations containing derivatives), since the determining equation
cannot be split into equations forming an overdetermined system. Therefore, instead of searching
for Backlund transformations in the form of arbitrary functions  =  (a , , g ), = (a , , g ), one
uses the superposition of some standard transformation containing the derivative and a point transformation which can be found by the direct method. The standard dependence on the derivative
can be introduced by means of an RF-pair, which amounts to a transformation of successively increasing and decreasing the order of the equation (this transformation is not equivalent to the identity
transformation). An additional point-transformation is necessary, since the equation obtained by an
RF-pair is usually outside the original class.

2003 by Chapman & Hall/CRC

Page 75

1 b . Suppose that any equation of the original class can be solved for the independent variable  :

 ( ,   ,    ) =  .
Termwise differentiation of this equation with respect to  yields the following autonomous equation:



  + N      + N        = 1,


N
N
N
"
whose order can be reduced with the substitution  = H ( ). This pair of transformations is called
N

a first RF-pair.

2 b . Suppose that any equation of the original class can be solved for the dependent variable :

 ( ,   ,    ) = .
Then, termwise differentiation of this equation with respect to  brings us to the following equation
which does not explicitly contain :

N
N

+ N




    + N  

N

     =   .

The order of this equation can be reduced by means of the substitution  = H ( ). This pair of
transformations is called a second RF-pair.
Example 1. Consider transformations of the class of generalized EmdenFowler equations:

).

(1)

This class will be briefly denoted by the vector of essential parameters { 5 , , }. Application of the first RF-pair transforms
this equation to
;N  = ( 1); 1 (; )2 + 1 ; + 5 1 ; E 1 (; ) E 1 .
(2)
E
E
E
E
Now we have to find a point transformation that maps class (2) into class (1) (with another vector of parameters):
89 8 = :D^ ( ) .
(3)

Note that in this case, the desired transformation does not map the given class into itself as in Subsection 0.6.4, but is a
(1). Nevertheless, the method for finding transformations
mapping of the equations classes (2)

= ( , ),

= (, )

8 U  8

0)

is completely the same and involves solving the determining equation:

8  ) :i^ (8 ) + ( Ui Ui )(8 )3 + ( 8 Ui 8 i + 2 U 8 2U 8 )(8 )2


+ ( 
898 U 898 + 2 8 U 8 2 8  8 ) 8 + 8 898 8 898 = 1 ( 8 + 
8 )( 8 + U
8 )2
8
1
1
2 +1
1
E ( 8 + 8 ) E ( 8 + U
8 ) E .
+
( + 8 )2 ( 8 + U
8 ) + 5
(4)

E E E
E
E
8 U 0 and consider transformations
Following the procedure set out in Subsection 0.6.4, we omit the general case 8 
for which at least one of the above partial derivatives is zero.
1 O . Case  = 0, 8 = 0. Equation (4) has the form
8 U
:i^ (8 ) + 8 Ui (8 )2 U 898 8 = 1 8 (U )2 ( 8 )2

1
1
2 +1
1
1
E ( 8 ) E (U ) E ( 8 ) E .
+ ( 8 )2 U 8 + 5

E E E
E
E
E
and for 5 0, 1, 1, 2 can easily be solved by splitting,
=: , = .
(

8 U

1
+1

1
2

2003 by Chapman & Hall/CRC

Page 76

As a result, using an RF-pair, we obtain:

8 ) 1 , = : 1+1 , = 21
transformation of the variables, = ( : );
E

1
5

1
 + 1 , 2 , 5 transformation of the vector of essential parameters.
{ 5 , , } 
2 O . Case 8 = 0, U = 0. Similar calculations bring us to the formulas:
= ( 8 ) 1 , = 1+1 ,  = : 21
transformation of the variables, = ( : );

E 
2 +1
1
5

1 , 5 + 1 , transformation of the vector of essential parameters.


{5 , , }

=(

(5)

(6)

Transformation (6) can be obtained by successive application of transformation (5) and the hodograph transformation
(see Subsection 0.6.4).
The inverse transformations have a similar structure. For instance, the inverse of transformation (5) can be written (after
changing notation) as follows:

8 1 ,

where = ( : ).
E
Denoting the transformation (7) by , let us schematically represent its action on the parameters of the equation:

 1 1 , 5 5 + 1 , 2 + 1 transformation .
{ 5 , , } 
Applying the transformation three times, we obtain the original equation.
=

1
+1

= ( )

: 11 ,

(7)

(8)

It can be shown that all transformations which can be found from equation (4), without additional restrictions on the
and
parameters of the original and the transformed equations, are obtained by superposition of the transformations
(see Subsection 0.6.4, Example 3), which form a group of order 6. The parameters of these equations are given in Figure 1.
Example 2. Suppose that

= 0 in equation (1). Then, on the class of EmdenFowler equations

(briefly denoted by { ,

, 0} ),

(9)

(see Subsection 0.6.4, Example 3). Therefore, in this case, the group considered in the
one can define the transformation
previous example is prolonged to a group of order 12 (see Figure 2).
This prolongation takes place each time the third component of the parameter vector becomes equal to zero. This
happens, for instance, if = 1 in equation (9). In this case, the order of the group is equal to 24 (see Figure 3).

Example 3. The class of second-order equations

= ( ) ( ) (  )

(10)

admits a discrete group of transformations similar to that for the generalized EmdenFowler equation. Most simply, this
group can be obtained by inverting the transformation (6). Thus, we seek the parameters of the transformation as functions
of a single variable,
= ( ), = ( ),  = ( ).

Introducing a point generator

(see Subsection 0.6.4), we find a discrete group of transformations relating the equations
shown in Figure 4. The functions 1 ( 1 ), 1 ( 1 ), 1 ( 1 ) determine the original equation, while the corresponding functions
( ) with = 2, 3, are determined by the parametric formulas:
for the transformed equations, ( ), ( ),

2 (

2)

2 (

2)

2)

2 (
and

1,

1 (

1
[ 1 ( 1 )]3

3 (

3)

1 (

3 (

3)

3)

3 (

1)

1)

1
1

1,
1

1
1 ( 1) ,

X 1 (

1 (

X 1 (

1 (

1)

1 ( 
1

(11)

1)

1)

1)

1,

(12)

1 ),

where
= , = 1, 2, 3.
The above example allows us to eliminate singular points of the group of transformations defined by (7) for = 1,
= 1, = 1, 2. For these values of the parameters, the form (10) and the transformations (11), (12) should be used.

^_

References for Subsection 0.6.5: V. F. Zaitsev and A. D. Polyanin (1994), A. D. Polyanin and V. F. Zaitsev (1995).

2003 by Chapman & Hall/CRC

Page 77

Figure 1

Figure 2

2003 by Chapman & Hall/CRC

Page 78

Figure 3

Figure 4

2003 by Chapman & Hall/CRC

Page 79

Chapter 1

First-Order Differential Equations


1.1. Simplest Equations with Arbitrary Functions
Integrable in Closed Form*
1.1.1. Equations of the Form $ = ( )
Solution:**

= X ( )  + L .

1.1.2. Equations of the Form

$ = ( )

Solution:  = X

+L .
M
 ( )
= #  , where #  are roots of the algebraic (transcendental) equation
Particular solutions:
 ( #  ) = 0.

1.1.3. Separable Equations

$ = ( ) ( )

= Xv ( )  + L .
M
( )
M
= #  , where #  are roots of the algebraic (transcendental) equation
Particular solutions:
( #  ) = 0.
x3y'z|{}~/
The equation of the form  1 ( ) 1 ( )  =  2 ( ) 2 ( ) is reduced to the form 1.1.3 by
dividing both sides by  1 1 .
Solution: X

1.1.4. Linear Equation


Solution:

=L  + X

( )$ = 1 ( ) + 0 ( )
 ( )
 ,
 0
( ) M

1.1.5. Bernoulli Equation

where

 ( ) = X

( )$ = 1 ( ) + ( )

Here, B is an arbitrary number. The substitution ( ) =


2 = (1 B )  ( ) + (1 B )  ( ).
1

Solution:
1

 1 ( )
 .
( ) M

 = L  + (1 B )  XZ   ( )  ,
( ) M

where

 leads to a linear equation:

 ( ) = (1 B ) X

 1 ( )
 .
( ) M

* Special cases of equations 1.1.11.1.5 for specific functions  ,  0 ,  1 ,   , and are not discussed
in this book; such cases can readily be recognized by the appearance of equations investigated, and
the solution can be obtained using the general formulas given in Section 1.1.
** Hereinafter we shall often use the term solution to mean general solution.

2003 by Chapman & Hall/CRC

Page 81

1.1.6. Homogeneous Equation

$ = (w )

The substitution ( ) = 
 leads to a separable equation:   =  ( ) .
Solution: X

= ln | | + L .

 ( )

= #   , where #  are roots of the algebraic (transcendental) equation

Particular solutions:
#   ( #  ) = 0.

1.2. Riccati Equation ( ) = 2 ( )

+ 1( ) + 0( )

1.2.1. Preliminary Remarks


For  2 0, we obtain a linear equation (see Subsection 1.1.4); and for  0 0, we have a Bernoulli
equation (see Subsection 1.1.5 with B = 2), whose solutions were given previously. Below we
discuss equations with  0  2 1 0.
1 b . Given a particular solution
as:
=

0 (  ) + ` (  ) L X` (  )

0(

 ) of the Riccati equation, the general solution can be written

 2 ( )

( ) M


, where ` ( ) = exp X V 2  2 ( ) 0 ( ) +  1 ( )W M
.
( )

To the particular solution 0 ( ) there corresponds L = F .


Often only particular solutions will be given for the specific equations presented below in
Subsections 1.2.21.2.8. The general solutions of these equations can be obtained by the above
formulas.
2 b . The substitution

( ) = exp O X


M P

reduces the Riccati equation to a second-order linear equation:

 2 2    + V  2   (  2 )   1  2 W  +  0  22 = 0.
The latter often may be easier to solve than the original Riccati equation. Specific second-order
linear equations are outlined in Section 2.1.

1.2.2. Equations Containing Power Functions


1.2.2-1. Equations of the form ( ) =  2 ( )
1.

+  0 ( ).

+ .

For  = 0, we have a separable equation of the form 1.1.2. For  0, the substitution 'a = ' + (
leads to an equation of the form 1.2.2.4: g  =  2 + 'a .
2.

+ 3 .

Particular solution:
3.

=  

+ .

This is a special case of equation 1.2.2.27 with = 0 and = 0.

2003 by Chapman & Hall/CRC

Page 82

1.2. RICCATI EQUATION ( )

4.

2 ( )

+ 1 ( ) +

0 (

Special Riccati equation, B is an arbitrary number.


1 2
= , where ( ) =  L 1
Solution:

1
2

 " 
P

+L

1
2

 "  S ,
P

, = 12 (B + 2); + ( H ) and n + ( H ) are the Bessel functions, B 2. For the case B = 2, see

equation 1.2.2.13.
5.

Particular solution:
6.

=   .
1

.
1

 
B +1
B
(
.
=   leads to an equation of the form 1.2.2.38:   +  2 +
= ' +
B +1
B +1

For the case B = 1, see equation 1.2.2.13. For B 1, the transformation =

7.

+ L , where =
M
2
+ + 1

Solution:   ln  = X
8.

B +1
B +1
2 b . For B = 1 and

+ 
 +1 .

 


+1

B +1
, =
.
 

leads to a Riccati equation of the form 1.2.2.4:

1, the transformation

=  +

+1

= 1


leads to a Riccati

2+
1
.
+1
+1

= 1, the original equation is a separable equation. In this case we have the

solution: ln | | = X
9.

+1

equation of the form 1.2.2.4:  =


3 b . For B =

1 b . For B 1, the substitution =  

 =

+1

+ (

M2


+ ) (

+

+L .
+ )

1
 + 
, = [( ( + )2 + ( ( ( + )], where =  ( , leads
( +

M
M
M
M
= 2 + , 2  .
to an equation of the form 1.2.2.4:
The transformation =

10.

Particular solution:
11.
12.

13.

= (

 + .
0 = '

+2

+ .

 = (
The substitution = 1
 leads to an equation of the form 1.2.2.6: I

+  2  + ' 

( 2 + 2 )( + 2 ) + 0 + 0 = 0.
The substitution = 
leads to a second-order linear equation of the form 2.1.2.108:
(  2  +  2 )   + (  0  +  0 ) = 0.

Solution:

+ .

+ +  = 0.

 2  O

 
 2  + L
2  + 1
P

, where is a root of the quadratic equation

2003 by Chapman & Hall/CRC

Page 83

14.

+ (1 2 )

Particular solution:
15.

=  + '

( + 1).

+ .

The substitution = 

+ # , where # is a root of the quadratic equation !#


leads to an equation of the form 1.2.2.35:  u =  2 + (1 2 !# ) + '  .
16.

2 (

+ ) + 41 (1

The transformation = ' + + ( ,

form 1.2.2.4: 3  =
17.

( 2

18.

+ 2

+ 2 )(

)  .
2

)+

).

2 (

2

 +

leads to an equation of the

= 0.

2.
1

= +

4 2

1)2 (

tan O
)+

+L
2

.
+ s = 0.

The substitution = 
leads to a second-order linear equation of the form 2.1.2.218:
 2 ( 1)2   + ( ' 2 + ( + ) = 0.

20.

21.

+1

+ )2 ( + 2 ) +

= 0.

The substitution = 
leads to a second-order linear equation of the form 2.1.2.234:
(  2 + ' + ( )2   + #* = 0.
=

+ .

The substitution =  

+ # , where # is a root of the quadratic equation !


 #
leads to an equation of the form 1.2.2.35:   =  2 + (B 2 !# ) + ( + .
22.

+ ) =

B# + = 0,

0 = 1  .

Particular solution:
23.

# + ( = 0,

The substitution = 
leads to a second-order linear equation of the form 2.1.2.179:
( ( 2  2 +  2  +  2 )   +  0 = 0.

Solution:
19.

+  (

+ )( 2 ) +

1)w

B/  1 +   +

0 =
  + ' + + (

Particular solution:

2
1

1)

= 0.

1.2.2-2. Other equations.


24.

+ .

2

=
2   =  2  ( ( + , )  .

The substitution

25.

Particular solution:

leads to a second-order linear equation of the form 2.1.2.12:

= 1
 .

2003 by Chapman & Hall/CRC

Page 84

1.2. RICCATI EQUATION ( )

26.

27.

28.

+ (

= ( + 1)w

0
2

35.

36.
37.

38.

39.
40.

+ .

=.

+
=  

+
0

+1
1

= ( .

0 (

85

  + +( .
0 = '

+ )

   +  )1 .
0 = (
+
0

+ ) +

 1 ^ +

= (  .

= 2 + + 2  .
The transformation a =  , = 

Particular solution:
34.

+ 1 ( ) +

2.

Particular solution:
33.

Particular solution:
32.

The substitution = 
leads to a second-order linear equation of the form 2.1.2.31:
   (  + )  + (  2 + ' + ( ) = 0.

Particular solution:
31.

2 ( )

+  ) +

Particular solution:
30.

The substitution = 
leads to a second-order linear equation of the form 2.1.2.45:
     + '   1 = 0.

Particular solution:
29.

+2

g = 
leads to a separable equation:  3

+( .

= 2 + + .
The transformation =  , =   leads to the special Riccati equation of the form 1.2.2.4:

(
B
 = 2 + + , where = 2.

= 2 + ( + ) + 2 .
 =  
The substitution =   leads to a separable equation: 3

= 2 + + .
The substitution = 

     + '  = 0.

(

+  + ( ).

leads to a second-order linear equation of the form 2.1.2.67:

+ 3 2 + 2 + 1 + 0 = 0.
The substitution  3 = 
leads to a second-order linear equation of the form 2.1.2.64:
   +  2  +  3 (  1  +  0 ) = 0.
= 2 + + .
 = 
The substitution =   leads to a separable equation:  3

= 2 + x 2 +2 .
Particular solution: 0 = ' + .

+ (  + B ) + ( .

2003 by Chapman & Hall/CRC

Page 85

41.

= +

Solution:
42.

43.

) +

+ (

  ++

 tan O

+ (

B +

+L

.
.

The transformation =    , =   leads to a special Riccati equation of the form 1.2.2.4:


B 2
.
(B +  )  =  2 + (  , where , =
B +
2

) + .

For B = 0, this is a separable equation. For B 0, the solution is:

BIX
44.

2 +

+ (

=  +L ,
M
+ +(

) +

1 )

where

=   .

 =   + +
The substitution =   leads to a separable equation: 3

45.

46.

2 )(

) + (

(

+  + ( ).

= 0.

The substitution = 
leads to a second-order linear equation of the form 2.1.2.108:
(  2  +  2 )   + (  1  +  1 )  + (  0  +  0 ) = 0.

+ ) =

 (

)2 +  ( +

+ .

The substitution a =  + ' leads to a first-order linear equation with respect to  =  (a ):


  +( .
( ha 2 + a + =/ + ( ) g = 

47.

= 2

= 2

+ 3

Particular solution:
49.
50.

51.

52.

53.

The substitution = 
+ (

 .
0 = 

Particular solution:
48.

.
=   12  .

+ .

 = 
leads to a separable equation:  p

+ s.

+

+ (  + 1) + ( .

+ .

The substitution ( = 
leads to a second-order linear equation of the form 2.1.2.139:
 2    (  +  )  + ( (  2 +  + = ) = 0.
+

The substitution  = 
leads to a second-order linear equation of the form 2.1.2.132:
 2   '  +  ( (  + ) = 0.
+ s .

The substitution  = 
leads to a second-order linear equation of the form 2.1.2.133:
 2   '  +   ( (  + ) = 0.
+ (

+ )

+

+

+ .
The substitution ( = 
leads to a second-order linear equation of the form 2.1.2.146:
 2   (   +  )
  + ( (  2  +   + = ) = 0.

2003 by Chapman & Hall/CRC

Page 86

1.2. RICCATI EQUATION ( )

54.

55.

= (

+

+ )

+ (

2 ( )

+ )

+ 1 ( ) +

0 (

87

 = ( 2
The substitution = 1
 leads to an equation of the form 1.2.2.53:  2 2

(   +  ) +  2  +   + = .
(

1)

+ (

+ 1) = 0.

2 1

The substitution

 +
2

(

leads to an equation of the same form:

( )

1)  + ( 1)(

2 + 1) = 0.

If = B is a positive integer, then by using the above substitution, the original equation can
be reduced to an equation of the same form in which = 1, i.e., to an equation of the form
1.2.2.58 with  = 1,  = 1.
56.

+ ) +



Particular solution:
57.

+ )



 +
 .

+ = 0.

+

+

+  ) = 0.

1
 +

leads to an equation of the same form: ( 

( )
 +
 2 + (2  + ) + = 0.
O =

P
=

The substitution

58.

+ ) +

Solution:
59.

=  + OX
+ ) =



+ ) =

Particular solution:

61.

62.

+L

+

+ ) + ( )

=  + # , where # =

+ (

+  )  +

= 0.

+ (2

+
)

1
2

+
.

 A

 2 4 4 (  .

+ (
) +

=  .

+ 2 ) =

Particular solution:

M2

Particular solutions:
60.

+ (1 )

+ (

1 )

( +

1 )

2)

+ (

1 )

2.

=  .

+ 2 ) =

+ (

+ 0.

Let and be roots of the system of the quadratic equations

+ (

 2) + 

= 0,

+ h 1 + (

(

= 0,

where the first equation is solved independently (in the general case there are four roots). If
some roots satisfy the condition 2 + " 1 +  1 +  0 " 2 = 0, the original equation possesses
a particular solution: 0 =  + .

2003 by Chapman & Hall/CRC

Page 87

88
63.

FIRST-ORDER DIFFERENTIAL EQUATIONS


( )( ) +

+ ( +

)( + ) = 0.

To the case , = 0 there corresponds a separable equation. To , = 1 there corresponds a linear


equation. For , 1 and , 0, with the aid of the substitution ,! ( ) = + , ( +  ), we obtain
the general solution:

+ )( +

64.

65.

66.

67.

+, ( +  )   
O
=L
+ , ( +   )  hP
1
1
+
=L
+, ( +  )  

+ 2 )( +

69.

70.

71.

72.

73.

) + (

+ 1 ) +

if  =  .

= 0.

The substitution = 
leads to a second-order linear equation of the form 2.1.2.179:
( ( 2  2 +  2  +  2 )    + (  1  +  1 )  +  0 = 0.
+ (

+ ) + s .

The substitution  = 
leads to a second-order linear equation of the form 2.1.2.183:
 3   ( ' 2 + ( )  + !  = 0.

+ (

+ ) + 

+ .

The substitution  = 
leads to a second-order linear equation of the form 2.1.2.186:
 3    ( ' + ( )  +  (  + ) = 0.
2

The substitution
 ( 2 +  )   + ( '
68.

if   ,

2(

+ )( +

= 
+ ( )

) + (

+ ) + s = 0.

leads to a second-order linear equation of the form 2.1.2.190:


 +  = 0.

) + (

+ ) + 

+

= 0.

The substitution = 
leads to a second-order linear equation of the form 2.1.2.194:
 2 ( +  )    +  ( ' + ( )  + (  + ) = 0.

+ )( ) 2 +

Solution: ln
= L + 2 X 
+

2(

+ )( +

1)2 ( +

) + (

= 0.

+ ' + (

+ ) + s = 0.

The substitution = 
leads to a second-order linear equation of the form 2.1.2.219:
 2 ( 2 +  )   +  ( ' 2 + ( )  +  = 0.

)+

1) +

+ s = 0.

The substitution = 
leads to a second-order linear equation of the form 2.1.2.227:
 ( 2 1)2   + ' ( 2 1)  + ( ( 2 +  + ) = 0.
M

+1

+ .

 # 2 (  + B ) # + = 0,
The substitution =   + # , where # is a root of the quadratic equation !
M
u

2
+ (B +  2 !# ) + ( + .
leads to an equation of the form 1.2.2.35:   = 

( 

+ ) =

The transformation a =  
, ( H +  ).

+ (

 )

.
, H =   leads to a separable equation: [a 2 + ( + 'B )a + = ]H g  =

2003 by Chapman & Hall/CRC

Page 88

1.2. RICCATI EQUATION ( )

74.

75.
76.

2 ( )

+ 1 ( ) +

0 (

89

2 (

1)( + 2 ) + ( + ) + + s = 0.
The substitution = 
leads to a second-order linear equation of the form 2.1.2.254:
 2 (   1)   + (  +  )  + (&  + ) = 0.

( + + ) = ^ 2
Particular solution: 0 = 1
 .
( + + )
Particular solution:

( + + ) =  
Particular solution: 0 = .

78.

+ )(

= .

77.

+

) + s

+ .



 (

Particular solutions: 0 = A" .



+ 

) = 0.

1.2.3. Equations Containing Exponential Functions


1.2.3-1. Equations with exponential functions.
1.



The substitution a = 
2.

 

Particular solution:
3.

4.

5.



+ 'a .

=   .
2

+ .

The substitution . = 
leads to a second-order linear equation of the form 2.1.3.10:

    + . (   + ( ) = 0.

+ ( )





= 2 +  2 ( 

1.2.2.4: 3  =

 8

+ 
+

 

2 2

+ (2 )2 ( 

=.

+ )

 .

 6
2

1
4

+ ,

 4

The transformation =  2  , = 

2  =

=   .

The transformation =  

8.

+ 



Particular solution:
7.

 

2 2

The substitution . = 
leads to a second-order linear equation of the form 2.1.3.5:


  + . (  +   + (  2  ) = 0.

Particular solution:
6.

leads to an equation of the form 1.2.2.35: a g = 

+ ' + ( ).

O  

 


P

leads to an equation of the form

 O

1
leads to an equation of the form 1.2.2.3:
2P

2003 by Chapman & Hall/CRC

Page 89

90
9.

10.

FIRST-ORDER DIFFERENTIAL EQUATIONS

 

 s



+ 

 





14.

= 

15.

 

16.

17.

18.

19.

=   

 (2 + )

+ ( + )H + ( .

 (  )

=   .


 

( )

=   .

 s



=    .

+ [ 

+ 

 

 

( + )


The substitution = 

 

] + 

 =  ( 
leads to a separable equation: 3
+ 

 (2

+1)

  (

+ )

g  = a 2

 



( 

 

( 

+  

)2 +
0

= .



+ )( ) + 

Particular solution:

+ 'a

+ 'a

=   .

+ ) =
0



+  + ( ).


The substitution a =   leads to an equation of the form 1.2.2.52: ,!a 2 g = a 2
(a  +1 + a 2(  +1) .
M

Particular solution:

21.


, a
The substitution a =   leads to an equation of the form 1.2.2.51: !
(a (  + ) J' + .
M

Particular solution:
20.

  

 ( + )

Particular solution:

.

2 ( +2 )

Particular solution:

=   .

+ 

+ 'a

leads to a separable equation: H   = !H

Particular solution:

 ( +2 )


 

 

=   .

Particular solution:
13.

The substitution H = 
12.

0.


The substitution a =   leads to an equation of the form 1.2.2.4: , g = 
Particular solution:

11.




   + '  


0 =
   +   +(

I


2  

= 0.

1.2.3-2. Equations with power and exponential functions.


22.



Particular solution:

+ 
0

= 1
 .

2003 by Chapman & Hall/CRC

Page 90

1.2. RICCATI EQUATION ( )

23.

 

Particular solution:
25.



Particular solution:
26.

= 



29.

= ( + 1)

34.

 

Solution:
36.

2  
2

 2

+ 2 

=   .



=   .
=  

+ ) +

=  

 



2 2  
1



) +

 =     ( 
leads to a separable equation: 3

+ (

+( .

 

 

2 2

=  

+  + ( ).

 +L

) + 

M
= X 
+ +(

Particular solution:

   +( .
0 = '



= '   tan O  X  

Solution: X
37.

)2 +

Particular solution:
35.

=   .


The substitution = 

  (

 2

 +1  

2
+ (  

=  2
=

= '  .

= 2 (  

91

=   .



Particular solution:
33.

Particular solution:
32.

0 (

=   .

Particular solution:
31.

Particular solution:
30.

 



Particular solution:

 2

  .
0 = '





+ 

Particular solution:
28.

+ 1 ( ) +

Particular solution:
27.



 

2 ( )

H

=  + L , where H =  .
!H + H + 

Solution: X
24.

 

.
.

 + L , where =   .

2003 by Chapman & Hall/CRC

Page 91

92
38.

FIRST-ORDER DIFFERENTIAL EQUATIONS

 
2



2
2
=    J 2 tan O /X    J 2  + L .
M
P

Solution:
40.

2
2
=    J 2 tan O  Z
X   J 2  + L .
M
P

Solution:
39.

2 ) = + exp(  ) + exp(2  ).
= 1
 , =  2  leads to a Riccati equation of the form 1.2.3.3:
The transformations
 = 2 +  +    + (  2  .

4 (

1.2.4. Equations Containing Hyperbolic Functions


1.2.4-1. Equations with hyperbolic sine and cosine.
1.

Particular solution:
2.
3.

= 2 + sinh(
Particular solution:

=  cosh(  ).

+ sinh(
=  .

sinh(

= [ sinh2 (

0
2

sinh3 (
0

= coth(  ).

6.

[ sinh( ) + ] = 2 + sinh(

Particular solution: 0 = .

7.

[ sinh(

9.
10.



)+

.
2

sinh(

).

+  + cosh .

The transformation  = 2a , = 
leads to the modified Mathieu equation 2.1.4.9:
g  g (  2  cosh 2a ) = 0, where  = 4 ,  = 2= .

= 2 + cosh(
Particular solution:

= +
2

+ cosh(
=  .

cosh ( )

Particular solution:
11.

).

sinh( ) = 0.
 cosh(  )
=
.
 sinh(  ) + 

) + ]( 2 ) +

Particular solution:

).

sinh2 (

).

) 2.

= cosh(  ).

) ]

Particular solution:

8.

+ sinh (
= 1
 .

sinh2 (

= 2 + sinh ( )

Particular solution:
5.

Particular solution:
4.

+  sinh(

= [ cosh2 (

+ cosh (
= 1
 .

) ]

Particular solution:

).

cosh2 (

).

= tanh(  ).

2003 by Chapman & Hall/CRC

Page 92

1.2. RICCATI EQUATION ( )

12.

2 = [

Particular solution:
13.

)]

cosh(
cosh (

) sinh

2 ( )

+ 1 ( ) +

cosh(

 .

1
2

= tanh 

+ +

0 (

93

).

).

1
The transformation = coth(  ), = sinh2 (  ) sinh(  ) cosh(  ) leads to an equation

of the form 1.2.2.4: 3  =


14.

sinh(

 .

) cosh (

+ sinh(

The transformation = cosh(  ),

2  =

15.

+  

cosh(

 .

The transformation = sinh(  ),

2  =

16.

+  

[ cosh(

Particular solution:
17.

[ cosh(

leads to an equation of the form 1.2.2.4:

).

leads to an equation of the form 1.2.2.4:

cosh(

).

= .

) + ](

Particular solution:

+ cosh(

).

 .

) + ]

) sinh (

+ cosh(

cosh( ) = 0.
 sinh(  )
=
.
 cosh(  ) + 
2

)+

1.2.4-2. Equations with hyperbolic tangent and cotangent.


18.

Particular solution:
19.

+ 3 

21.

[ tanh(

Particular solution:
22.

+ tanh(

coth (

Particular solution:

).

tanh(

).

= .

).

=  coth(  ).

+ 3  ( + ) coth2 (

Particular solution:
24.

= 1
 .

( + ) coth2 (

Particular solution:
23.

+ tanh (

).

=  tanh(  ) coth(  ).

tanh (

) + ]

( + ) tanh2 (
0

Particular solution:

).

=  tanh(  ).

Particular solution:
20.

( + ) tanh2 (

).

=  coth(  ) tanh(  ).

= 1
 .

coth (

).

2003 by Chapman & Hall/CRC

Page 93

94

FIRST-ORDER DIFFERENTIAL EQUATIONS

25.

[ coth( ) + ] = 2 + coth(

Particular solution: 0 = .

26.

Particular solution:
27.

tanh2 (

)2

coth2 (

coth(

).

).

= tanh(  ) + coth(  ).

2 ( + ) tanh2 (

Particular solution:

) ( + ) coth2 (

).

=  tanh(  ) +  coth(  ).

1.2.5. Equations Containing Logarithmic Functions


1.2.5-1. Equations of the form ( ) =  2 ( )
1.

= (ln )

Particular solution:
2.
3.

4.

(ln ) .

= ' + .

+ 'a + ( .

= 2 2 ln2 ( ) + .
Particular solution: 0 =  ln( ).

= B 1:

ln2  (

)+

ln 


0 =  ln (  ).

(

).

1
2

leads to an equation of the form 1.2.2.3:

1
leads to an equation of the form 1.2.2.4:
2

= 2 + 2 ln2 .
Particular solution: 0 =  ln  .

= 2 2 + ln2 + ln + .
The transformation = ln  , = 
2  = 2 +  2 + ' + ( 1 .
4

+ ( ln + ) + 14 .

The transformation =  ln  + ( ,

2  =

)(

ln(

+  

 .




2 ) = 1.

Particular solution:

= [ ln( )]1 .

1.2.5-2. Equations of the form ( ) =  2 ( )


10.

= 2 + ln  + ln2  +2 .
The substitution a = ln  leads to an equation of the form 1.2.2.6 with ,
g  =  2 + 'a  + (a 2  +2 .

8.

9.

Particular solution:

7.

+  0 ( ).

= 2 + ln + .
g
The substitution  =  leads to an equation of the form 1.2.2.1: g = 

5.
6.

ln(

Particular solution:

ln(
0

+  1 ( )  +  0 ( ).

) 2.

=.

2003 by Chapman & Hall/CRC

Page 94

1.2. RICCATI EQUATION ( )

11.
12.

ln (

Particular solution:

+1

ln

= ' ln  .

= ( + 1)w

Particular solution:
14.

= (ln )

= (ln )  (

= (ln )

18.

ln (

Solution:
19.

20.

Solution: X
21.

22.
23.

M2

+ ln + .

(ln )

2 (ln ) .

+ L , where H = 

!H + 

+ ln

95

= ( .

0 (

(ln ) .

2 2

= '  tan  /X 

= (

Solution:

+1

   +( .
0 = '

= ( + ln )2 .

+ 1 ( ) +

).

)2 +

+ (ln )

Solution: ln  = X

= ' ln  .

Particular solution:
17.

(ln )

Particular solution:
16.

(ln )

+1

=  

2 ( )

+ ln + .

Particular solution:
15.

ln (

0 = 1  .

Particular solution:
13.

ln (

+  ln  .
).

ln (  )  + L2S .

) .
2


1
+   =L .
+  ln 
B

(ln )

+ (

ln ) + ln .

= X 
M
2+ +(

ln 

 + L , where =   .

= 2 2 2 + 2 ln .
The substitution  2 = 
leads to a second-order linear equation of the form 2.1.5.27:
 2   +   + (   )2 ln * = 0.

( ln + ) = 2 + (ln )
Particular solution: 0 = .

( ln + ) = (ln ) 2 +
Particular solution: 0 = .

(ln ) .

(ln ) +

1.2.6. Equations Containing Trigonometric Functions


1.2.6-1. Equations with sine.
1.

=  2 +  + sin( ).
The substitution 2a = 2  + leads to an equation of the form 1.2.6.14: g =

+ + = cos a .

2003 by Chapman & Hall/CRC

Page 95

96
2.

FIRST-ORDER DIFFERENTIAL EQUATIONS

+  sin(

Particular solution:
3.

+ sin (

)+

sin2 (

).

=  cos(  ).
+ ) sin

+ ).

sin(  +  )
sin2 (  +  )
, =
+ cot(  +  )S leads to an equation
The transformation =
sin(  +  )
sin(   ) Q

3


of the form 1.2.2.4: = 2 + #*  , where # = ( [ sin (   )]2 .


4.

sin(

Particular solution:
5.

sin(

=[ +

)]

sin2 (

Particular solution:
9.

= ( + 1)

Particular solution:
10.

sin  (

11.

sin (

[ sin(

) + ] =

Particular solution:
13.

[ sin(

).

1
2

=  

2
0

).

(sin ) .

.
)2 +

2

sin (

   +( .
0 = '

).

.

= cot(  ).

+
1

+ sin(

= .

).

sin + (  )  + L2S .
M

sin(

).

sin( ) = 0.
 cos(  )
=
.
 sin(  ) + 

) + ]( 2 ) 

Particular solution:

 +

1
4

+ sin2 (

sin(

  X  
= '   tan
Q

Solution:
12.

= tan 

+
)(

Particular solution:

).

 +1 (sin )

= cos(  ).

)]

2 = [ + sin(
Particular solution:

8.

sin3 (

Particular solution:
7.

+ sin (

0 = 1  .

=  .

sin (

Particular solution:
6.

+ sin(

1.2.6-2. Equations with cosine.


14.

15.



+  + cos .

The transformation  = 2a , = 
leads to a Mathieu equation of the form 2.1.6.29:
g  g + (  2  cos 2a ) = 0, where  = 4 ,  = 2= .

+  cos(

Particular solution:

)+

cos2 (

).

=  sin(  ).

2003 by Chapman & Hall/CRC

Page 96

1.2. RICCATI EQUATION ( )

16.

17.
18.
19.

2 ( )

= 2 + cos( ) + cos(
Particular solution: 0 =  .

= 2 + cos (
Particular solution:

cos(

+ cos (
= 1
 .

cos3 (

)]

cos2 (

Particular solution:

= ( + 1)

cos  (

cos (

Solution:
25.
26.

[ cos(

=  

Particular solution:

2

cos (

+ cos(

= .

2
0

).

).

).

+
cos (  )  + L2S .
M

cos(

).

cos( ) = 0.
 sin(  )
=
.
 cos(  ) + 

) 
2

(cos ) .

)2 +

 /X 
= '   tan
Q

) + ](

   +( .
0 = '

[ cos( ) + ] =
Particular solution:

 +1 (cos )

Particular solution:

= tan(  ).

+
)(

+ cos2 (

Particular solution:

).

21.

=[ +

).

= sin(  ).

97

) 2.

2 = [ + + cos( )] 2 + + cos(
Particular solution: 0 = tan  12   .

24.

0 (

= 2 + 2 + cos ( + ) cos 4 ( + ).
The substitution  = H z 2 leads to an equation of the form 1.2.6.3:  =
( sin  ( H +  ) sin  4 ( H +  ).

20.

23.

+ 1 ( ) +

Particular solution:

22.

1.2.6-3. Equations with tangent.


27.
28.
29.

30.

= 2 +  + ( ) tan2 ( ).
Particular solution: 0 =  tan(  ).

= 2 + 2 + 3  + ( ) tan2 ( ).
Particular solution: 0 =  tan(  ) cot(  ).

= 2 + tan + .
The substitution  = 

   tan *  + !( = 0.

leads to a second-order linear equation of the form 2.1.6.53:

= 2 + 2 tan + ( 1) tan2 .
The substitution = +  tan  leads to a separable equation of the form 1.1.2:  = 

+ .

2003 by Chapman & Hall/CRC

Page 97

98

FIRST-ORDER DIFFERENTIAL EQUATIONS

31.

= 2 + tan( ) + tan(
Particular solution: 0 =  .

32.

tan (

= ( + 1)

tan (

Particular solution:
35.

tan (

Solution:
37.

.
+2

).

)+

2

tan (

+
1

tan2 (

)2 +

+ tan(

= .

).

)+

tan + (  )  + L2S .
M

2
0

(tan ) .

=  tan(  ).

  X  
= '   tan
Q

[ tan( ) + ] =
Particular solution:

   +( .
0 = '

tan (

tan

+
)(

Particular solution:
36.

=  

tan (

 +1 (tan )

Particular solution:
34.

0 = 1  .

Particular solution:
33.

tan(

).

1.2.6-4. Equations with cotangent.


38.

= 2 +  + ( ) cot2 ( ).
Particular solution: 0 =  cot(  ).

39.

= 2 + 2 + 3  + ( ) cot2 ( ).
Particular solution: 0 = tan(  )  cot(  ).

40.

41.
42.
43.

= 2 2 cot(
Particular solution:

= 2 + cot( ) + cot(
Particular solution: 0 =  .
= 2 + cot (
Particular solution:

= ( + 1) 

cot (

Solution:
46.

+ cot (
= 1
 .

=  

.
)2 +

   +( .
0 = '

 /X 
= '   tan
Q

[ cot( ) + ] =
Particular solution:

 +1 (cot )

+
)(

cot (

) 2.

Particular solution:
45.

Particular solution:
44.

+ 2 2.
=  cot(  )  cot( ' ).

+ cot(

= .

2
0

2 2

).
(cot ) .

cot (

.
).

+
cot (  )  + L S .
M

cot(

).

2003 by Chapman & Hall/CRC

Page 98

1.2. RICCATI EQUATION ( )

2 ( )

+ 1 ( ) +

0 (

99

1.2.6-5. Equations containing combinations of trigonometric functions.


47.

48.

+ sin (

) cos

).

This is a special case of equation 1.2.6.3 with  = 0 and  =


2.

sin(

) cos

+ sin(

The transformation = cos(  ),

2  =

49.

+  

cos(

51.

+  

sin(

+ cos(

) sin (

Particular solution:
52.

53.

sin

+1

(2 ) =

+ x cot +

+ cos2

tan2 (

(sin )2 .

+ B 2

+1

H + .

1
2

sin(

Particular solution:

 +L

= cot(  ) tan(  ).

cot2 (

+ 
sin(

) + cos2 (

 .

) + ( ) cot2 (

) sin (

cos(  )

tan(

0 = 1 cos(  ).

).

=  tan(  )  cot(  ).

tan2 (

)2

3
4

 +L
M

+ 2 + ( ) tan2 (

form 1.2.2.4: 3  =

cos2' .

The transformation = sin(  ),

59.

=  sin +  cot O/X sin + 

Particular solution:
58.

=  cos +  cot O X cos+ 

Particular solution:
57.

leads to an equation of the form 1.2.2.4:

=  cot  .

x tan +

Solution:
56.

).

tan + (1 ) cot2 .

Solution:
55.

.
The substitution H = tan   leads to a separable equation: 2  sin(2 )H   = !H

Particular solution:
54.

= 1
cos(  ).

sin2

cos(

).

 .

leads to an equation of the form 1.2.2.4:

cos

The transformation = sin(  ),

2  =

0 = 1 cos(  ).

Particular solution:
50.

cos (

).
=

 .

sin(

).

).

sin(  )
leads to an equation of the
2 cos2 (  )

).

2003 by Chapman & Hall/CRC

Page 99

100

FIRST-ORDER DIFFERENTIAL EQUATIONS

1.2.7. Equations Containing Inverse Trigonometric Functions


1.2.7-1. Equations containing arcsine.
1.

+ (arcsin )

Particular solution:
2.

(arcsin )

= ( + 1)

+ (arcsin ) ( 

= (arcsin )

= (arcsin )
= (arcsin )

=  .

=  + .



)2 +

= ( 2 2 +
The substitution H =  

1).

(arcsin ) .

2 2

= '  tan "/XZ 

(arcsin ) .

  + + .
0 = 

= (arcsin )

+1

(arcsin )

+ 

= (arcsin ) (

Solution:
9.

Particular solution:
8.

 + .
0 = '

Particular solution:
7.

Particular solution:
6.


0 = 

Particular solution:
5.

+ (arcsin ) .

= 1
 .

Particular solution:
4.

+  (arcsin ) .

=  .

Particular solution:
3.

(arcsin ) .

(arcsin  ) 

 + L2S .

+ )(arcsin ) .
leads to a separable equation: H   =  

(arcsin  ) + ( !H

+ H + ( ).

1.2.7-2. Equations containing arccosine.


10.

+ (arccos )

Particular solution:
11.

= ( + 1)

= (arccos )
= (arccos )

Particular solution:

= 1
 .


0 = 

Particular solution:
14.

+ (arccos ) .

+ (arccos ) ( 

Particular solution:
13.

+  (arccos ) .

=  .

(arccos )

Particular solution:
12.

0
2

=  .

 + .
0 = '

+1

1).

(arccos ) .

(arccos )

2003 by Chapman & Hall/CRC

Page 100

1.2. RICCATI EQUATION ( )

15.

= (arccos )

Particular solution:
16.

2 2

2 ( )

1 ( )

0 (

(arccos ) .

(arccos  ) 

+ )(arccos )

 + L2S .

leads to a separable equation: H   =  

The substitution H =  

101

(arccos ) .

)2 +

= '  tan "/XZ 

= (



  + + .
0 = 

= (arccos )

Solution:
18.

=  + .

= (arccos ) (

Particular solution:
17.

+ 

(arccos  ) + ( !H

+ H + ( ).

(arctan  ) + ( !H

+ H + ( ).

1.2.7-3. Equations containing arctangent.


19.

+ (arctan )

Particular solution:
20.

= ( + 1)

= (arctan )

= (arctan )

= (arctan )

Particular solution:
25.

= (arctan )

= (

1).

(arctan ) .

(arctan ) .

2 2

2 2

)2 +

  + + .
0 = 

(arctan )
1

= '  tan "/XZ 

= ' + .

+1

=  .

= (arctan ) (

Solution:
27.

=  

Particular solution:
26.

= 1
 .

 + .
0 = '

Particular solution:
24.

+ (arctan ) .

+ +

Particular solution:
23.

(arctan ) .

+ (arctan ) ( 

Particular solution:
22.

=  .

(arctan )

Particular solution:
21.

(arctan ) .

(arctan  ) 

+ )(arctan )

 + L2S .

leads to a separable equation: H   =  

The substitution H =  

1.2.7-4. Equations containing arccotangent.


28.

+ (arccot )

Particular solution:

(arccot ) .

=  .

2003 by Chapman & Hall/CRC

Page 101

102
29.

FIRST-ORDER DIFFERENTIAL EQUATIONS

(arccot )

Particular solution:
30.

= ( + 1)

+ (arccot ) ( 


0 = 

= (arccot )

= (arccot )

= (arccot )

Particular solution:
34.

= ( 2 2 +
The substitution H =  

2 2

= '  tan " X  

)2 +

1).

(arccot ) .

  + + .
0 = 

= (arccot )

Solution:
36.

= ' + .

= (arccot ) (

(arccot )

+
0

Particular solution:
35.

 + .
0 = '

Particular solution:
33.

=  .

+1

+ +

Particular solution:
32.

= 1
 .

Particular solution:
31.

+ (arccot ) .

(arccot ) .
1

(arccot ) .

(arccot  ) 

 + L2S .

+ )(arccot ) .
leads to a separable equation: H   =  

(arccot  ) + ( !H

+ H + ( ).

1.2.8. Equations with Arbitrary Functions


]

Notation:  =  ( ) and = ( ) are arbitrary functions;  ,  , B , and are arbitrary parameters.


1.2.8-1. Equations containing arbitrary functions (but not containing their derivatives).

1.

! .

Particular solution:
2.

Particular solution:
3.

= .

=.

= 1
 .

+! .

Particular solution:

4.

= ! 2 ! + 1 .
Particular solution: 0 =   .

5.

= ( + 1)w

Particular solution:

2 2

   .
0 = 

Particular solution:
6.

+1 !


0 = 

.
! .
.

2003 by Chapman & Hall/CRC

Page 102

1.2. RICCATI EQUATION ( )

7.

Solution:
8.

9.

2 !

+ +

2 !



+ (

+ "I

+ "I +

"I





Solution:
16.



19.

 ( )( H 2 + !H +  ).

! 2

2 2 ("

! ).

 

+ ) +

"I


Particular solution:

=   .

=  

+ "I + 

+ 2 

if  < 0.




0
2

if  < 0.

+ .

  !
"



=   .

=   .

if  > 0,

leads to a separable equation: H   =  ( )( H

+ 

if  > 0,

+ ( ! ) +



103

  !

2 2

0 (

=   .

Particular solution:
20.

Particular solution:
=

"

#!

=   .

The substitution H = 
18.

!

   tan O lX      + L
M 
P

|  |   tanh O |  | X     + L

P
M

! ( 

 !

 2 !

+ 

Particular solution:
17.

+ 

Particular solution:
15.

= .

 !

Particular solution:
14.

1 ( )

leads to a separable equation: H   =  

   .
0 = 

Particular solution:
13.

) +

Particular solution:
12.

   .
0 = 

Particular solution:
11.

" .

Particular solution:
10.

2 ( )

  tan O  X   1   + L
M
P
|  |   tanh O |  | X  1   + L

M
P

The substitution H =  

=  

! 2


 

2 2

  !

2 2

("

+ !H +  ).

! ).

2003 by Chapman & Hall/CRC

Page 103

104
21.

FIRST-ORDER DIFFERENTIAL EQUATIONS

Solution:




+ !

    2J 2 tan O lX    2J 2 
2
|  |   J 2 tanh O |  | XZ 

22.

= ! 2 tanh2 ( )( ! + ) +  .
Particular solution: 0 =  tanh(  ).

23.

24.
25.
26.

27.
28.

30.

if  < 0.

)( !

= ! 2 coth2 (
Particular solution:

+ ) +  .
=  coth(  ).

= ! 2 2 ! +  sinh( ) 2 ! sinh2 (
Particular solution: 0 =  cosh(  ).

= ! 2 +
Particular solution:

).

(ln )2 .
0 =  ln  .

= ! ( + ln )2 .
 ( )
1
 =L .
+X
Solution:
+  ln 
 M

= ! 2 ln
Particular solution:

= ln

!
0

! (

= sin( ) 2 +
Particular solution:

+ ln + .
=  ln  .

ln ) ! .
1
.
=
 ( ln   )

cos( ) ! .

0 = 1 cos(  ).

= ! 2 2 ! +  sin( ) + 2 ! sin2 (
Particular solution: 0 =  cos(  ).

31.

= ! 2 2 ! +  cos( ) + 2 ! cos2 (
Particular solution: 0 =  sin(  ).

32.

33.

if  > 0,

Particular solution:
29.

 +L
M 2J 2 P
  +L
M
P

= ! 2 tan2 (
Particular solution:

= ! 2 cot2 (
Particular solution:

).
).

)( !
0

) +  .
=  tan(  ).

) +  .
=  cot(  ).

)( !

1.2.8-2. Equations containing arbitrary functions and their derivatives.


34.
35.

= 2 ! 2 + ! .
Particular solution:

= ! 2 !"I + "
Particular solution:

= .

.
0

= .

2003 by Chapman & Hall/CRC

Page 104

1.2. RICCATI EQUATION ( )

36.

= !

!"I

= " ( ! )2 +

"

"
!

! 2

!

!

 !

Particular solution:
41.

+"




!
!

= 1
 .

= .

=
 .

1 ( )

0 (

105

= .



0 = 1  .
2

  tan O  X  

42.

+ !

Solution:

+ " ( ! ) = 0.

Particular solution:
40.

Particular solution:
39.

2 ( )

Particular solution:
38.

" .

Particular solution:
37.

|  |  tanh O |  | X

 +L
 

if  > 0,

 +L
P

if  < 0.

Particular solution:


=  
 .

1.2.9. Some Transformations


]

Notation:  , , and % are arbitrary composite functions of their argument, which is written in
parentheses following the function name (the argument is a function of  ).
1.

2.

3.

2 ! (

+ ).

The transformation =  +  , = 


 leads to the equation  =

(1 

).

The transformation = 1
 , = 

1
!IO

( + )4
The transformation

4.

+  ( ).

+
.
+ P

 + 
,
( +
M

4 ! ( )

+  ( ).

 =
 leads to the equation 3

= 1 [( ( + )2 + ( ( ( + )],

M
M
u

leads to a simpler equation: = 2 + 2  ( ).


=

where

=
M

( ,

+ 1.

The substitution =

1
2

leads to the equation  =

+  ( ).

2003 by Chapman & Hall/CRC

Page 105

106
5.

FIRST-ORDER DIFFERENTIAL EQUATIONS

2 ! (

+ ) + 14 (1

The transformation =   +  ,

2  =

6.

7.

B

).
1

1 B 

leads to a simpler equation:
2 B

+ " ( ) + % ( ).

The substitution = 1


 = % ( )
leads to an equation of the same form: p

1
4

 2 ! ( 

 

 

+
!IO

(   + )4
 

 2


2
2 2 
2
= ( (  +  M ) + (    M ,
e  
2  
leads to a simpler equation: u  = 2 + ( e )2  ( ).

10.

11.

 coth( )  .
The transformation = coth(  ), =
equation: 3  = 2 + 2  ( ).

+ ! ( ln + ) + 41 .

)!

+ sin 4 (

(cot(

The transformation = cot(  ),

2  =

13.

2
2

 ( ).

+ cos 4 (

14.

sinh2 (  )

 ( ).

+ sin 4 (

+ ) !IO

=
M

( ,

1
2

sinh(2  ) leads to a simpler

1
2

sinh(2  ) leads to a simpler

1
leads to a simpler equation: u  =
2

+

 ( ).

)).
= sin2 (  )

)u
!  tan(

The transformation = tan(  ),

2  =

where

)!

+ cosh 4 (

1
The transformation =  ln  +  , = 
12.

 ( ).

)!

+ sinh 4 (

 tanh( )  .
The transformation = tanh(  ), = 1 cosh2 (  ) +
equation:  = 2 + 2  ( ).

+
.
+ P

4
The transformation

( ) +  ( ).


1
  leads to a simpler equation:  = 2 +
2


   + 
=
,

(  +
M

9.

The transformation =  , =
8.

+ ( B )2  ( ).

= ! ( )

+ cot(  )S leads to a simpler equation:

) .
= cos2 (  )

sin(
sin(

tan(  )S leads to a simpler equation:

+ )
.
+ )P

sin(  +  )
sin2 (  +  )
, =
+ cot(  +  )S leads to a simpler
sin(  +  )
sin(   ) Q
3

2
2
+ [ sin(   )]  ( ).
equation: =
The transformation =

2003 by Chapman & Hall/CRC

Page 106

107

1.3. ABEL EQUATIONS OF THE SECOND KIND

1.3. Abel Equations of the Second Kind


1.3.1. Equations of the Form !$ = ( )
1.3.1-1. Preliminary remarks. Classification tables.
For the sake of convenience, listed in Tables 58 are all the Abel equations discussed in Section 1.3.
Tables 57 classify Abel equations in which the functions  are of the same form; Table 8 gives
other Abel equations. In Table 5, equations are arranged in accordance with the growth of the
parameter . In Table 6, equations are arranged in accordance with the growth of the parameter  .

In Table 7, equations are arranged in accordance with the growth of the parameter . The rightmost
column of the tables indicates the equation numbers where the corresponding solutions are written
out.
TABLE 5
Solvable Abel equations of the form ! =  + #* + , # is an arbitrary parameter

arbitrary
7
4
5
2
2
2
5
3
5
3
5
3
7
5

Equation

2( + 1)

( + 3)2

15 4
6
12
0
2
3
16
9
100
63
4
5
36

Equation

1.3.1.10

1.3.1.16

1.3.1.56
1.3.1.54
1.3.1.47
1.3.1.33
1.3.1.19
1.3.1.30
1.3.1.23
1.3.1.48
1.3.1.27

1
2
1
2
1
2
1
2
0
0
1
2
2
2

2
9
4
25
0
20
arbitrary
0
12
49
6
25
6
25

1.3.1.26
1.3.1.22
1.3.1.32
1.3.1.55
1.3.1.2
1.3.1.1
1.3.1.53
1.3.1.45
1.3.1.46

Solvable Abel equations of the form !

1
3
5
5
11
1
3
1
3
1
3
1
3
1
5
0
2

3
7
5
13
11
5
3
5
3
5
3
5
3
4
5
1
2
3

arbitrary
2 +1
2
4
0
5
36
33
196
3
16
3
16
3
16
15
4
10
49
2
9
4
9

TABLE 6
=  + h#* K + #

'& , # is an arbitrary parameter

Equation

1.3.1.5

1.3.1.13

1
arbitrary
286#
3
arbitrary
3
5
6
13 #
5
arbitrary
2

1
arbitrary
770 #
9
arbitrary
12
12
3
7 #
20
arbitrary
2

1.3.1.7
1.3.1.62
1.3.1.69
1.3.1.61
1.3.1.40
1.3.1.15
1.3.1.60
1.3.1.68
1.3.1.3
1.3.1.14

2003 by Chapman & Hall/CRC

Page 107

108

FIRST-ORDER DIFFERENTIAL EQUATIONS


TABLE 7
Solvable Abel equations of the form !   =  + ./# (  1 J
# is an arbitrary parameter

arbitrary 0
2( 1)

( 3)2

1 4
30
121
12
49
12
49
12
49
12
49
12
49
12
49
12
49
12
49
12
49
6
25
6
25
28
121
2
9
2
9
2
9
10
49
4
25
4
25
0
0
0
0
0
0
0
2
2
20

arbitrary
2
( 3)2

1 4
3
242
arbitrary
1
98
6
49
2
49
4
49
1
49
6
49
2
49
1
2
25
6
25
2
121
arbitrary
arbitrary
6
2
49
arbitrary
1
50
arbitrary
1
B +2
B +2
1
2
arbitrary
2
2
arbitrary

( + 3)

1
21
arbitrary
25
1
5
10
5
3
1
3
49 + 3 $
2
2
5
0
0
0
4
0
7
0
1
1
1
1
1
0
10
10
0

+ # + =/#

J ),

2 1 2

Equation

arbitrary

1.3.1.2

3 ( + 3)

1.3.1.12

3
6
0
10
5
15
10
65
12
55
15
196 + 75 $
16
6
4
15
arbitrary
arbitrary
2
12
arbitrary
6
arbitrary
arbitrary
(B + 1)(B + 3)
2B + 3
0
3
0
30
30
arbitrary

1.3.1.17
1.3.1.29
1.3.1.53
1.3.1.25
1.3.1.38
1.3.1.24
1.3.1.31
1.3.1.52
1.3.1.28
1.3.1.58
1.3.1.64
1.3.1.20
1.3.1.39
1.3.1.51
1.3.1.3
1.3.1.26
1.3.1.11
1.3.1.57
1.3.1.22
1.3.1.59
1.3.1.32
1.3.1.36
1.3.1.34
1.3.1.35
1.3.1.37
1.3.1.4
1.3.1.1
1.3.1.50
1.3.1.49
1.3.1.55

+3 +9

5
35
0
41
8
34
27
262
23
166
12
49 15$
2
19
7
106
arbitrary
0
1
61
0
49
0
2
2(B + 2)
2(B + 2)
2
4
arbitrary
19
31
0

Given below in this section are all solvable Abel equations known so far. The equations are
arranged into groups, in which all solutions are expressed in terms of the same functions. Notation
is given before each group.
In most cases the solutions are presented in parametric form:

 =  1 (U , L ),

=

2(

U , L ),

where U is the parameter and L is an arbitrary constant.

2003 by Chapman & Hall/CRC

Page 108

109

1.3. ABEL EQUATIONS OF THE SECOND KIND

TABLE 8
Other solvable Abel equations of the form !  

=  ( )

Function  ( )

#* 

, $I  + , $

#*

9
625

Equation
1.3.1.6
(particular solution)

 

2 2 1

1.3.1.44

27
 #* J + 34 # 2  1 J 3 625
# 4  5 J 3
4 5 J 3
6
25
 + 75 #* 1 J 3 + 313 # 2  1 J 3 100

3 #
6
25
 +  1 J 3 +  + ( 1 J 3 +  2 J 3
M
(coefficients  ,  , ( , and are related by an equality)
M
21  + 7 # 2  123 1 J 7 + 280 #* 5 J 7 400 # 2  9 J 7 
3
4

100

1 3

3
2

1.3.1.66
1.3.1.67
1.3.1.65
1.3.1.70

#*

1.3.1.63

+ $I + L

1.3.1.18

+ 4#

3
9  2 6 2
 + 2 2
32
64  + 

1.3.1.43

6 2 + 5  2
3
 + 2 2
8
16  + 

1.3.1.21

6 2 + 9 #
3
 + 2
8
16  + #

1.3.1.41

30 2 + 33 #
9
 + 2
32
64  + #

1.3.1.42

# + $ exp(2
# )
# [exp(2
# ) 1]

 2  2
 2  2




 (  + 1)  
+    

+  

1.3.1.8
1.3.1.9
+

1.3.1.73
(particular solution)

1.3.1.74
(particular solution)
1.3.1.75
(particular solution)

2  2 sin(2  ) + 2  sin(  )
1.3.1-2. Solvable equations and their solutions.
1.
2.

=  .
Solution:  =

# ln | + # | + L .

=  +( ,
 0.
Solution in parametric form:

 = L exp O X
U

U #P

$
#

= L3U exp O X

U #P

2003 by Chapman & Hall/CRC

Page 109

110
3.

FIRST-ORDER DIFFERENTIAL EQUATIONS


= 92 +  + ( 1 ) 2 .
1 b . Solution in parametric form with # > 0:

 = 2

2
(2 , 1) L3U  ( , 2)U , 1

,
L3U  + U + 1

where # = 23  ( ,

+1
+ , 2 L3U  + U
,

L3U + U + 1

= 23  3 J 2 (2 , 1)( , 2)( , + 1).

, + 1), $

( , 1)2 L3U 

= 6 

2 b . Solution in parametric form with # < 0:

 = [2  + * ( L 1 3 L 2 ) sin U (3 L 1 + L 2 ) cos U ]2 ,
= 6 , ( L 12 + L 22 ) 2 [ L 1 ( 2 2 ) 2 L 2 ]  + * sin U
[2 L 1 + L 2 ( 2 2 )]  +* cos U/. ,
where =  (  + * + L

sin U + L

cos U )2 , # = 19  (3

2 ), $

2
27

 (9

+ 5 2 ).

3 b . For the case # = 0, see equation 1.3.1.26.


4.

= 2 ( 1) 2 + 4 + 3
Solution in parametric form:

1 ) 2 ).

 = 14  (3 A 2U01 )2 ,
where

5.


X


X

=X

 = O



Here,

where

= 4 # + 1.

P
2



(U












# = 12  1 J 2 ,

+ U ),

= arctan U L ,

= 1 + U 2,

= U

= 1 U 2.

1,

= (U + 1) O

2
+ U # ) exp

+ U # ) exp

1 2

$O

2U + 1
arctan

2
2U + 1

2 13

2U + 1
(U + U # )
2U + 1 +
2U + 1
2
U
arctan
L + 2 $ X exp

M
2
U
L + 2 $wX exp
2U + 1 M
2



(U



1 2






1 01
2

1
U 1
U
= ln
L ,
M
U 2 1 2 U + 1

U
1
1+U
= ln
L ,

M
1U 2
2 1 U

=  + ( 1 ( 2
Solution in parametric form:

1+U

= A 01 (

2U + 1
L + 2 $wX
2U + 1 +

2 13

1 2

if

< 0,

if

= 0,

if

> 0,

if

< 0,

if

= 0,

if

> 0,

2003 by Chapman & Hall/CRC

Page 110

111

1.3. ABEL EQUATIONS OF THE SECOND KIND

6.

 

Particular solution:
7.

/( 2 2 1 .

#
=  $I 

, $

=  1  2 3 .
Solution in parametric form:

 = U

(U ln |1 + U | L )1 J 2 ,

where # =  2
2.
8.

/(

=

1+U

1
(U ln |1 + U | L )1 J 2 U (U ln |1 + U | L )1 J 2 S ,
2

=  + (4 2 )65 .
Solution in parametric form:

U 2 + #2$
# ln U + U 2 + #2$ + L

 = # ln
9.

 = # ln

+1

(arctan U L ) ,

# V
U + (U
U

+1

) - 2 U L ,

+ = 1 AIU +

2( + 1)
+
( + 3)2
Solution in parametric form:

1 +
+3P

= 92 + 6  2 (1 + 2 
Solution in parametric form:

 =#
where

1 2, 3 2 ,

# .

1)(arctan U L )W .

2
1

+1

 =

3)2

where # =

= ! + +

U L ,

2
1

O + + +

2
U ,
1 P

= 12 #

2U ),

1 2 .

) J

+1 1 2

> 0.

A 6U 1 J 2 )2 ,

[( 3) + + +3U ]2 ,

= X (1 AIU +

+ .

+3)

 1J

,0

 + = + + U .

+1 ,

1 ) 2 ),

2( 1)
2
V
+
( 3)2
( 3)2
Solution in parametric form:

+ = +

+3
U + +
1

+1
O

where # = A2
2

12.

 =

11.

In the solutions of equations 1015, the following notation is used:

+ , - = X (1 AIU +
10.

=  (  2 )65 1).
Solution in parametric form:

# ln U + U 2 + #2$ + L

U 2 + #2$

=U

1 2

+(4

+3

+9)  +3

+ [A ( 1)U +

+1

+3) 

2 1 2

W.

+ 2 + +2U + ],

2003 by Chapman & Hall/CRC

Page 111

112
13.

FIRST-ORDER DIFFERENTIAL EQUATIONS

 =
14.

+1
+  1 
4 2
Solution in parametric form:

1 2

U 1 J 2 2+

 =
15.

U V 1 T (2 + 1)U + +1 W 2+

,
2  U 1 J 2 2+ 1 J 2

= 49 + 2  2 + 2 
Solution in parametric form:

1
U
3#

 3,

1
U
9#

2 5 3

= 2 # ,

= +

, 3 2.

3 (U

AIU 4 3 ).

 = U 1 J 2 3 J 2  3 J 2 ,
= 14 U 1 J 2
1 4J 3
 , = 5 J 3 ,  =  5 J 3 .
where # = 24
]

3
= 16
+ 5  1 ) 3 12 
Solution in parametric form:

where

J 

J (

3 2

1 2

2 3

2U  U

),

In the solutions of equations 1618, the following notation is used:

16.

 = X exp(T"U 2 ) U L ,
M

=  1 .
Solution in parametric form:

 =  

17.

exp(T"U 2 ),

= 41 + 14  ( 1 ) 2 + 5 
Solution in parametric form:

1
16

2
 V 3 A 8U exp(AU 2 )W ,

=A

7 2
A

V exp(T"U 2 ) A 2UW ,

=  

 =
18.

= 2U X exp(T U 2 ) U L2SrA exp(T U 2 ).


M
Q

+ 3

1 ) 2 ).

=   exp(AU 2 ) V (2U

# = T 2 2.

A 1)  exp(A"U 2 ) A*U W ,

where # = 14  .

.
8 2
Solution in parametric form:

 = A  (  )1 ( 2 T 2  2 ),
]

where

=  (  )1 [exp(T U 2 ) 2  2 ].

In the solutions of equations 1921, the following notation is used:

= U (U + 1) ln L  U + U + 1  ,

19.

U +1
,
U

= 2 +  2 .
Solution in parametric form:

 =1

 = 13 !

20.

J U ,

2 3

6
2
= 25
+ 25
 (2 1 )
Solution in parametric form:

 = U

(5 u 3U )2 ,

J  U u  ,

= !

2 3 2
3

+ 19 

+ 6

= 5 U

U +1
ln L  U + U + 1  .

where

1 ) 2 ).

[(2U + 3) 2U 2 ],

3
# = 243
2  .

where

# =  .

2003 by Chapman & Hall/CRC

Page 112

113

1.3. ABEL EQUATIONS OF THE SECOND KIND

21.

3
37 2
+
+
8
8
Solution in parametric form:

7
16

 2 2U 2 
 =
,
2 2 U 
]

4U 

U 

4 2

In the solutions of equations 2225, the following notation is used:

22.

= A (U

4
= 25
+  1 ) 2 .
Solution in parametric form:

. 2.

3 2 9 8
,
3
4

4
2 (14

12
1
= 49
+ 98
 (25
Solution in parametric form:

. .

4
3 (21 2 4

= A (U

6U

+ 4 L3U 3).

4 3
( 22
3

3 ),

# =A

where

4
5

 5 .

16

+ 34 

2 2
3) ,

2 4 ),

2 1 2

4
2
3 (4
3
2

+ 41 
= 21 

. .

+ 15 

= 28 
1 2

1 2 9 8
( 32
3
4

+ 10 

# = A 9 2 (10  )2 J 3.

where

).

3U

1 ) 2 ).

. .

4
2
4 (9 2 4
3

.
T

2
2

2
4

. .

where

# = 3  .

. .

where

# = 8  .

2 3 ),

2
2 3 ),

In the solutions of equations 2629, the following notation is used:


1
3

= exp( 3 U ) + L sin U ,

= 2 exp( 3 U ) L sin U + 3 L cos U ,

= 2 exp( 3 U ) L sin U 3 L cos U ,

= 92 +  1 ) 2 .
Solution in parametric form:

=4

8 8

1 3

2
2.

 = 3

2 2
1
2,

= 2

5
= 36
+  7 ) 5 .
Solution in parametric form:

2 2
1 ( 2

8 8

1 3 ),

where

# = 16(3  )3 J 2.

 = 48 
28.

2 2
2) ,

27.

= 9

12
2
= 49
+ 49
 (5 1 )
Solution in parametric form:

26.

= 4

 =
]

4 3
,
3

9
= 100
+  5 ) 3 .
Solution in parametric form:

 =

25.

3U + L ,

 = 10 
24.

=U

 = 5
23.

1),

8 5J 2 8
1

5 4
,
4

12
6
= 49
+ 49
 (3
Solution in parametric form:

 =

8 8

4
2 (7 1 3

2 2
2) ,

= 5

1 2

J 8

1 2 5 4
(8 13
1
4

+ 23 

= 7 

+ 12 

8 8

8 8

8 8

2 4 ),

where

# = (48 )2 J 5  2 .

1 ) 2 ).

4
2
1 2 (4 1 2

8 8

2
1 3

8 28
2

3 ),

where

# = 
2.

2003 by Chapman & Hall/CRC

Page 113

114
29.

FIRST-ORDER DIFFERENTIAL EQUATIONS


30
3
= 121
+ 242
 (21
Solution in parametric form:

 =
]

8 8

6
1 (11 2 4

64

3 2
1) ,

+ 35 

1 2

= 11 

+ 6

2 1 2

6
2
4( 4
1

).

8 28

+ 32

8 38

where # = 32  .

2 ),

In the solutions of equations 30 and 31, the following notation is used:


{ = tanh(U + L ) + tan U , { = tanh(U + L ) tan U ,
91
9 2
9
=
cosh
U

sin(
U
+
L
),
1
2 = sinh U + cos(U + L ),
3 = sinh U cos(U + L ).

30.

3
= 16
+  5 ) 3 .
1 b . Solution in parametric form with # < 0:

 = 8

3 2
,
1

= 3

{ {

3 2
(2
1

1 2 ),

2 b . Solution in parametric form with # > 0:


31.

 = 4

9 3 J 2 9 3 J 2
,
1

9 1 J 2 9 3 J 2 9
(

= 3

2
4
49

2
1

12
= 49
+  (10 1 ) 2 + 27  + 10 
1 b . Solution in parametric form with # < 0:

 =  (10 7

{ {

1 2)

= 7

{ ({
1

3
1

+3

9 9

2 3 ),

2 1 2

{ {

2
1 2

2 b . Solution in parametric form with # > 0:

 =
]

9 9

4
1 (7 2 3

5 12 ),

= 7 

9 4 9 9
1

3
2( 2

# = 12 8 J 3 .

where

# = 3 2 (4  )2 J 3 .

where

).

4 2 ),

9 9

2
2 3

+2

where

9 29

1 3 ),

# = 2  .

where

# =  .

In the solutions of equations 3243, the following notation is used:


:
L (U ) + L 2 n (U ) for the upper sign,
= 1
L 1  (U ) + L 2 ; (U ) for the lower sign,

:
:
:
: :
:
:
:
 = U ( )  + @ ,
= 1 J 3 , 1 = U  + 13 , 2 = 12 ACU 2 2 , 3 = A 23 U 2 3 2 1 2 ,
*
*
where (U ) and n (U ) are the Bessel functions, and  (U ) and ; (U ) are the modified Bessel
functions.

: :
The solutions of equations 3243 contain only the ratio 
, where the prime
:*
denotes differentiation with respect to U . Therefore, for symmetry, function is defined in terms of
two arbitrary constants L 1 and L 2 (instead, we can set, for instance, L 1 = 1 and L 2 = L ).
x3y'z|{}~/

32.

33.

34.

=  1 ) 2 .
Solution in parametric form:

J :

4 3

=  2 .
Solution in parametric form:

:
 = 2 U 4 J 3

= U

4 3

J :

2,

= A 3 U

2 3

J :

12 ,

1
2 ,

=  ( + 2)[ 1 ) 2 + 2( + 2) 
Solution in parametric form:

 =

35.

 = U

2 [  ( @ + 1) ]2 ,

=

=  ( + 2)[ 1 ) 2 + 2( + 2) 
Solution in parametric form:

 =   2 [U 2 rA (2 @ )  ] ,
1
.
where # = T/@  , @ =
B +2

3,

+ ( + 1)( + 3) 

2(  2 2 @

1
2

!AhU

+ (2 + 3) 

# =T

where

2 1 2

].

where # = @  , @ =

2 ),

# = 36 3 .

where

2 1 2

 J

1 3 2
.
3

B +2

].

= A U 2  2 [  2 + 2(1 @ ) ACU

2 ],

2003 by Chapman & Hall/CRC

Page 114

115

1.3. ABEL EQUATIONS OF THE SECOND KIND

36.

=  1) 2 + 2 2 + (
Solution in parametric form:

where $
37.

1 ) 2 .

 =#

= (1 @ 2 ) #

: 

2 (U

)2 ,

=#


T U 2 ) 2 ],
2 [U 2 ( )2 ( @ 2 C

and the prime denotes differentiation with respect to U .

= 2 2  1) 2.
Solution in parametric form:

:
:
 =  ( 0 )2 (U

:
A 2 0 )2 ,

:
:
:
= A U ( 0 )2 [U ( 0 )2 + 2

:  :
0 AIU

2
0 ],

where # =  and the prime denotes differentiation with respect to U .


38.

12
6
= 49
+ 49
 ( 1) 2 + 8 
Solution in parametric form:

 = 3
where # = 2 3  ;
39.

and

7U

2 2

3)

+ 7

3
= 16
+ 3  1 ) 3 12 
Solution in parametric form:

+ 4

= 5 U

2 5 3

2
1 ),

1 ) 2 ).

2
2

2(

3 ),

where

# = 
2.

2
3 2

J ,

3 2 3 2

1
J and  3 J 2 are expressed in term of modified Bessel functions; # = 8  4 J 3 .

3 2

J
J

1
2
2 (2

1
2
2 (3 2

T 3 22 ),
2
1

12

A 4U

1 ),

= 32  2 .

9
15 7 2
+
T]
32
32
Solution in parametric form:

7
64 2 ]
T

J
:
= A 14 
 U 1 3 J 2

 = 21 U
2

1 2
1
1 2
1

3 2

where 

4
2
1 (3

2
3  3 J 2 2 3 J 2  3 J 2 U
=
:
4
U 3J 2 3J 2  1J 2

J
:
= T 18 
 U 1 3 J 2

where 

  33J J 22
 =
,
:
U 3 J 2 33J J 22

 = 41 U

42.

3
37 2
3 2
+
A 2A
7 2
8
8
16
A
Solution in parametric form:

= 28 U

),

1 ) 2 ).

are expressed in term of modified Bessel functions.

6
6
= 25
+ 25
 (2 1 )
Solution in parametric form:

where
41.

 = U

40.

4
2
1 (5 1

+ 5

3 2

J
J

3 2
2
3 2
2

1 2
(2 2 3 3
3
1 2
2
(3 23
3

TCU

A 3 23 ),
3

2
3 ),

= 6 2.

2003 by Chapman & Hall/CRC

Page 115

116
43.

FIRST-ORDER DIFFERENTIAL EQUATIONS

3
3 7

32
32

15

7
64

Solution in parametric form:

3 2
2

 = 12 
]

1
2
3 ( 3

3
2 ),

1
24

3 2
2

1
2
3 (3 3

3
2

12

A 4U

3 ).

In the solutions of equations 4452, the following notation is used:

<

=U

A (4 <

A (4 <

1) + 3U 2 <IT 1,

1) A 2U'< 2 ,

=U

= U 2 <pT 1,

A (4 <

1) + 2 < .

= < (U ) is given implicitly as follows: U = X

<

L 2 . The upper
A (4 < 3 1)

sign in this formula corresponds to the classical elliptic Weierstrass function < = < (U + L 2 , 0, 1).
Here, the function

44.

9
625

Solution in parametric form:

 = 5  U 2 I
< T
45.

6
= 25
+

1
2

,

= U 2

4,

where

6 1
.
125

# =A

Solution in parametric form:

 = 5 U 2 < ,
46.

6
25

= U 2

where

# =A

where

# =A

4,

6 1
.
125

Solution in parametric form:


= U 2

 = 5 ! 2 ,
47.

= 12 +

5 2

4,

6 1
.
125

Solution in parametric form:

 = -<
48.

63
4

5 3

= -<

4 7
,
3

6 7

<

4 7
(14 2
3

6 7

3),

where

# = T 147  7 J 2 .

Solution in parametric form:

 = 2 !
49.

3 2
3

9 8
,
4

= !

= 2 + 2  (10

1 2

1 2
3

+ 31 

9 8
(9
4

+ 30 

2
3

T 16 </

2 1 2

2
4 ),

where

2
2J 3
# = 128
.
3  (2  )

).

Solution in parametric form:

 = -<

VU

where # =  .

A (4 <

1) 3 <"W ,

= 2 U'<

V<

A (4 <

1) A 2U'<

AIU!W ,

2003 by Chapman & Hall/CRC

Page 116

117

1.3. ABEL EQUATIONS OF THE SECOND KIND

50.

= 2 + 2  (10 1 ) 2 + 19 
Solution in parametric form:

2
2 (

 = !

51.

=X

</

2
4

+ 15 

2
1

+ 7

2 1 2

4
3

2
4 (6

</

# =  .

where

).

3 (7

= 56 !

2 ),

),

# = A 2  .

where

4 ),

# = T 3  .

where

(incomplete elliptic integral of the second kind),

2=U

+ L ),

( $

3 = 4U 12 T 22 .

1),

12
= 49
+  1) 2.
Solution in parametric form:

= 2  ( + L )4 (  + Lu 2U 2 ),

where

12
49

# =A

= 6 +  4 .
Solution in parametric form:

7 .

3 5 2 5
,
1

= U

= 20 +  1 ) 2 .
Solution in parametric form:

3 5 2 5
(5
1

$ 1 2),

where

# = T 150  5 .

 = 

4 3 2
2,
1

+  7 .
= 15
4
Solution in parametric form:

= 4 

4 3 2
( 2
1

T 9U 12 ),

where

# = A 108 3 J 2.

 =  11 J

57.

2 2
3) ,

T 15

A (4U 1)
A (4U 3 1), 1 = U (2 lTCU

 = U

56.

+ 106 

3
2

</ 3 T 2U
1
 (5 1 ) 2 + 262  + 65  2 1 ) 2 ).
49

 = 7 U 2 ( + L )4 ,

55.

A 6

1 ) 2 ).

In the solutions of equations 5360, the following notation is used:

54.

2
2 (

= A 44 -< 2

12
= 49
+
Solution in parametric form:

53.

1 2

5)2 ,

4
3 (28

= 2 !

28
2
= 121
+ 121
 (5
Solution in parametric form:

 = !
]

 =  (22 < 2

52.

2)

+ 30 

= 12


3 12 ),
= 4910 + 492  (4 1 ) 2 + 61  + 12 2 1 ) 2 ).
2 3 8
,
3

3 2 3 8
( 2 3
1
3

# =A

where

3 8
.
4

Solution in parametric form:

 =  (7 $ 1 3)2 ,

58.

= 14  1 [A (10U

12
2
= 49
+ 49
 ( 1 ) 2 + 166 
Solution in parametric form:

 = 

4
2 (42

U 12 T 5 22 )2 ,

+ 55 

= T 84 

1)

2 1 2

2 4
1 2 (3

],

where

# =  .

).

U 22 + 2 T 12U

2 2
1 ),

where

# =A  .

2003 by Chapman & Hall/CRC

Page 117

118
59.

FIRST-ORDER DIFFERENTIAL EQUATIONS


4
1
= 25
+ 50
 (7 1 )
Solution in parametric form:

 = 
60.

+ 49 

+ 6

1 ) 2 ).

16 12 )2, = 5 14 3 (A 3 32 22 3 + 8 12 3 ),
15
+ 6  1 ) 3 3  2 5 ) 3 .

4
1 (5 2 3

2 3 2 3 4
,
1
2

3
= 16
+

1 3

1 2 1 2 3 4
(2
1
2

= U

The substitution  = U

5 3

U 22 + 2 3U

J leads to an equation
!  = 3 U 5 J 2 + 9 U
2

32

32 *
# U

32 $ ,

= 3#
2, 

1J 2
 +  53 $  U 1 J

coincident with equation 1.3.3.13 when B = 1


2, ( = 0,  = 3
4,

5
= 36
+

3 5

7 5

> 0.

The transformation  =  13 U + #
$ 

5 4

equation of the form 1.3.1.3:

63.

= (

where # = 31  (2  )1 J 3 .

2 2
1 ),

3 2

62.

# = 8  .

where

= 4
Solution in parametric form:

 = 2 U 3 J
61.

5
6

= 3 $ .

leads to an

*  = 2 U + 2 # + O 5 1 J 2 1 .
U
*
9
3$
27 $P

)1 ) 2 .

4(  2

4(  2 2 +  1 +  0 )
+  1 +  0)
,
=

+
, where parameters
4# 2
4# 2
2
 2 ,  1 , and  0 are found from the relations $ = 4#3 2  0 and L =  1 4  0  2 , leads to a Riccati
equation:
A/,  = ( 41 +  2 ) 2 +  1 + #* +  0 .
The transformation  =

For L > 0, we can set  2 = 0,  1 = L , and  0 = $ .


In books by Zaitsev & Polyanin (1993, 1994) it is shown that the original equation is
reducible to the degenerate hypergeometric equation.

64.

12
1
= 49
+ 3  ( 49
+

The substitution 
( = 0,  = # , and

65.

66.

( ) 1)

+ 3

4
( 49

5
2

)+

15
4

1
( 49
+

5
4

( )

1 2

# ) leads to an equation of the form 1.3.3.13 with B = 3,  = 4


7,
#
$ ):
M
!  = (4 2 + 5 # ) V 48 4 + 12 #  2 $  2 + 3 # 2 W 3 .
49
7
= ( + 54
= 12 ( 72

V (2  ) 1 ) 3 3 ( (2  + 1) + ( 2 (1 3  ) 1 ) 3 >( 3 2 ) 3 W .
2
1 2 2
$ 2 (3 2 $ )
= +
ShO +
leads to a Riccati
The transformation  = 3 ,

Q
5 ($
+ 1)
$P
equation:
 2 2 25 $ 3 + 254 #2$ 6   = $I 2 +  25 $ 3  + 35 $ 2 .

6
= 25
+

= 34 32 
The transformation

 = # 3J 2 

J ,

3 2

4
75

1 ) 3 + 3  2 1 ) 3
4

= 3#

J  

3 2

3
25

27
625

 4

5 3

12  + 12  

leads to an equation of the form 1.3.1.46: *u  =

J ,

3 2
6
25

where

 =  2

6
25

2.

2003 by Chapman & Hall/CRC

Page 118

119

1.3. ABEL EQUATIONS OF THE SECOND KIND

67.

6
= 25
+

Denote # =

7
100

 1)

7
5

31
3

 2

1 3

 4

100
3

 and perform the transformation


7
 = 3J 2,
= 20
 + 87 35  507  2

5 3

3
10

 .

=H

21
20

=H

where

3
 :
As a result we obtain an equation of the form 1.3.4.30 with B = 17 , ( = 10

68.

10
= 49
+

Denote # = 8

VH

13
5

 2

3
10

  + 87 H 2 97 ! H + 17  2 W  = 21

1 5

7
20

 3

4 5

+ 2H .

. The transformation

= 4 U +  3 J 5
7

leads to an equation of the form 1.3.1.64 with $ = 1 49:


*  = 12 U + 39  2 15  3 U 1 J 2 .
U =  3

69.

1
16

 3J 5

5
112

49

 )

33
2 5 11
= 196
+ 286
770
3
9
= 85 2 . The transformation

Denote #

J ,

2 5

98

13 11

   8J

U =

15 3
448

 (123 1 ) 7 + 280  5 ) 7 400  2 9 ) 7 ).

11

14
11

3 11

,

7
9

J ,

7 4

11

39 2
56

=
+

Denote # = 1  . The transformation


21
100

 =

 + 4 + 75  +

35 2
3

3 2 1
50

J ,

3 4

where

leads to an equation of the form 1.3.4.30 with B = 3, ( = 21


20  :

VH

71.

784

= 3 U +  8 J
7

leads to an equation of the form 1.3.1.64 with $ = 1 49:


*  = 12 U + 39  2 15  3 U 1 J 2 .
49
98
784
70.

39 2
42

21
20

  + 4 H 2 7 !H + 3  2 W  = 34

+ 2H .

= + .
1 b . For 3, the transformation

 ,

U =$

3)

+ 1S ,

= 2( 3)$

O ' +

2
2

+

leads to an equation

*  = 2( 1) "U $IU 1 J 2 + [2 3  ( 3)2 ] $ 2 + [2 +  ( 3)2 ] $ 3 U


*

( 3)2

2 b . Let 1 and  > 1 4. Denote

(B + 2)(B + + 1)
1
1
 =
2 , where B = B 1, 2 = OA
3 .

(2B + + 3)
2
P
1 + 4

Then the transformation


1
+
2B + + 3

J .

1 2

B +2
O  +
,
B = B 1, 2
1 P

reduce the original equation to the classical EmdenFowler equation I  = *


#  2+ , where
2
2B + + 3
# = O

, which is discussed below in Section 2.3.


1 P

 = +

+2
1

+2
1

2003 by Chapman & Hall/CRC

Page 119

120
72.

FIRST-ORDER DIFFERENTIAL EQUATIONS

Denote # =

 2

+1
+
+ 2)2
2(

, $
2

+ 2)2 

2
U = ( + 2)   +

leads to the equation

+1

( 3

2(

+ 2)3 

+2

+1

.
. The transformation

' + ,

= 2( + 1) U +  + + + 2 
+2

*  = 2 ( + 1) U +  + 'U
*
( + 2)2

1 2

(see Table 7 with = 0 in Subsection 1.3.1).


73.

2  2

(  + 1) 

Particular solution:
74.

2  2

76.

= 2



+ 

Particular solution:
75.

=  

=  

sin(2

 .
+

= 2  sin(  ).

( ! + )2
!
( ! )3

2! !


=    +

= (

+ ) + 1.

= (

= Oi

5.

1
2



).

 +
,
 

= ! ( ).


=    +

 +
.
 

!$ = ( ) + 1

' leads to a Riccati equation with respect to  =  ( ):

+ )2 + 1.

The substitution  = (  +  )1 leads to an equation of the form 1.3.1.33: ! = + (  )2 .


1

+ 1.
 leads to a Bernoulli equation:   +  +  2 

The substitution =
4.

) + 2 sin(

The substitution =
  = 12  2 + ' + .

3.




+ +
.


1.3.2. Equations of the Form

2.

+ .

Particular solution:

Particular solutions:

1.

= (

+ )1 )

= 3(

3) 2 + 8

= 0.

+ 1.
2
The substitution H = (  +  )1 J

)1 )

leads to an equation of the form 1.3.1.2: ! =

+ 1.

The substitution H = 12  (  1 J 2 + 8)1 J


!  = 2 H + 1  2 H 3.
9
5184
1

+ 12 !
 H .

leads to an equation of the form 1.3.1.10 with

= 3:

2003 by Chapman & Hall/CRC

Page 120

121

1.3. ABEL EQUATIONS OF THE SECOND KIND

6.

7.

= (

2 ) 3 2 1 1 ) 3 )
3



+ 1.

The substitution =
8.

= ( 

 

  )

 = 2.
leads to a Riccati equation: 3  3 J 2 2

leads to an equation of the form 1.3.1.16: ! =


+ 1.

The transformation =
9.

+ 1.

The transformation  =  3 J 2 3 ,

+  .

 



  , =  leads to a Riccati equation: 3  =


= cosh + 1.
This is a special case of equation 1.3.3.75 with  = 0 and ( = 1.

10.

= sinh + 1.
This is a special case of equation 1.3.3.76 with  = 0 and ( = 1.

11.

cos(

+ 1.

The transformation  =

 = 2U

12.

+ ( )1 .

1
8

+  U .

sin(

+ 1.

The substitution  = +

arctan

= U

8 
16 2 +

leads to a Riccati equation:

leads to an equation of the form 1.3.2.11: ! =  cos( ) + 1.

1.3.3. Equations of the Form

!$ = 1 ( ) + 0 ( )

1.3.3-1. Preliminary remarks.


With the aid of the substitution = Xv 1 ( )  , these equations are reducible to the form:

!  = +  ( ),

where

 ( ) =  0 ( )
 1 ( ),

(1)

and by means of the substitution H = X 0 ( )  they can be reduced to the form:

!  = ( H ) + 1,

where ( H ) =  1 ( )
 0 ( ).

(2)

Specific equations of the form (1) and (2) are outlined in Subsection 1.3.1 and 1.3.2, respectively.
1.3.3-2. Solvable equations and their solutions.
1.

2.

= ( + 3 ) + 3 2 2 2 .
The substitution =  2 a + ' leads to a linear equation with respect to  =  (a ): (2a 2 + a + ( ) g =
a +  .

= (3 + ) 2 3 2 + .
The substitution =  +  2 leads to a Bernoulli equation with respect to  =  ( ):
( 2 +  + ( ) ? =  +  2 .

2003 by Chapman & Hall/CRC

Page 121

122
3.
4.

5.

6.

FIRST-ORDER DIFFERENTIAL EQUATIONS


2 = (7 + 5 ) 3 2 3 2 2 3 2 .
This is a special case of equation 1.3.3.11 with

= [(3 ) 1] + ( 1)( 3 2 ).

The transformation  = u
H , = H + 1 +  2   leads to an equation *  = + !H + H +
whose solvable cases are outlined in Subsection 1.3.1 (see Table 5).

+ ( 2 + ) + = 0.
The substitution H = 21  2 leads to an equation of the form 1.3.2.1 with respect to
!  = (2 !H +  ) + 1.

+ (1 1 ) = 2 .
Solution in parametric form:

9.

10.

11.

+ ln ),

=  (U + 

),

 *

=X

where

(1 1 ) = 2 .
Solution in parametric form:

+L .

 = 12  exp(T"U 2 ) 
8.

= ( H ):

 = (U +  *
7.

= 3
2, , = 1
2.

=T

1
2

 

V 2U 2 p
 A exp(T U 2 )W ,

where  = X exp(T"U 2 ) M

[(1 + 2 ) + ] 2 ( + ).

The transformation  = , =  +   +1 +   leads to a separable equation: 3 =

+L .

= ( ) )

+ [

(2 + 1)

+ ( + 1) 2 ]

2 1

  .


 leads to an Abel equation of the form 1.3.1.2:
   O
The substitution = 
B +1
P
!  = + (B + 1)( 2 .
= [ (2 + )  + ] 1 + ( 2 2   + ) 2 1 .
The substitution =   ( +   ) leads to a Bernoulli equation with respect to  =  ( ):
(B 2  ( ) ? =    +1 .

= [ (2 + ) 2  + (2 )]w  1 ( 2 4  + 2  + 2 ) 2 2  1 .
The transformation H =   , =  + (a +   + '  ) leads to a Riccati equation with respect
to H = H (a ):
(1)
( a 2 + 2   ( ) H g  =  , H 2 + ,!aH + , .

a 2+(
The substitution H =
 ,

order linear equation:

2g 
, where ( 0 = ( 2   , reduces equation (1) to a second-

( a 2 + ( 0 )2 g   g + (2

The transformation =

a + (( 0

+ , )a ( a 2 + ( 0 ) g  +  , 2 = 0.

(2)

, = (1 2 ) J 2 , where

)
equation 2.1.2.226:
equation (2) to the Legendre

+,

, brings

(1 2 )   2  + [ @ ( @ + 1) 2 (1 2 )1 ] = 0,
where @ is a root of the quadratic equation @

+@ +

 ,

= 0.

2003 by Chapman & Hall/CRC

Page 122

123

1.3. ABEL EQUATIONS OF THE SECOND KIND

12.

13.

14.

15.

+ 2@ 3) + 2@ + 3]

= [(

A.

= [ ( 1) + (2 + )]  1 ( + )  2 [ + ( + )] 2  1 ( + )2  3 .
1
1
S   +  (  +  )  leads to an equation of the form
The substitution = + 
Q
 (  +  )
1.3.4.5: ( +   +1 + '  )2 = [ B/  +  ( + B )  1 ] .

V ( 1) + 1W 1 =
Solution in parametric form:

1)(U +

+1

+ 1)

U
+L .
M
U + +1 + 1

=X

(1 1 ) 2 ) = 2
Solution in parametric form:

where

+ ln O

U +

+ 1)( 1).

+1

+1

+
=  V1+( U U

) W ,

1 ) 2 .

:  2
*),

= A   2 U

V ( :  )2 A : 2 W ,

U ) + L 2 n 0 (U ) for the upper sign,


U ) + L 2 ; 0 (U ) for the lower sign,
0 ( U ) and n 0 ( U ) are the Bessel functions, and 0 ( U ) and ; 0 ( U ) are the modified Bessel

functions.

19.

12

 = T 2 U

2@ + 3) + @ 2]

+ (

= [ (2 + 1) 2 + + (2 1)] 2 (, 2 4 + 3 + 2 + ^ + ) 2 ) 2 3.
Here,  ,  , ( , , and B are arbitrary numbers.
M
The substitution =   a +   +1 + '  1 leads to a Riccati equation with respect to  =  (a ):
(B/a 2 + (a + 2B  ) g =  2 + a +  .

where

18.

+ @ 1)

+ [(

17.

= #*  - +2 + +- 1  2- +  1- leads to the generalized


The transformation  =  ,
#  2+ (2  )- , which is discussed in Section 2.5.
EmdenFowler equation 3  = *

 =

16.

1 0(

1 0(

+ 3( + )2 ) 3 7 ) 3 .
1
1  +  1 J 3
+ O
leads to a separable equation for
The substitution =
P

3

2 =  1 J 3 (  +  )2 J 3 ( 1  3 3 ).
9

= 3(

3 = (7 

+ )1 )

5 ) 3

+ 6s 2 )

1 3

+ 6( s 1)

= ( ):

2 3

+ 2(  + 5 )(  + 3s + 4 ) 1 ) 3 ,
 = s(3s + 4 ).

1
2J 3
leads to an equation of the
The transformation  = ( +  ) , = ( + 4 + 3 #* )
form 1.3.4.10 with  = 1
3: [( +  ) + (4 + 3 ) ]p  = 23 2 + 2(3 + ) + 2 .

In the solutions of equations 19 and 20, the following notation is used:

 = U (U + 1) ln L  U + U + 1  ,

=1

+ 12 (6 1) 1 = 12 2 ( 1)(4 1)
Solution in parametric form:

 = U

2 2

=  (1 U

U +1
ln L  U + U + 1  .

2 2

U 2

).

2003 by Chapman & Hall/CRC

Page 123

124

FIRST-ORDER DIFFERENTIAL EQUATIONS


1 2
21 (1 + 2 2 ) = 16
(3 + 4
Solution in parametric form:

20.

 = (U
]

J ,

1 2



).

J (

1 1 1 2

= 14 !(U

4U! 2 ),

 = ( 2.

In the solutions of equations 2123, the following notation is used:

1
14

(13

20) 9 ) 7

+
=
Solution in parametric form:

21.

 = 64

8 38 8

1)(

8)

8 38

3
4,
3

= 

256
25

J 8

8 38
3

8 3 8  2 J 7 8
1 4

2
4 ,

=  (25

+7

256
25

8 28
2

3 ).

J 8

3
3

15 7
3

8 38

11 7

+7

2
4

8 38 
1 4

23 13

3 5 13 16 13
( 42
3)
4

+ 25

3
3

8 38

208

1 4 ).

{ {

U + L ) cos U ,
2 = tanh( U + L ) + tan U ,
3 = tanh( U + L ) tan U ,
4 = 3 2 3,
9
9
9
9
9 9
92
1 = cosh U sin(U + L ),
2 = sinh U + cos(U + L ),
3 = sinh U cos(U + L ),
4 = 3 2 3 2 1.
1 = cosh(

1
+ 15
(13 18) 7 ) 5 = 154 2 ( 1)( 6)
1 b . Solution in parametric form with # < 0:

{ 3 {

=  (12 2)2 J

2,

2 b . Solution in parametric form with # > 0:

 =6

9 2 9 3 9
1 2

=  (6

3,

+ 12 (5 + 1) 1 ) 2 = 2 (1
Solution in parametric form:

 =

26.

64

In the solutions of equations 2427, the following notation is used:

 = 12

25.

1
+ 26
(19 + 85) 18 ) 13 = 263 2 ( 1)( + 25)
Solution in parametric form:

 = 25

2 7 10 7
( 32
2
3

23.

11 7

5
+ 56
(23 16) 9 ) 7 = 563 2 ( 1)(25 32)
Solution in parametric form:

 = 256
25

24.

(
2

=  (4 1 )6 J

2
2 3 ,

3
14

22.

= + L sin( 3 U ),
2 = 2 L sin( 3 U ) + 3 L cos( 3 U ),
8 8  8 8
8 8
8
8 8 
8 8
8 8
= 2 1 ( 2) ( 1 ) 2 1 2 ,
4 = 2 1 ( 3 ) 5( 1 )
3 + 1 3.

= exp(3U ),

{ 2{
1

9 29

1 3)

9 5

9 5
( 13
1

J 9

3,

9 9 9

14 2 4
9 1 2 4,

2
1 2

+ 12 2 ).

).

2
2 ,

= 

3
3
2 ( 1

{ {

= 

28
9

{ 6{
1

2 5

= 

9 9  J 9


2 5
14 2
9 1 4

4
2

+ 4 2 ).

4
1

5 2
9 4

9 5

2
1 2

2 b . Solution in parametric form with # > 0:

 =

{ {

J 9 + 59 9 2 69 2 9 ).
2 3
1 3

3
+ 35
(19 14) 7 ) 5 = 354 2( 1)(9 14)
1 b . Solution in parametric form with # < 0:

{ 4{

2 5 9 5 3
( 2
2

 = 289

5
9

2
3

28
9

9 9 

14 2
9 1 4

.

2003 by Chapman & Hall/CRC

Page 124

125

1.3. ABEL EQUATIONS OF THE SECOND KIND

27.

3
+ 10
(3 + 7) 13 ) 10 = 15 2( 1)( + 9)
1 b . Solution in parametric form with # < 0:

{ 4{

 =9

2
3 ,

=  (9

8 5

4 3 10 7 5
( 32
1)
3

20

2 b . Solution in parametric form with # > 0:

9 49

2
2 4,

 = 92
]

9  J 9 J 9


9 4 3 10 7 5
4
2 2

= 4  

2
4

9 29

1 4

4
1 ).

9 

9 4
2 2

In the solutions of equations 2830, the following notation is used:

.
.

28.

= A (U

= A (U

. . 2.
2 3

= A (U

6U

+ 12 L3U 8 L

+ 27).

3U + L ,
3

45U

2
4 ,

=  (16

. . 2.
2 3

9 5

32
27

. 2.
3

3
6,
4

= A  (3

4)

9 5

J 

4 5 2 5 5
3
6
4

3
+ 14
(3 + 11) 10 ) 7 = 141 2 ( 1)( 27)
Solution in parametric form:

. 3.
4

2
6 ,

= T/ (3

4)

A 15

9 5

9 7

13 7

+ 4 L3U 3),

3 2 5
( 42
4)

3
+ 20
(13 8) 7 ) 5 = 201 2 ( 1)(27 32)
Solution in parametric form:

8 7
( 62
6

. 2.
2

T 16

. .

2
2 3 ).

2
6

T 8

. 2.
3

27
4

3
4

.

T 28

. 2.
3

A 27

3
4 ).

In the solutions of equations 3138, the following notation is used:

=X

31.

+ 20 L3U

 = T 27
]

=U

1
+ 10
(7 12) 7 ) 5 = 101 2 ( 1)( 16)
Solution in parametric form:

 =
30.

15U

 = A 16

29.

1),

U U
(incomplete elliptic integral of the second kind),
M
A (4U 3 1)

A (4U 3 1), 1 = U (2 T\L 1 + L ), 2 = U 1 ( $ 1 1),

3 = 4U 12 T 22 ,
21 ( + 1) 7 ) 4

4 = 2 3 8 12 ,
5 = 2 (
1 2
( 1)(3 + 5) 5 ) 2 .

+ L ) U 2.

= 4
Solution in parametric form:

 = 16 U

3 1
1

= 13  

1
6

U 

J V (11U 3 2) 1 2
1
2 W .

3 4 1 4
1

The lower sign is taken in the notation adopted.


32.

21 ( + 1) 7 ) 4 = 14 2 ( 1)( + 5)
Solution in parametric form:

 = 31 U

2 2
1 2,

5 2

= 13  (U 1 )1 J 2 

3 4
1
(2
3 2

U 22 + 2 3U

2 2
1 ).

The lower sign is taken in the notation adopted.

2003 by Chapman & Hall/CRC

Page 125

126
33.

FIRST-ORDER DIFFERENTIAL EQUATIONS


1
14
(4 + 3) 8 ) 7 = 141 2 ( 1)(16 + 5)
Solution in parametric form:


+ 16 (13 3) 2 ) 3

3 2 1 2
16 1 2 3 ,

 =A

34.

1
28
(8 1) 8 ) 7 =
Solution in parametric form:

 = TI

36.

1
28

2(

2 3 1
1 3 4 ,

1 1
(
6

( + L )2 5 ,

=A

2
1 2 ).

A 7 42 A 32

2
1 4 ).

[(1 A 2U 3 )( + L ) U 2 ].

 ( + L )4

576
125

T 16

2
2 3

42 A 4 12 4 ).
A

3 2
5 [ 5

1
+ 42
(39 4) 9 ) 7 = 421 2 ( 1)(9 16)
Solution in parametric form:

11 7

+ 5U

5 TCU 3 (

+ L )2 ].

= 13  [16U 3 ( + L )2 55 ]2 J 7 [3

2
5 ,

A 7

9 7

2
5

+ 7U

5T

48U 3 ( + L )2 ].

In the solutions of equations 3941, the following notation is used:

+ ( 2) 1 = 2 2 ( 1)
Solution in parametric form:

+ (1 2 ) = 2 1 .
Solution in parametric form:

= 2UIA exp(T"U 2 ).

+ (3 2) 1 = 2 2 ( 1)2
Solution in parametric form:

 = 12 exp(T"U 2 ) 

41.

= 36  ( + L )2

=  V 1 T 2U

 = U exp(T U 2 )  ,
40.

3 12 3 6 1 7
(3 33
32 1 3 4

1
= 32
s

 = X exp(T U 2 ) U + L ,
M

39.

52 (3 10) 4 = 15 2 ( 1)(8 5)
Solution in parametric form:
5
24

1 3

1)(32 + 3)

+ L )1 ,

8 3 3
( 3
3 4

 = A 16U 3 ( + L )2
]

= 

(5 4) 4 = 2 ( 1)(3 1)
Solution in parametric form:

 =A
38.

3 2
3 4 ,

 =A
37.

9 7

13 12 6 2 1 7
(3 32
16 1 2 3

1
= 16
 

= 16 2 ( 1)(5 1)
Solution in parametric form:

 = T[

35.

U exp(T"U 2 ) 

W.

.
= 12  V 2 exp(T"U 2 ) 

Te

2 2

W.


 =
T 2


=
T 8

1 1

V exp(T U 2 ) 2  2 W .

2003 by Chapman & Hall/CRC

Page 126

127

1.3. ABEL EQUATIONS OF THE SECOND KIND

In the solutions of equations 4252, the following notation is used:

=U

=U

A (4 <
A (4 <

3
3

<

1) + 3U 2 <IT 1,

1) + 2 < ,

=U

= U 2 <pT 1,

A (4 <

A (4 <

1) 4U 2 <CA 6,

= < (U ) is defined implicitly as follows: U = X

1) A 2U'< 2 ,

=U

A (4 <

1) < .

<

L . The upper sign


1)

in the above relations corresponds to the classical Weierstrass elliptic


function < = < (U + L , 0, 1).

Here, the function

42.

1
4

(3

4)

5 2

1
4

2(

1)( + 2)

= 12 U 

3 1 2
(
6

A (4 <

Solution in parametric form:

<

2 2

 = 23 U

6,

2
3

+ 2 </

2
6

3U

<

2 3

).

The upper sign is taken in the notation adopted.


43.

(33 + 2)

1
30

1 2
= 30
( 1)(9 4)

6 5

7 5

Solution in parametric form:

44.

1
8

 = 4U 2 < 3

2
6 ,

8)

= 18 2 ( 1)(3 4)

5 2

=A

1
3

 (4U 2 < 3

8 1 5
(3
6)
4

+ 5 </

2
6

T 12U 2 < 3 ).

Solution in parametric form:

 = 43 1

2
2 ,

= 14  

4
3

3 1 2
(
1

2
1

+ 3

3
2 ).

The lower sign is taken in the notation adopted.


45.

(17 + 18)

1
30

22 15

1 2
= 30
( 1)( + 4)

29 15

Solution in parametric form:


2
1

 = A 4
46.

1
13

(6

13)

3
2,

5 2

= A !

16 15
(4
1

2 )7 J 15 (

1 2
= 26
( 1)( 13)

2
1

3
2 ).

T 4

Solution in parametric form:

 =

13
6

2
2

5,

1
= 13
 

13
6

3 1 2
(2
5

2
5

A 13 2

3
2 ).

The upper sign is taken in the notation adopted.


47.

1
30

(24 + 11)

27 20

1 2
= 60
( 1)(9 + 1)

17 10

Solution in parametric form:

 = 4

3
2

2
5 ,

= 13  (4

3 7 20
2)

13 10
(3
5

2
5

+ 20

12

3
2 ).

The upper sign is taken in the notation adopted.

2003 by Chapman & Hall/CRC

Page 127

128
48.

FIRST-ORDER DIFFERENTIAL EQUATIONS


52 (3 + 2) 8 ) 5 =
Solution in parametric form:

1
3

 =T

49.

<

2
3

51.

1
5

2 (

11 5

= T/ (3 </

3 5
(3
4

9 5

2 5
2

= !

2
2

 = 2

3
4 ,
= 15 2 (

51 ( + 4) 8 ) 5
Solution in parametric form:

<

1 3
3

3 5
3

A (4 <

1),

< 2 A
Q

(4 <

1) S

3 5

(2 1) 5 ) 2 = 21 2 ( 1)(3 + 1)
Solution in parametric form:

 = 16 U

<

1 2

4<

=  V 6U 

1,

4<

2
3

3
4

2
4

AC 4 ).

2
3

2
2 ).

10

14 < 3

2
3

2
2 ).

+ 4

+ 2 A (4 <

1) S .

3 1 2

<

4<

The upper sign is taken in the notation adopted.

TB< 2

1

3
4

3 5

</

9 5
(
4

9 2 5
(2
4)
11 5

1)(3 + 7)

2 5 1
3
3

= 16 !

1
2

=  (4

2 2 5
3)

1)(27 + 8)

3
4 ,

1)(8 + 1)

3
+ 10
(13 3) 4 ) 5 = 101 2 ( 1)(27 7)
Solution in parametric form:

 = 13

52.

2 (

4,

56 (4 + 1) 7 ) 5 =
Solution in parametric form:

 = 2

50.

2
3

1
5

1 + 2U'<

2U  .

1S ,

In the solutions of equations 5355, the following notation is used:

53.

54.

6) 7 ) 5

+ L3U 1,

+
=
Solution in parametric form:
1
5

1
5

(21

2
5

2
2

 = A 3U

+ 19) 7 ) 5

(
2

1
3 ,

2
2

1)(

25

2
3 ,

 =

2
1

+ 4) 9 ) 5 .

= A 

2
3,

A 1,

= 

=U

A 3U + L .

3 5
3

4) 9 ) 5 .

2 5
1

5 2

J V (1 A 5U 2 )

2 2 5
2)

1)(9

3 7 ) 4 = 14 2 ( 1)( 9)
Solution in parametric form:

=U

=  (3U

+
=
Solution in parametric form:

 = A/

55.

= A"U

4 5
2

6 5
(
3

2
3

Te

T 3U

2
2

2
2

W.

2
1 ).

1 2
1

3 2
(
3

2
3

2
2

2
1 ).

The lower sign is taken in the notation adopted.

56.

In the solutions of equations 56 and 57, the following notation is used:

V ( + 1) 1W

%  =X U 

exp(T"U 2 ) U + L .

= ( + 1)(
2

Solution in parametric form:


2 2
 =
U  +1 exp(AU 2 ) % : ,
, +1
, 1
.
where =

=

1)

, +1
2

U 

2
+1

exp(T"U 2 ) % 1
: A ( , + 1)U

2003 by Chapman & Hall/CRC

Page 128

129

1.3. ABEL EQUATIONS OF THE SECOND KIND

57.

[( 2) + 2 3]
Solution in parametric form:

2
2

 = T 2U
58.

21 [(4 7) 4 + 5]
Solution in parametric form:

)2

1
2

)3 J 5

: 
*

= U

(2 3)( 1)2

= 12  (U

J (2

7 5

+@

, @ =

2(1 )
2 exp(

= L 1  (U ) + L 2 ; (U ),
where
modified Bessel functions.

12

=  (T 2 %  )1  exp V T ( , 2)U 2 W U

 exp(AU 2 ) %  ,

 = (U

2)( 1)2

2(

12

T"U 2 )+ O

2
A 4U 2 %  S .
2,
P

+5 U

),

1,
;  (U ) and
3 2,

; (U

) are the

In the solutions of equations 59 and 60, the following notation is used:

U
M
+L .
U  +

 =X
59.

V (2 1) lW 1 =
1 b . Solution in parametric form:

60.

= (   )  .

V ( + 1) W 1 ( )
1 b . Solution in parametric form:

61.

.
1

V (2 3) + 1W  =
Solution in parametric form:

2(

[(U  +  )  U ].

2 (

=   (   ) 

= (U !  ) 

 = U  1 ,
2 b . Particular solution:

= U   

 = U  1 ,
2 b . Particular solution:

)2

[U (U  +  )  ].

2)[( 1) + 1]

2(1 )

1
U +1
 2 (1 , )(2 , 3)(U + 1) 1  + (2 , ) uS ,
(1 , )(2 , )
Q
2 

U (1 , )(U + 1) 1  S ,
=
1 ,Q

 =

where

(U + 1) 1 

U +L .

In the solutions of equations 6266, the following notation is used:

 = 1 AIU 
62.

=X

+1 ,

 = X 1 AhU 

+1 1 2

V ( + 2 3) + 3 2 /W
Solution in parametric form:

 =U

  2  ,

U +L ,

= U 

  =   U ,
2

V ( + 1)

B +1 
U
1  !A
2 ,

+1

 = X 1 AhU 

+1

 

1
2

( + 2 3) + 2W

  2 + U 

2(1 )
2

U +L .
12

2003 by Chapman & Hall/CRC

Page 129

130
63.

FIRST-ORDER DIFFERENTIAL EQUATIONS

V ( + 2) 2W

2 +1

Solution in parametric form:

= A 2 

 = A 2 1  ,
64.

2 +1
+4
2 =
P
+2
Solution in parametric form:

65.

B +2
.
B +1

3 + 5
+
+3
2

66.

pO

2   2 ,

+2

+ 2)

+1
+3

= U

+4
+3

68.

69.

= ( 

+ ) + 

= [ (2
+ )-



The transformation  = ln ,

+ ]

  

= !( 

3 +2

+ 4) ( 1)( + 2)W

1
 2 U   + U 
B +2
'B

2( +2)
+3

2
+3

 2U 
Q 
P

+1

+ 2 + 5
+
+1

2 + (A 1) 

1
+3

+1

= ( 

+ ) + [



 

 

4
S
+1

B +3
B +1

D

(2

+ 2 + ) +

+5
+3

U   S .

J O 2 + 1 U .
 1  B P
1

+ (

C 2



+ )

+ 1) 

+ ) 

+ ).

 2D

].

+ ' leads to an equation of the form 1.3.1.2: ! =

(2

g = a +  .
= a +  leads to a linear equation: (a 2 + a + ( )2

2 2

3 +2

2 ,
P

The substitution =  ( +  ) leads to a linear equation with respect to 


( 2 + ' ( )  =  + .
71.

The substitution H =  leads to an equation of the form 1.3.3.10: ! = [  (2 + )H  +  ] H 


(  2 H 2   H! + ( ) H 2  1 .

The substitution =
70.

( + 1)

 = (U +1 J1 ,

+ (

2( + 3) Q

 

2 O + 1 +

Solution in parametric form:

67.

1+

1
+1P

Solution in parametric form:


+1

V 2

2 + 1W

1
 +1

!     (  A 2  ).

+1

=  (2  )      

 =U 

 = 2 'U  2   ,
where  = A

V ( + 1)

+ ( .

=  ( ):

+ ).


The substitution =  ( +  2 ) leads to a Riccati equation with respect to  =  ( ):

2
2
 + ' + ( ) =  + .
( 

2003 by Chapman & Hall/CRC

Page 130

131

1.3. ABEL EQUATIONS OF THE SECOND KIND

72.

) 

+ (1 + 2

2  2

Solution in parametric form:

 =

where  = X U
73.

exp(A"U 2 )  ,

2

V 2U 2 exp(A"U 2 ) pA 1W exp[2 exp(AU 2 )  ],

exp(T"U 2 ) U + L .

V 1 + 2 + 2 ( + 1) WE

( +1)

+ 1)(1 +

)  2(

+1)

Solution in parametric form:

 = O 2B/U  +

= X (1 AIU 

where
74.

1
exp V (B + 1)U  W ,
B +1P

+ (1 + 2

) J

+1 1 2

U
+ B
exp V (B + 1)U  W ,
1 AIU 
P

= U  O

U +L .

1 ) 2) exp(2 1 ) 2)

3 ) 2 exp(4 1 ) 2 ).

Solution in parametric form:


4

 = (U

2,

= !(U

ACU

) exp(2 'U

).

Here,

 = (T ( )1 J 2 ,

: 
*

=U

+ ,

U ) + L 2 n 1 (U ) for the upper sign,


U ) + L 2 ; 1 (U ) for the lower sign,

1 1(

1 1(

where 1 (U ) and n 1 (U ) are the Bessel functions, and 1 (U ) and ;


functions.
75.

= ( cosh + ) sinh + .

The transformation a =  sinh  , = 


76.

= ( sinh + ) cosh + .

The transformation a =  cosh  , = 


77.

cos(

leads to a Riccati equation: 2( 'a + ( ) g  =  2 +2a +  .

= ( + ) 

 = + .
leads to a linear equation: ( 2 +  +  )2

sin(

, = (H +

)

leads to a Riccati equation: (H 2 + !H +  )2 =

+H .

) + .

The substitution H = 12 
80.

leads to a Riccati equation: 2( 'a + ( ) g  =  2 +2a  .

= (2 ln2 + 2 ln + ) + ( ln4 ln2 + ).

The transformation  = 
79.

U ) are the modified Bessel

= (2 ln + + 1) + ( ln2 ln + ).

The transformation  = 
78.

1(

leads to an Abel equation of the form 1.3.2.11: !" =  cos(2 H ) +1.

) + .

The substitution H = 12 

leads to an Abel equation of the form 1.3.2.12: !" =  sin(2 H ) +1.

2003 by Chapman & Hall/CRC

Page 131

132

FIRST-ORDER DIFFERENTIAL EQUATIONS

1.3.4. Equations of the Form [ 1 ( ) + 0 ( )] = 2 ( )

2+

1 ( ) + 0 ( )

1.3.4-1. Preliminary remarks.


With the aid of the substitution

=O

1P

= exp O X

where

2
1

 ,
M P

(1)

these equations are reducible to a simpler form:

*  =  ( ) +  ( ),
1
0
where

= M

0 2
,
12


(2)
0
1

0  1 02  2
+ 3
12
1

Specific Abel equations of the form (2) are outlined in 1.3.11.3.3. In the degenerate cases with
 0 0 or  1 0, the variables in equation (2) are separable.
1.3.4-2. Solvable equations and their solutions.
1.
2.

(  + ( + ) + ( +
Solution: # 2 + ,! 2 + 2( $I

+
+

= 0.
+ ' ) = L .

( + + ) =  +  + .
   leads to the equation
The substitution = 

"  = (  + ) + ( ! ) + = 

which is separable with  = . For  , the substitution = (  + ) leads to an equation


of the form 1.3.1.1 or 1.3.1.2:

*  = +

3.

4.

5.

6.

( ! ) +

(=  ),

where

= + .

( + 2 + + ) = 2 + 2 + x + ( + ) + s.
The substitution = H + ,! leads to a Riccati equation with respect to  =  ( H ):
( +

[ !H

+ ) + 
2,
 +
Solution: 2 +
+1

+(

1
+1

, ) H + (, ]  =  ,!
+

+ 2( #* 

+
+

+ (  + , ) + H + ( .

= 0.
+ ' ) = L .

( + +1 + ) = ( + 1 ) .
The substitution =   (  ) leads to a Bernoulli equation with respect to  =  ( ):
  2.
[B 2 + ( 'B + ( ) ( ] ? =  + 

+ s.


The transformation =  , =   leads to an equation *3  = + #* + $I + , where

# =   2 , $ =   2 ( , = (  B )
 (see Subsection 1.3.1).

7.

= 2 + (2 + 1) + 2 2 + .
The substitution = +  leads to a Bernoulli equation with respect to 
(B 2 +  + ( ) ? =  +  2 .

=  ( ):

2003 by Chapman & Hall/CRC

Page 132

133

1.3. ABEL EQUATIONS OF THE SECOND KIND

8.

9.

10.

11.

2 = (1 ) 2 + [ (2 + 1) + 2 1] 2 2 .
The transformation  = 2 , = a +  2 + 1 leads to a Riccati equation: (B/a 2 + 2 B  ) g  =
 2 + a + 1.
(  ]I + ( (m ) = =] 2 + F + ( ( F} ) .
The transformation  = + , , = leads to a linear equation with respect to
[( L # ) 2 + c\ ]2  = #* + $ .

[(3 + s) + (4 + 3s) ] = 2 2 + 2(3 + s) + 2 .


The substitution =  2 + (3 + ) +  leads to an Abel equation of the form 1.3.3.3:
2*2 = (7  + 5  ) 3  2 3 2 ( 2 3  2 , where  = + 2 , ( = 12  (13 + 6 ).

[(4

+ s) + (4 + 3s) ]

The substitution = 34 
2*2 = (7  + 5  ) 3 

12.

(2 

14.

15.
16.

17.

+ +
The substitution
[ #2H

13.

19.

3
2

+ 2(3 + s) + 2 .

+ (3 + ) +  leads to an Abel equation of the form 1.3.3.3:


2 ( 2 3  2 , where  = + 2 , ( = 18  (60 + 25 ).

2 3

+ ) =  2 + ] 2 2 + x + ( + ) + s.
=H +!
,  leads to a Riccati equation with respect to  =  (H ):
+(

 , ) H + (, ]  = 2 #3,!

+ (2 #2H +  , +  ) + !H + ( .

1
+ 1)
1 + .
2

+
S =
Q
+3
2
+3
1

1
2+
+  leads to an equation of the form 1.3.3.4:
The substitution =
2
+3
2

+ (1

2(

) 

*  = [(3 ) 1] + ( 1)(


),

where

 = # 2( + 1)( + 3)2 .

(2 + ) = (2 ) 2 + (1 ) + 2 +  +2 .
The transformation H = 
 , = #* + +  H 2 +  H ( leads to a separable equation:

*  = (2 !H +  )( H 2 +  H ( ).

+ )
= +

+ .
Solution: ( + ) +  +  = L ( ' 
2

)2 .

(2  + ( 2 + ) =  2 + ( ] + ( ) 2 + .
= H + ,! leads to a Riccati equation with respect to 
The substitution

H  + .
( #2H 2 + ( , )  = (2 #3, + $ ) 2 + 2 #2

( + ( 2 +
The substitution

=  ( H ):

= F 2 + G + H 2 + I .
= "H leads to a linear equation with respect to  =  (H ):
[( c

18.

= ( ):

# )H

+ ( $ ) H +  ]  = ( #2H + $ ) + ,

(  + ( 2 + ) =  2 + ( + ( ] + ) + (m + .
This is a special case of equation 1.3.4.22. Solutions: = L3  and #

(2  + ( 2 + )
=
The substitution
[ #* 2 + ( L $ ) + c ]

+ $I + , = 0.

+ = + F 2 + I =B I 2 ' .


  + leads to a linear equation with respect to 
 = (2 #* + $ ) + 2#* + , .
=



=  ( ):

2003 by Chapman & Hall/CRC

Page 133

134
20.

21.

22.

FIRST-ORDER DIFFERENTIAL EQUATIONS


(  + ( 2 + ) =  2 + = + F 2 + ( I ) =B ' .
The substitution =  + leads to a linear equation with respect to  =  ( ): [( L $ ) + c ]  =
( #* + $ ) + #* + , .

(  + ]I + ( 2 + (m ) = =] 2 + F + ( F ( ) .
The transformation  = , , = leads to a linear equation with respect to
[( L # ) 2 + ( c $ ) ]2  = ( #* + $ ) , $ .
( + ( 2 +
Jacobi equation.

1 )



= ( ):

2.

1 b . With the help of the transformation  =  + , = + , where and are the parameters
which are determined by solving the algebraic system
2

#2 + $C

+  1 +  1 + (

#*

= 0,

+ $C +  2 +  2 + (

= 0,

we obtain the equation

( #  + $  2 +  1  +  1 )  = # 2 + $  +  2  +  2 ,


where  1 = 2 $C + #* +  1,  2 = $I +  2 ,  1 = #2 +  1 ,  2 = 2 #* + $C +  2 . The transformation


H =  , = 1
 leads to a linear equation: 

[  1 H

+ ( 1  2 ) H  2 ]  = (  1 H +  1 ) + #2H + $ .

2 b . The original equation can be also rewritten in the form


(

  )(B 3  +

+ , 3 )   (B 1  +

+ , 1) + B 2 +

+,

= 0.

The solution of this equation in parametric form can be obtained from the solution of the
following system of constant coefficient linear differential equations:

using the formulas  (a ) = 

23.

(

+(

2
+
+

24.

25.

+ , 1 3,
2  2 + , 2 3,
3  2 + , 3 3,

and (a ) = 

1 2

 .
2
3

 , ) H + (, ]  = ( #3, + $ )

+ ( #2H +  , +  ) + !H + ( .

(2  + ( 2 + + + ) =  2 + ( ] + ( ) 2 + I + + s.
The substitution = H + ,! leads to a Riccati equation with respect to  =  ( H ):

+ (, )  = (2 #3, + $ )

( #2H

+(

+ (2 #2H +  , +  ) + !H + ( .

(2  ] 2 + + + ) =  2 + x + ( + ) + s.
,  leads to a Riccati equation with respect to  =  (H ):
The substitution = H + !
[ #2H

26.

1+

1+
1

+ ) = / + /( 2 + x + ( + ) + s.
= H + ,! leads to a Riccati equation with respect to  =  ( H ):

The substitution
[(

1 

(  1 ) g = B 1 
(  2 ) g = B 2 
(  3 ) g = B 3 

 , ) H + (, ]  = #3,!

+ (2 #2H +  , +  ) + !H + ( .

(2  + ( 2 + + ) =  2 + ( ] + ( ) 2 + x + s.
The substitution = H + ,! leads to a Riccati equation with respect to  =  ( H ):
[ #2H

+(

 , ) H + , ]  = (2 #3, + $ )

+ 2 #2H + !H + ( .

2003 by Chapman & Hall/CRC

Page 134

135

1.3. ABEL EQUATIONS OF THE SECOND KIND

27.

(2 

+(

28.

+(

[

+ ) =



( )

+ ( ] +

+ s.

= H + ,! leads to a Riccati equation with respect to  =  ( H ):

The substitution
[ #2H

+ (   , ) H + (, ]  = (2 #3, + $ )

+ ( 1) 

( ]

+ ( ) ]

= 

+ (2 #2H +  , +  ) + !H + ( .

+(

( ] + (

+ ( 1) (m

This is a special case of equation 1.3.4.22. Solution in parametric form:

a + #3L3a 
,
a +L

 =

'a $L3a 
.
a +L

The solution can be presented in implicit form as well:

L  (#

+ ) + (1 )

29.

[(

30.

[( + ) + ( + 1)

+ $I )  + [ # (  ) + $ (   )] 

(2 + 1) +

1.3.4.8:


2 H * = (1 B )

32.

(

+ 2

' ) = 0.

The substitution =  +  leads to an equation of the form 1.3.4.8:


where  = (2B 1)  ( .
2!*  = (1 B ) 2 + [  (2B + 1) + 2B 1]  2 B 2  B ,


The transformation H =

31.

+ (2 1) ] = 2

(2

3B 1 1
,
B 1

+ 3)

3 1

+ 0 ) +

+ 2

H B ,

+ 2) +

0 ]

 =

(3B 1) ( + B (2B + 1)


.
B 1

The transformation H =  , = ( + +1 +  ( + 1)

equation: *3 = ( + 1)2 (2 !H +  )( !H 2 + H ).

[(

3B 1 
B
leads to an equation of the form
+
B 1
B 1

+ [  (2B + 1) H + 2B 1]  2
B H

+ ) = (

2 2

+ (

+ 1)

leads to a separable

0.

This is a Riccati equation with respect to  =  ( ). The substitution  =

leads to a second-order linear equation:

 2   [(  2 )  +  1  2 ]  +  0  22 = 0,



33.

[(12

2 2

+ 1) + 4

= 2 (3

2 2

! + 
;
+( 1 +( 0

+(
2

+

+(
2

2


= 1, 2, 3.

+ 2 ^ + 3 2 ).

The substitution =  (3  2 2 + 2 ( + 3  2 ) leads to an Abel equation of the form 1.3.3.3:


2*2 = (7  + 5  ) 3  2 3 2 ( 2 3  2 .

[(

34.

1)( 

Solution: #*
35.

 =

where

(2

+ ( ) +
+ c\

+ ) = 4

(F

+ G + H )]
= [  (1 ) ( ]
2

+ p + $

+  = L3  + .

The transformation H =  , = 2


*2 = 2(2 !H +  )(2!H 2 + 2 H , ).

2 2

+ .

+ 2 '

+ [ F (2 )

(

(1 ) H

leads to a separable equation:

2003 by Chapman & Hall/CRC

Page 135

136
36.
37.

38.

FIRST-ORDER DIFFERENTIAL EQUATIONS


( + + 2 ) = 2 + + .
The transformation a = 
 , H =   2 leads to a linear equation: ( ( a ) H g  = (B 2)(!H + a +  ).
(2 + ) = (3 + )w 2 (2 + ) + 
 H
The transformation H =   , = #*  + + + (B + )( !
(


*

2
2
= (B + ) (2 !H +  ) O !H + H
.
equation:

B + P

w
2

.
+ H ) ( leads to a separable

= 2 + (2 + 1) + 2 J 2  + .
The transformation  = ln , = +  leads to a Bernoulli equation with respect to = ( ):
(B 2 +  + ( ) ?  = +  2 .

1.3.5. Some Types of First- and Second-Order Equations Reducible


to Abel Equations of the Second Kind
]

Notation:  , , % ,  , Y , j , ` ,  , and 4 are arbitrary functions of their arguments.


1.3.5-1. Quasi-homogeneous equations.

1 b . Let us consider a quasi-homogeneous equation of the form

 ( )

+1

  + ( ) + #*  = 0.

In the special case = 0 this equation is homogeneous.


, = #*  + ( H ) @ H! ( H ) leads to an Abel equation:
The transformation H = 

*  = [( + @ )  +  @ H!  ] + " ( @ H! ).

2 b . A quasi-homogeneous equation of the form

 ( )

  + ( ) +    [ % ( ) +1   +  ( )] = 0
, =    to an Abel equation: 
can be reduced by the transformation H = 
, [ (H ) @ H! ( H )] +  (H ) @ H!% (H ) .  = " (H ) 2 + "% (H ) .
+1

1.3.5-2. Equations of the theory of chemical reactors and the combustion theory.
In the theory of chemical reactors and the combustion theory, one encounters equations of the form

       =  ( ).
The substitution ( ) =  
 leads to the Abel equation *  = 


are given in Subsection 1.3.1.

 ( ), whose solvable cases

1.3.5-3. Equations of the theory of nonlinear oscillations.


1 b . Let us consider equations of the theory of nonlinear oscillations of the form

    + Y ( )   + = 0.

The substitution H ( ) =   leads to the Abel equation

HH  + Y ( ) H + = 0,

which is reduced, with the aid of the substitution U = 12 (  2 ), to the following form:
Yu A  2U 
HH  = (U ) H + 1, where (U ) = A
.
 2U
*

(1)

(2)

Specific cases of equation (2) are outlined in Subsection 1.3.2.

2003 by Chapman & Hall/CRC

Page 136

137

1.3. ABEL EQUATIONS OF THE SECOND KIND

2 b . An equation of the theory of nonlinear oscillations of the form

    + ` (   ) + = 0
can be reduced by the transformation H =  , = ` (  ) to an Abel equation of the form (1):
*  + `  ( H ) + H = 0.
1.3.5-4. Second-order homogeneous equations of various types.
1 b . A homogeneous equation with respect to the independent variable has the form

    =  ( )   +  ( ).
The substitution ( ) =    leads to an Abel equation: *3 = [ ( ) + 1] +  ( ).

2 b . A generalized homogeneous equation
    = (   )   +   1  (   )


can be reduced by the transformation a =   , =   (  + ,

) to an Abel equation:

" g = [ (a ) + 2 , + 1] +  (a ) ,!a (a ) , ( , + 1)a .


To the EmdenFowler equation, discussed in Section 2.3, there correspond (a ) = 0,  (a ) = #*a + ,
and , = +  +2
1 .
3 b . A generalized homogeneous equation


    =  9 : O
can be reduced by the transformation =


P

+ 

  , =  9

+2

4O

:


P

to an Abel equation:

[  ( ) + 4 ( ) + 2 ] K = [( 1) + + 2] .
To the generalized EmdenFowler equation, discussed in Section 2.5, there correspond = B 0 ,
= + 0 ,  ( ) = #* - , and 4 ( ) = 0.

1.3.5-5. Second-order equations invariant under some transformations.


1 b . An equation invariant under dilatationtranslation transformation has the form


The transformation

    =  9  :   (   ) +  2 (   ).
=    , =   9 +2  :   leads
 equation:
 to an Abel

[  ( ) + ( ) + ] = ( + + 2) .

2 b . An equation invariant under translationdilatation transformation has the form

    =  9

The transformation =   
, =  9 :


 :
*O
1

+ rO


P

leads to an Abel equation:

[  ( ) + ( ) 2 ]  = [( 1) + ] .

2003 by Chapman & Hall/CRC

Page 137

138

FIRST-ORDER DIFFERENTIAL EQUATIONS

1.4. Equations Containing Polynomial Functions of


1.4.1. Abel Equations of the First Kind
 = 3 ( ) 3 + 2 ( ) 2 + 1 ( ) + 0 ( )
1.4.1-1. Preliminary remarks.
1 b . If 0 = 0 ( ) is a particular solution of the equation in question, the substitution
reduces it to an Abel equation of the second kind:

*  = (3 

2
3 0

+ 2

2 0

+  1 )

(3 

= 1


+  2 )  3 ,

3 0

which is discussed in Section 1.3. For  0 ( ) 0, we can choose

0 as a particular solution.

2 b . The transformation
2

= X  3

 ,

= O

3

 2
 S ,
= exp X O 1 2
Q
3  3 PuM

where

brings the original equation to the normal form:

 =

+ ` ( ),

` =

where

 3

 0+


1
M O
3 

M

2
3

 1  2 2  23
S .
+
3 3
27  32

1.4.1-2. Solvable equations and their solutions.


1.

2.

3.

3 ) 2 .

This is a special case of equation 1.4.1.9 with B = 1


2.
+ 3

+ (

6.

+ .

+ ) 2 .

= 1
leads to an Abel equation of the form 1.3.2.1: "  = (  +  ) + 1.

+ (

+ )2 2 .

= 1
leads to an Abel equation of the form 1.3.2.2: "  = (  +  )2 +1.

The substitution
5.

The substitution = 1
+  leads to an Abel equation of the form 1.3.2.1: "  = 3  + 1.

The substitution
4.

+ (

+ )1 )

2 2

The substitution = 1
leads to an Abel equation of the form 1.3.2.4: "  = (  +  )1 J 2 +1.
+ 3

3 3

This is a special case of equation 1.4.1.10 with B = 0 and


7.

+ 3

= 1.

leads to a separable equation:   = 

The substitution = 
8.

3 4

+ '

+ .

This is a special case of equation 1.4.1.10 with B =

= 1.

2003 by Chapman & Hall/CRC

Page 138

1.4. EQUATIONS CONTAINING POLYNOMIAL FUNCTIONS OF

9.

11.

139

= 2 +1 3 + 2 .
The substitution =   +1 leads to a separable equation:  3 =  3 + (B + 1) +  .
For  = 13 (B + 1) # 2 and  = 23 # (B + 1), the solution is written in parametric form:

 = exp O
10.

B +1P

= #O 1 +

exp(  ),

U*P

 = U 13 ln |U + 13 | + L .

where

= 3 + 3 + 2 1 2 3 +3 .
The substitution = + ' + leads to a Bernoulli equation:  =  

= 3 + 3 + 2 +  2 3 +3 + ^ + 
The substitution = + ' + leads to a Bernoulli equation:  =  

9 = (

13.

= 4 3 + ( 2 1) + .
The substitution =  leads to a separable equation: p =  ( 

15.
16.
17.

18.

19.

20.

3   2  

+2

+ .

.
+( (  3   2  

+2

+ ) .

(9 + 2 + 9 w 1 )( 1 + )  2 .
1
1
3

, the substitution = O +
For =
  1 + +    leads to the Abel
3  (1 )
 + ' + P
equation *2 = +  + ' + , which is discussed in Subsection 1.3.1.

12.

14.

+ )2 

+1 3

= 3 + 3 2 2 3 3 .
 = 
The substitution = + '  leads to a Bernoulli equation: u

= 2 +1 3 + ( ) + 1 .
 = 
The substitution =   leads to a separable equation: 3

= +2 3 + (
The substitution = 

1) + 1 .
 =  
leads to a separable equation: p

= 3 3 2 4 + 2 3 6 + 2 3.
  2 + 1

The transformation  = 1
, = 
*2  = 3  2 + 1.

+  + ( ).
1

3   2  2 

+ +( .

(

+  + ( ).

leads to an equation of the form 1.3.2.2:

= ( + ) 3 + 2 .
The substitution = 1
 leads to an equation *u = +  + ' + , which is discussed in
Subsection 1.3.1.

= (  2 + ( + = )1 )
The substitution = 1 

( #* 2 + $I + L )1 J 2 .

+ 2.
leads to an Abel equation of the form 1.3.1.63: *I =

2 3

6

= 16 ) 9  25

Solution in parametric form:

34 9 3

6
U
25 

 =A
where
The function

<

=T

<

2
27

 9

125 
( !
108

= U 2 <pT 1,

) 

2
25

1)

25 18 7 9 7 18

11 18

J U J < J

=U

A (4 <

= < (U ) is defined implicitly by U = X

6 61 18
.
25

1
1

1
2 (18

</ 1 A 5

2 ),

1) + 2 < .

A (4 <

<

1)

the formulas corresponds to the classical Weierstrass elliptic


function
3

L . The upper sign in

<

= < (U + L , 0, 1).

2003 by Chapman & Hall/CRC

Page 139

140
21.

FIRST-ORDER DIFFERENTIAL EQUATIONS

+ 

+ 3


3   + 1.

 

= 13

2 3  3

1
= +  

 

2 2

The substitution

23.


leads to an Abel equation of the form 1.3.2.7: *I =   + 1.

= 1


The substitution
22.

1 2

 

2
3

+ 

*p =

leads to an Abel equation of the form 1.3.2.7:

Solution in parametric form:

 =
24.

 2



25.

= 





+ 3 
+ 3 



+ 3 

The substitution =
29.

= 

= [ + exp(2

+ 3 

= 23

"  = + (6  )1 .

33.

+ 

 

( + )

+^
2 
3

+ '

 

( + )


+  

 

2)


C

+ 2

  

2 ( +2 )


C

+
2

+ [(3 2 + ) 

+ ( ( + ) + .

+ ( ( 3   2   ) .

  

3 ( +3 )

 

( +2 )

3   2  ( 

  2  ( 

 

+2 )


leads to a Bernoulli equation:  =  


+  

 )]

  

3 ( +3 )


leads to a Bernoulli equation:  =  

.

leads to a Bernoulli equation:  =   




+  

( + )

exp(2

The substitution

32.

+ s] + ( 2 + ) 

 

+2 )

+ V 3   2  ( 

 

( +3 )


leads to a Bernoulli equation:  =    3 + [ (  ( 

  W

+2 )

(s
) 

 

+2 )

.
.

+ ] .

+ 2.

= 1
leads to an equation of the form 1.3.1.8: "  =   exp(2
 ).

The substitution
31.

 = U 13 ln |U + 13 | + L .

where

leads to a separable equation:  = 

 

The substitution =
30.

 ,

 

+ ^ +

( + )

The substitution =
28.

U P


+  leads to a Bernoulli equation:  =  

The substitution =
27.

+ 3 

The substitution =
26.

=  

The substitution

= O 1 +

= O

+

) exp(

).

exp(  2 ) leads to an equation of the form 1.3.1.16:

= exp(2

exp(2
2

+ (1 2

3 )
= O

The substitution

"  = + (9  )1 

) exp( 3).
2
+  exp(  3 ) leads to an equation of the form
The transformation =  2 , = O
3 
P
1.3.1.32: "  = pA 2(9  )1 1 J 2 .

3 )

1
2 

4
3

+ (1

+ 2 (1 3 ) exp( 3 ).
1
+  2 exp(  3 ) leads to an equation of the form 1.3.1.33:

3 

2003 by Chapman & Hall/CRC

Page 140

1.4. EQUATIONS CONTAINING POLYNOMIAL FUNCTIONS OF

34.

= 29 1 1 ) 2 exp(2 3 ) 2 )
Solution in parametric form:

J :

4 3

 = 'U
where

2 3 2
,
3

 =T

J :

=U

: 
*

2 3

141

1 ) 2 (2 3 ) 2 1) exp( 3 ) 2).

3
4

1
= U

12 ,

1 2
2 (

1
3

:
,

T\ 1 2 ) exp  A

2
1

ACU

2
3

3
1

,

J U ) + L 2 n 1 J 3 (U ) for the upper sign,


J U ) + L 2 ; 1 J 3 (U ) for the lower sign,
1 J 3 ( U ) and n 1 J 3 ( U ) are the Bessel functions, and 1 J 3 ( U ) and ; 1 J 3 ( U ) are the modified Bessel
=

1 1 3(

1 1 3(

functions.
35.

)4 ) 3 exp O

Solution in parametric form:

 =

2  
,
U

3 P

3 J U J

1
27

2(

)13 ) 6 (2

2
4(U + 1)  2 + 4U 2  3U
2
7
J
6
2(U + 1)  U 2
 U )

2 3 2 3 (2

+ 27 2 ) exp O

exp O

3U

6 P

where  = U ln |1 + U | + L .
36.

+ cosh(

The transformation a =

37.

2 a  .

+ sinh(

The transformation a =

38.

39.

40.
41.
42.

43.

2 a +  .

.
1

sinh(  ),

= 

cosh(  ),

= 

g =
leads to a Riccati equation: 2  3

.
1

g =
leads to a Riccati equation: 2  3

= 3 + 3 2 cosh2 2 3 cosh3 + sinh .


=  cosh  + 1
 leads to an Abel equation of the form 1.3.2.9:
The substitution
*2 = 3  cosh  + 1.

= 3 + 3 2 sinh2 2 3 sinh3 + cosh .


=  sinh  + 1
 leads to an Abel equation of the form 1.3.2.10:
The substitution
*2 = 3  sinh  + 1.

= 3 + cos( ) 2 .
The substitution = 1
leads to an Abel equation of the form 1.3.2.11: "  =  cos(  ) +1.

= 3 + sin( ) 2 .
The substitution = 1
leads to an Abel equation of the form 1.3.2.12: "  =  sin(  ) +1.

= 3 + 3 2 cos2 ( ) +  sin( ) + 2 3 cos3 ( ).


The substitution =  cos(  ) + 1
 leads to an Abel equation of the form 1.3.2.11:
*2 = 3  cos(  ) + 1.

= 3 + 3 2 sin2 ( ) +  cos(
The substitution =  sin(  ) + 1

*2 = 3  sin(  ) + 1.

) 2 3 sin3 ( ).
leads to an Abel equation of the form 1.3.2.12:

2003 by Chapman & Hall/CRC

Page 141

142

FIRST-ORDER DIFFERENTIAL EQUATIONS

44.

!

+ OD!"

The substitution

"

" P

= ! ( ),

In equations 4447, the following notation is used:

= " ( ),

"

= % ( ).

+ !" 3 .

 =  2 ( 
leads to a separable equation: p

+  + ( ).

= ! 3 + 3 !% 2 + (" + 3 !% 2 ) + !% 3 + "/% % .


  = (  ) +  (  )
The substitution = + % ( ) leads to a Bernoulli equation: p

45.

46.

2(

"
 " +

Solution: X

)3

!"I

1
+ L = ln |  +  |, where =
.
M
3 +1

 (  +  )

+ L"
" ! + ! " .
% +
+ P
! "
" !

 



+




X (  )2 %  S .
= L exp
Solution: |  | | | 

 +
Q  +
M

47.

!

= ( ! )( " ) Oi

1.4.2. Equations of the Form


( M 22 2 + M 12 0 + M 11 2 + M

0)

 = N

22

+N

0 + N

12

11

+N

1.4.2-1. Preliminary remarks. Some transformations.


1 b . For # 22 = 0, this is an Abel equation (see Subsection 1.3.4). For $

with respect to  =  ( ).

H = 
 ,
2 b . The transformation


[( #

H $ 0) + #

22

H 3 + (#

12

=
22 )

11

= 0, this is an Abel equation

leads to an Abel equation


of the second kind:


H 2 + (#

11

12 )

H $

11 ]

 = 2#

+ 2( #

22

H 2+#

12

H +#

11 )

3 b . The transformation  =  + , = + , where and are parameters, which are determined


by solving the second-order algebraic system

22

+#

12

+ #

11

+#

= 0,

22

+$

12

+ $

11

+$

= 0,

leads to the equation


[#

22

+#

 +#

12

11

The transformation = 

{[  2

+ (

+ # 12 ) + (2 # 11 + # 12 ) )]  
= $ 22 2 + $ 12  + $ 11  2 + (2 $ 22 + $ 12 ) + (2 $ 11 + $ 12 ) .
 , = 1
 reduces this equation to an Abel equation of the second kind:
+ # 22 3 + ( # 12 $ 22 ) 2 + ( # 11 $ 12 ) $ 11 } 
= (  +  ) 2 + ( # 2 + # + # ) ,

+ (2 #

 2 )  1 ]

22

where 

= 2#

11

+#

4 b . The substitution

22

12

11

,  1 = 2 $ 11 + $ 12 ,  2 = 2 # 22 + # 12 , and  2 = 2 $ 22 + $ 12 .
= a +  , where parameter is determined by solving the cubic equation

12

(#

22

2+#

12

+#

11 )

22

2$

12

11

= 0,

leads to an Abel equation of the second kind with respect to  =  (a ):


[ ua + ( $
where

22

= 2$

#
22

22

)a 2 + $

+$

12

(2 #

0
22

]  g = ( #

+#

22

2+#

12

+#

11 )

+ (2 #

22

+#

12 )

a + #

22

a 2 + # 0,

12 ).

2003 by Chapman & Hall/CRC

Page 142

143

1.4. EQUATIONS CONTAINING POLYNOMIAL FUNCTIONS OF

1.4.2-2. Solvable equations and their solutions.


1.
2.

3.

4.

(  2 + 2 )
Solution: #

= 2 + (
3
$I 3 + 3(

+ .
 ) = L .

(  2 + ( 2 2 ( ) = =] 2 + 2 (
The transformation  = +  , =
( #* 2 + $ ) + 2 !$ .

7.
8.
9.
10.
11.
12.
13.
14.

+ L3

 =
+ $I )2

(  2 2 ] + (m 2 ) = ( 2 + 2 (m ] 3 2 + .
The substitution = H + ,! leads to a Riccati equation with respect to  =  ( H ):
2

+  ]

 = , ( $ #3, )

+ #2H 2 .

(  2 + 2 ( + ] 2 2 ) = ( 2 + 2 ] 2 + (m 2 2 + .
The substitution = H + ,! leads to a Riccati equation with respect to  =  ( H ):
[( $ #3, ) H

6.

.
leads to a linear equation: ( #*

(  2 + ( + = 2 ) = F 2 + G + H 2 .
Homogeneous equation. The substitution H = 
 leads to a separable equation: "H   =
( #2H 2 + $CH + L )1 [ #2H 3 + ( c $ ) H 2 + ( L ) H +  ].

[( #3, + $ ) H

5.

(  2 + ( + =
The substitution
(  2 + 2 (
Solution: #

+  ]

 = 2 , ( #3, + $ )

+ 2( #3, + $ ) H + #2H 2 .

+ ) = ]I 2 + (m + = 2 + .
= H + ,! leads to a Riccati equation with respect to  =  ( H ):
2

(   , )  = ( #3,

+ $, + L )

+ F 2 + ) = (
3
p 3 + 3( $I 2 + c\

2
2

2F
+

+ (2 #3, + $ ) H + #2H

+G 2+ .
' ) = L .

(  2 2  + ( 2 +  ( ) =  2 + 2 ( ( 2 +
This is a special case of equation 1.4.2.21 with  = 1 and  = 1.

(  2 + 2  + ( 2 +  ( ) =  2 + 2 ( + ( 2  +
This is a special case of equation 1.4.2.21 with  = 1 and  = 1.

(  2 4  + ( 2 + 4  ( ) = 2  2 + 2 ( 2 (
This is a special case of equation 1.4.2.21 with  = 1 and  = 2.

.
.

+ 8

(  2 + 4  + ( 2 + 4  ( ) = 2  2 + 2 ( + 2 ( 2 8 
This is a special case of equation 1.4.2.21 with  = 1 and  = 2.

(  2 6  + ( 2 + 9  ( ) = 3  2 + 2 ( 3 (
This is a special case of equation 1.4.2.21 with  = 1 and  = 3.

+ 27 

(  2 + 6  + ( 2 + 9  ( ) = 3  2 + 2 ( + 3 ( 2 27 
This is a special case of equation 1.4.2.21 with  = 1 and  = 3.

2(  2  + ( 2 +  4 ( ) =  2 + 4 ( ( 2 +
This is a special case of equation 1.4.2.21 with  = 2 and  = 1.

+ .

2( .
+ 2( .
3( .
+ 3( .

4( .

2003 by Chapman & Hall/CRC

Page 143

144

FIRST-ORDER DIFFERENTIAL EQUATIONS


2(  2 +  + ( 2 +  4 ( ) =  2 + 4 ( + ( 2  + 4 ( .
This is a special case of equation 1.4.2.21 with  = 2 and  = 1.

15.

( 2 2 + 2 + 2 3 ) = 2 + 2 2 + 3 2 .
This is a special case of equation 1.4.2.21 with # = 1 and $ = 1.

16.

( 2 2 2 + 2 + 3 ) = 2 2 + 2 + 3 + 2 .
This is a special case of equation 1.4.2.21 with # = 1 and $ = 1.

17.

( 2 2 + 2 2 + 2 2 3 ) = 2 + 4 2 2 +
This is a special case of equation 1.4.2.21 with # = 1 and $ = 2.

18.

( 2 2 2 2 + 2 + 2 3 ) = 2 4 + 2 2 + 3 + 2
This is a special case of equation 1.4.2.21 with # = 1 and $ = 2.

19.

(  2 + ( + =
The substitution

20.

[( c

+ ) = F 2 + (2 ] + ( 2 F ) + ( ] 2 + F} + = )
= H + ,! leads to a Riccati equation with respect to  =  ( H ):

#3, ) H

( !#*

+   , ]  = ( #3,

(  2 2  + ( 2 +
The transformation  = +  ,

21.

+ #*

+ $, + L )

=  2 + 2 ( (
= +  leads to a linear equation:

( )

+ !$I $ )  = ( !#*

+ (2 #3, + $ ) H + #2H

+ .

+ .

3

(

2 #* + !$ ) + 2 2 $ 2  2 # .

1.4.3. Equations of the Form ( M 22 2 + M


= N 22 2 + N 12 0 + N 11 2 + N 2 + N

12
1

0 + M

11

+M

+M

)

1.4.3-1. Preliminary remarks.


1 b . For # 22 = 0, this is an Abel equation (see Subsection 1.3.4). For $
with respect to  =  ( ).

11

= 0 this is an Abel equation

2 b . The transformation = 
 , = 1
 leads to an Abel equation of the second kind:

{[ #

2 + (#

2)

$ 1 ] + #

22

3 + (#

12

$ 11 } 
= ( # 2 + # 1 ) 2 + ( # 22 2 + #

2 + (#

22 )

11

12 )

12

+#

11 )

3 b . In Item 3 b of Subsection 1.4.4, another transformation is given which reduces the original
equation to an Abel equation of the second kind.
4 b . Dynamical systems of the second-order

M
M


a

( , ),

M
M

= ( , ),

(1)

which describe the behavior of simplest Lagrangian and Hamiltonian systems in mechanics, are
often reduced to equations of the type in question if

( , ) =  ( , )( #

( , ) =  ( , )($

22
22

2
2

+#
+$

12
12




+#
+$

11
11




+#

+$

+#

 ),
+ $ 1  ),
1

(2)

where  =  ( , ) is an arbitrary function.

2003 by Chapman & Hall/CRC

Page 144

1.4. EQUATIONS CONTAINING POLYNOMIAL FUNCTIONS OF

145

In particular, dynamical systems (1) with functions (2) and  1 arise in analyzing complex
. and are substituted by their Taylor-series
equilibrium states. In this case, the functions
expansions in the vicinity of the equilibrium state  = = 0 with the first and second order terms
retained.
Whenever a solution of the ordinary differential equation
(#

22

+#

12

+#

11

+#

+#

is obtained in parametric form,  =  ( , L


(1), (2) is determined by

 =  ( , L 1 ),

= ( , L

 )  = $

22

= ( , L

1 ),

1 ),

12

+$

11

+$

+$

the corresponding solution of system

 

+ L 2.
M
 ( , L 1 ), ( , L 1 ) 

a =X

1 ),

+$

The last relation defines an implicit dependence of the parameter on a , = (a , L 1 , L 2 ), and makes
it possible to establish, with the aid of the first two formulas, the dependence of  and on a .
1.4.3-2. Solvable equations and their solutions.
1.

( 2 2 + ) = 2 2 +
Solution in parametric form:

 = a + L |a |1  4g ,

2.

( 2 2 + ) = 2 2 2
Solution in parametric form:

= a + L |a |1 

4.

( 2 2 + ) = 2
Solution in parametric form:
( 2 2 + + 2 ) =
Solution in parametric form:

 = a + L  2g ,

g
= L3a 2  J .

+ 2 +

= a + L 

( 2 2 + + 2 ) = 2
Solution in parametric form:

6.

( 2 2 + 2 ) = 4
Solution in parametric form:

+ 2

 = 13 a + L |a |2 J 3  J g ,

7.

( 2 2 + + 3 ) =
Solution in parametric form:

+ 4

 = 12 a + L |a |1  J g ,

g.

= 2a + L3a
2

g.

= a + L |a |3 
+ 2

 = a + L3a 2  J g ,

 = a + L |a |3  4g ,

5.

g.

+ .

 = a + L3a 2  J g ,

3.

+ 2

g
 J .

g
= 23 a + L |a |2 J 3  J .
2

+ + 3

g
= 23 a + L |a |1  J .

2003 by Chapman & Hall/CRC

Page 145

146
8.

FIRST-ORDER DIFFERENTIAL EQUATIONS


( 2 + + ) =
Solution in parametric form:

 = a + L |a |1  J g ,

9.

10.

( 2 + + ) = 2
Solution in parametric form:
( 2 + 2 ) = 3
Solution in parametric form:

g
= a + L |a |1  J .

+ 2 .

 = a + L3a 2  g ,
2

+ 2

= L3a 2  .

 = 12 a + L |a |1 J 2  J g ,

11.

( 2 + 2 2 + + ) =
Solution in parametric form:

12.

( 2 + 2 2 + + ) = 2
Solution in parametric form:

( 2 + 2 2 + )
Solution in parametric form:

14.

15.

( 2 + 2 2 + 2 ) = 5
Solution in parametric form:
(

Solution:  =
16.

L ln | |

( 2 2 + 2 + + )
Solution in parametric form:

18.

2 + 2

+ 2

g.

+ 2

g
= 12 a + L |a |3 J 4  J .

 =
+

= 2 a + L 

+ 2

2 ln |a |

2
+
2 )

3 ln |a |
= 2(

+ L3a ,

( 2 2 + 2 + + 2 ) = 2(
Solution in parametric form:

( 2 2

 = 14 a + L |a |3 J 4  J g ,

 =
17.

g
 .

2 9

g
= a + L | a |3  J .

 = a + L  3g ,

= 2 a + L3a

 = a + L | a |3  J g ,

13.

+ 2

g
= a + L | a |1 J 2  J .

 = a + L3a 2  9g ,

+ 2

=
+

+ L3a ,
2

+ +

2 ln |a |

+ L3a .

) + + 2

2
+ L3a .
3 ln |a |
) + 2

Solution in parametric form:

 =

ln |a |

+ L3a ,

2
+ L3a .
ln |a |

2003 by Chapman & Hall/CRC

Page 146

1.4. EQUATIONS CONTAINING POLYNOMIAL FUNCTIONS OF

19.

( 2 + 2 + 2 + + 2 ) =
Solution in parametric form:

 =L
20.

O a 1J 3 +

 =L
21.

Oa 3 +

4a

P

a,

+ L3a ,

( 2 + 2 3 2 + + ) = 3
Solution in parametric form:

( 2 + 2 3 2 + + ) =
Solution in parametric form:

+ 2

 = a + L |a |1  8g ,

24.

( 2 + 2 3 2 + + 2 ) =
Solution in parametric form:

 = a + L |a |3 

25.

( 2 2 + + ) = 2
Solution in parametric form:

16

g,

+
 = (   )a + L |a |   4g ,

26.

( 2 + + ) = 2
Solution in parametric form:

( 2 + 2 2 + + ) =
Solution in parametric form:

( 2 2 + 2 + ) = (
Solution in parametric form:

 =

P

 0.

+ L3a ,

 0.

g
= 21 a + L3a 2  J .

+ 3

+ 3

1

+ L3a ,
 1 ln |a |

g.

.
16

g.

+  4g

 ,

= L |a |
2

+ g
  ,

+ ( ) + 2

)+

 .

 0.

1
 
+ L3a ,
 1 ln |a |

+  9g

 ,

= 2(  2 )a + L |a | 2

= 3 a + L |a |3 

+ ( + ) .

+
 = (2   )a + L |a | 2    9g ,

28.

a,

= (   )a + L |a |

+ g
 = 'a + L |a |    ,

27.

4a

= 3 a + L |a |1 

+ 2

Oa 3 +

 0.

4a 3
+L
3

2 +

= L

 = 12 a + L3a 2  J g ,

23.

= L

4a 2
+ L3a ,
5 P

O a 1J 3 +

= L

147

+ 2 +

4a 3
1
+L
3

( 2 + 2 + 2 + 2 )
Solution in parametric form:

 =L
22.

4a 2
+ L3a ,
5 P

( 2 + 2 + 2 + )
Solution in parametric form:

 2 .

 1.

2003 by Chapman & Hall/CRC

Page 147

148
29.

30.

FIRST-ORDER DIFFERENTIAL EQUATIONS


( 2 + 2 3 2 + + ) =
Solution in parametric form:
( 3
2

+
 = (3   )a + L |a | 3   
+ 2

+
+ )

31.

( + 3

32.

16

+ 2

+
+ )

= + 3

(  + 1)

g,

33.

+ 2 ]

# +3

+ L |a | O

( 2 2  +  2 2 +
Solution in parametric form:

36.

38.

+ ( 3 ) + 4

+



= (  + 4) (  + 6)



a,

 3 .

+  g
+ 
 ,

 2  .
.

25

g,

= #3L

= (2

# 3.

2( # 1) 3
a + L 2a ,
 0.
3
) 2 2 +  2 + ] .

1+

#*a

 4 .

+ 2

+2
2a
g
+ L |a | O +3   J ,
# +3
O
2  2 +  3 2 +

] ]

+ (3 + ) 2

g,

=
2

16

+ L3a 2   J ,
=
+ L3a 2   J ,
# 1.
1#
1#
( 2 2  +  2 2 + + ) =  ( 2 2  +  2 2 ) + (  + + )  .
Solution in parametric form:
+
+
(1 # )2 2
(1 # )2 2
a + L3a ,
a +3
 = L 2 a O +  +
= #3L 2 a O +   +
L a,
(2 # )  + 
(2 # )  + 
where  +  0 and (2 # )  +  0.
[ 2 (  + 2) + (  + 1) 2 +
Solution in parametric form:

 =

37.

+



= 2(2  +  )a + L |a | 2

 Jg,

 =

35.

+2
+3

2( # 1) 3
a +L
 =L
1+
3
[ 2 2  + (2  1) 2 +
Solution in parametric form:
3

34.

+ ( 2 ) + 3

= 4(  4 )a + L |a | 4

Solution in parametric form:

 =

+ 2

25

= 3(  3 )a + L |a | 3

+
 = (4   )a + L |a | 4   

[ + 

g,

+
 = (2  +  )a + L |a | 2 +   g ,

Solution in parametric form:


2

= 3

Solution in parametric form:


2

+ L |a | O 

( 1)

Jg,

1#
[  2 + (  + 1) 2 + +
Solution in parametric form:
(  2 + ( + =
The substitution
[ #2H

 = a + L | a |2 O

= (

 Jg,

#*a

1#
+ 1)

( 1)

+ L |a | O 
2

= a + L | a |2 O

+1

]

Jg,
+

# 1.

 Jg.

= F 2 + G + H 2 + I .
= "H leads to a linear equation with respect to  =  (H ):
2

+ (c

+1

= 

$ )H

+ ( L ) H +  ]  = ( #2H

39.

(  + (

40.

(  2 + 2 ( + ] 2 2 + + ) = ( 2 + 2 ] 2
# , 2.
This is a special case of equation 1.4.3.57 with L = 3

+=

] ]

( = )

=F +G
2

+ $CH + L ) + , .

 (=

G )

The transformation  = + , = leads to a linear equation:



[ #* 3 + ( c $ ) 2 + ( L ) ] = ( #* 2 + $I + L ) + sL .

(m 2

2003 by Chapman & Hall/CRC

Page 148

1.4. EQUATIONS CONTAINING POLYNOMIAL FUNCTIONS OF

41.

( 

( 

( 

+ 2(

(m

] 2

] 2

+ 2 ]

This is a special case of equation 1.4.3.62 with L

42.

+ 2(

= #3, .

+ 2(

( 

( 

( 

( 

( 

( 

+ 2(

(m

(m

= #3, .

+ 2(

+ 2 ]

+ 2 ]

+ 2 ]

2 ]

+ +

+ +

+ 2 ]

]

+ I +

= $, .

]I

]I

(m

This is a special case of equation 1.4.3.59 with


50.

( 

2 ]

(m

51.

+ 2

 2

Solution in parametric form ( # 2, $

52.

2

 2

Solution in parametric form ( $


2

 =L
53.

[2

(  +3)

aQP

2, , 0):

# 1 2
a + L3a ,
( $ 2) ,

+(  +1)

+ ( >(

+ x

+ 2 (m

]

+ 2 (m

]

 2

) (

=  , .

+ 2 ]

= #3L

1) I + (  ( )
=  ( 2 2 

+ ((

+ x

+ 1) ( +

+
+

 (

1) (

0):

# +1 2
a + L3a ,
aO +
( # 2) $

 =L

=.

1) ( 2 >( ]
=  ( 2 + 2 

+ (

= #3, .

= (

This is a special case of equation 1.4.3.59 with

= #3, .

= (

+ 2 ]
2

= #3, .

+ 2 ]

= $, .

]

]I

+ 2 ]

This is a special case of equation 1.4.3.61 with $

49.

+ x

This is a special case of equation 1.4.3.62 with $

48.

]

This is a special case of equation 1.4.3.57 with $

47.

(m 2

This is a special case of equation 1.4.3.61 with L

46.

(m 2

149

+ 2 ]

=.

This is a special case of equation 1.4.3.62 with L

45.

This is a special case of equation 1.4.3.58 with


44.

+ 2 ]

This is a special case of equation 1.4.3.61 with L

43.

]
+  2
= #3L

# +1 2
a + L3a .
aO +
( # 2) $

= (  +1)

aQP
2

1) I  ( (

) + ( >(

1)

# 1 2
a + L3a .
( $ 2) ,

(3  +1)

+2 

+ ( >(

Solution in parametric form:

 =

1#

+ L |a |1 

P Jg,

#*a

1#

+ L |a |1 

P Jg,

# 1.

2003 by Chapman & Hall/CRC

Page 149

150
54.

FIRST-ORDER DIFFERENTIAL EQUATIONS

 =
55.

1#

+ L | a |3 

(  2 + ( + =
The substitution

 2 + 2(
(

ln |a |

+ ( >(

)  (  ( ) + ( (  ( ) .

+ L3a .

+ !H .

+ 2 $, + L )

+ [2( #3, + $ )H +  , +  ] + #2H

+ $, )

+ [2( #3, + $ )H +  , +  ] + #2H

2
+
+ )

=
( 2 + 2]
 2

+ (  , +  ) + #2H
+m
( 2

(  2 + 2 ( + =
The substitution

 , ] H

 = 2 , ( #3, + $ )

+ !H .

1)

+ !H .

Solution:

+ 

 , ] H  = ( #3,

+ 2 $, + L )

+ 2( #3, + $ ) H + #2H

+ !H .

+ + ) = ( 2 + 2 ] 2 + ( ] 2 + (m + = ) 2 + I +
= H + ,! leads to a Riccati equation with respect to  =  ( H ):
2

+ 2 $, + L )

1,

+ L |a | O 

P Jg,

+ [2( #3, + $ )H +  , +  ] + #2H

+

) .

+ ) = ( 2 +2 ] 2 + ( ] 2 + (m + = ) 2 + x
= H + ,! leads to a Riccati equation with respect to  =  ( H ):

 =
[  ( 

) .

+ !H .

+ [2( #3, + $ )H +  , +  ] + #2H

,!a

1,

+ L |a | O 

P Jg,

1)

) .

+ [2  ( + 1)] + ( ] 1) 2 + ( (m }
= (  ) 2 + [2  ( + 1)] + ( 
Solution in parametric form:

{( 

+ !H .

+ x + ( +
= H + ,! leads to a Riccati equation with respect to  =  ( H ):

(  2 +2 ( + =
The substitution

# 1.

+ [(2 #3, + $ )H +  , +  ] + #2H

 , ] H  = , ( $ #3, )

( $ #3, ) H 2   = ( #3,

64.

 , ] H  = ( #3,

+]
 2

[( $ #3, ) H

63.

(  2 2 ] + (m 2 + + ) = ( 2 + 2 (m ] 3 2 + x + ( +
The substitution = H + ,! leads to a Riccati equation with respect to  =  ( H ):

[( $ #3, ) H +

62.

+ L3a ,

P Jg,

+2 

(  2 + 2 ( (m 2 + + ) = ( 2 + 2 ] 2 ] 3 2 + x + ( +
The substitution = H + ,! leads to a Riccati equation with respect to  =  ( H ):

The substitution

61.

ln |a |

 = ( #3, 2 + $, + L )

[( #3, + $ ) H +

60.

+ L | a |3 

(  2 + 2 ( + = 2 + + ) = ( 2 + 2 ] 2 + ( ] 2 + (m + = ) 2 + +
The substitution = H + ,! leads to a Riccati equation with respect to  =  ( H ):

[( $ #3, ) H +

59.

1#

(  +3)

+ + ) = ]I 2 + (m + = 2 + + ( + ) .
= H + ,! leads to a Riccati equation with respect to  =  ( H ):

[( $ #3, ) H +   , ] H  = ( #3,

58.

#*a

(B  , ) H

57.

P Jg,

= (3  )

[  ( 2 2 + 2 )  (  ( ) + ( (  ( ) ] = ( (
Solution in parametric form:

 =
56.

[2 2 (3  +1) +(3  1) 2 + ( >(


Solution in parametric form:

+ !H .

(m

, 1.

) + (2   2  ) + ( >(4 ) ]
+ ( (  2 +  + 2 ) + ( >(4 ) + (2 (
+ =  2 #2. $C.  + . = L exp( # I
$  ).

 )

= 0.

2003 by Chapman & Hall/CRC

Page 150

1.4. EQUATIONS CONTAINING POLYNOMIAL FUNCTIONS OF

65.

The substitution

+  1 ) = ( 22 2 + (2  22 +  12 2 ( 22 )
+ (  22 2 + ( 22 +  11 ) 2 + ( 2 + (  2 + 
= H + ,! leads to a Riccati equation with respect to  =  ( H ):

[( $

(

22

22

12

22

, )H + $

11

 2

, ] H  = ( #

, 2+#

12

(

2

] 2

RT

+
=

H .

RT(m

+ @U(m

+ @U] @U(m

2)

, + # 11 )
+ [(2 # 22 , + #

22

In equations 6670, the following notation is used:

66.

151

12 )

]

H +#

(

R  ( )
2
( 2 2 (m

, + # 1 ] + #

0,

(m 2

H 2+#

]

+ ( .

RT]

22

Solution in parametric form:

 =
67.

[ /

/S

(m

+  L3a ,

ln |a |
@U
=

/

+ L3a .

ln |a |

+ (@3( R ) ]
2 /S + /(m

Solution in parametric form:

3
 = #*a +  L |a |- +1   g ,

68.

R )

( /S

 = #*a +  L |a | 
69.

( / 3 2 2 / 2 ( + />( 2
Solution in parametric form:

 = #3L
70.

V /

3 2

2 /

2
3

+1

exp O

0
,
Ca P

/>( 2

= $Ia + L |a | 

+ ) = /  2
(

, Ca 3 + 1 +  L

a,

= $L

+1

exp O
2

+ @U

Solution in parametric form ( 0 1):

 = #3L

Oa - +1 +

a 2 +  L3a ,
0 1 P

1.4.4. Equations of the Form ( M 22 2 + M


= N 22 2 + N 12 0 + N 11 2 + N 2 + N

= $L

Oa - +1 +

0 + M
1 + N

12

0
.
Ca P

+ / ( 3

, Ca 3 + 1 + L

(@U] + R ) WN
=/
 2 ( 2 2 />( 2 + / ( 3
2

/(m 2

2 />(
2
3

11

R )

3 g
.

= $Ia + L |a |- +1  

+ ( ( R ) 2 + @U @3(


= ( /(m + R ) 2 ( /S + R )
Solution in parametric form:

[ /

(@U] +

+ (@3(

a.

R )

@U(m

a 2 + L3a .
0 1 P

2+M

+M

+ M 0 )

1.4.4-1. Preliminary remarks. Some transformations.


1 b . With # 22 = 0, this is an Abel equation (see Subsection 1.3.4). With $
equation with respect to  =  ( ).
See Subsection 1.4.2 for the case # 2 = # 1 = $ 2 = $ 1 = 0.
See Subsection 1.4.3 for the case # 0 = $ 0 = 0.

11

= 0, this is an Abel

2003 by Chapman & Hall/CRC

Page 151

152

FIRST-ORDER DIFFERENTIAL EQUATIONS

2 b . The transformation  =  + , = + , where and are parameters, which are determined


by solving the second-order algebraic system
# 22 2 + # 12 + # 11 2 + # 2 + # 1 + # 0 = 0,

22

+$

12

+ $

11

+$

+$

+$

= 0,

leads to the equation


(1)
( # 22 2 + # 12  + # 11  2 +  2 +  1  )   = $ 22 2 + $ 12  + $ 11  2 +  2 +  1  ,
where
 2 = 2 # 22 + # 12 + # 2 ,
 1 = 2 # 11 + # 12 + # 1 ,
 2 = 2 $ 22 + $ 12 + $ 2 ,
 1 = 2 $ 11 + $ 12 + $ 1 .
The transformation = 
 , = 1
 reduces equation (1) to an Abel equation of the second kind:
{[  2 2 + (  1  2 )  1 ] + # 22 3 + ( # 12 $ 22 ) 2 + ( # 11 $ 12 ) $ 11 } 

= (  2 +  1 ) 2 + ( # 22 2 + # 12 + # 11 ) .
3 b . The substitution = H +  , where the parameter is determined by solving the cubic equation
( # 22 2 + # 12 + # 11 ) $ 22 2 $ 12 $ 11 = 0,
leads to an Abel equation of the second kind with respect to  =  ( H ):
[( pH + ) + ($ 22 # 22 ) H 2 + ( $ 2 # 2 ) H + $ 0 # 0 ] 
= ( # 22 2 + # 12 + # 11 ) 2 + [(2 # 22 + # 12 ) H + # 2 + # 1 ] + # 22 H 2 + # 2 H + # 0 ,
where = 2 $ 22 + $ 12 (2 # 22 + # 12 ), = $ 2 + $ 1 ( # 2 + # 1 ).
1.4.4-2. Solvable equations and their solutions.
1.
2.

3.

4.

5.

6.

7.

8.

( + + )2 = (  +  + )2 .
This is a special case of equation 1.7.1.6 with  ( H ) = H

 2+(
(

( +

$
 ) = ]
= 2+F

+ ( F ) .
The transformation  = + , = leads to a linear equation:
[ #* 3 + ( L $ ) 2 + c\ ]  = ( #* 2 + $I ) + , .

(  2 + 2  + ( 2 +  ( ) =  2 + 2 ( + F 2 + 2( ( F ) + F
The transformation  = + 1, = 1 leads to a linear equation:
( #* 3 #* 2 + $I + c )  = ( #* 2 + 2 #* + $ ) + 2( $ # ).

(  2 2  + ( 2 +  ( ) =
The transformation  = 1, =
( #* 3 + #* 2 + $I + L

 2 + 2 ( + = 2 + 2( ( + = ) + 
1 leads to a linear equation:
)  = ( #* 2 2 #* + $ ) + 2( # $ ).

(  2 + 2  + ( 2 +  ( ) =  2 + 2 ( + = 2 + 2( = ( ) 
The transformation  = 1, = + 1 leads to a linear equation:
# 2 + $I + L )  = ( #* 2 + 2 #* + $ ) + 2(# $ ).
( #* 3 *
(  2 2  + ( 2 +  ( ) =
The transformation  = + 1, =
( #* 3 + #* 2 + $I + L

 .

 2 + 2 ( + = 2 2( ( + = ) + 
+ 1 leads to a linear equation:

) = ( #* 2 2 #* + $ ) + 2( $ # ).

+= .

+= .

(  2 2  + ( 2 +  ( ) = =] 2 +2 ( + F 2 2(  + = ) 2( ( + F ) +2  + = + F .
The transformation  = + 1, = + 1 leads to a linear equation:

[ #* 3 + (2 # + L ) 2 + $I + c ] = ( #* 2 2 #* + $ ) + 2( $ # ).

(2  2 2  + ( 2 + 2  4 ( ) =  2 + 2 ( + F 2 2( ( + 2 F ) + 
This is a special case of equation 1.4.4.34 with = 2, = 1, and L = # .

+ 4F .

2003 by Chapman & Hall/CRC

Page 152

1.4. EQUATIONS CONTAINING POLYNOMIAL FUNCTIONS OF

9.

13.
14.
15.
16.
17.

2 #*


+ $I + L ) = ( #*

+ 2 #*

+ $I + L )  = ( #*

+ ( L 4 # )

19.
20.
21.
22.
23.

= )

+ 8

+= .

4 #* + $ ) + 2$ 8 # .

+ $I + 2 $ ]  = ( #*

+ 4 = 16  .

+ 4 #* + $ ) + 2$ 8 # .

(2  2 + 2  + ( 2 + 2  4 ( ) =  2 + 2 ( + F 2 + 2( ( 2 F ) + 4 F
This is a special case of equation 1.4.4.34 with = 2, = 1, and L = # .
(2  2 + 2  ( 2 + 2  + 4 ( ) =  2 2 ( F 2 + 2( ( 2 F ) 
This is a special case of equation 1.4.4.34 with = 2, = 1, and L = # .

(  2 + 2 ( + ] 2 2 + + + ) = ( 2 + 2 ]
This is a special case of equation 1.4.4.27 with L = #3, 2 .

+ (m

+ (m

+ (m

]

]

(  2 + 2 ( + ] 2 2 + + + ) = ( 2 + 2 ]
This is a special case of equation 1.4.4.32 with L = #3, 2 .
(  2 + 2 ( + ] 2 2 + + ) = ( 2 + 2 ]
This is a special case of equation 1.4.4.31 with L = #3, 2 .
(  2 + 2 ( (m 2 + + + ) = ( 2 + 2 ]
This is a special case of equation 1.4.4.28 with =  .

18.

8 .

+ 4 #* + $ ) + 2$ 8 # .

(  2 + 4  + ( 2 + 4  ( ) = =] 2 + 2 ( + 2 ( 2 + 4( = 2  ) + 2 (
The transformation  = + 1, = 2 leads to a linear equation:
[ #*

12.

(  2 4  + ( 2 + 4  ( ) = 2  2 + 2 ( + = 2 2(2 (
The transformation  = + 1, = + 2 leads to a linear equation:
( #*

11.

153

(  2 + 4  + ( 2 + 4  ( ) = 2  2 + 2 ( + = 2 2( = 2 ( ) + =
The transformation  = + 1, = 2 leads to a linear equation:
( #*

10.

(  2 + 2 ( (m 2 + + + ) = ( 2 + 2 ]
This is a special case of equation 1.4.4.32 with L = $, .

(  2 + 2 ( (m 2 + + ) = ( 2 + 2 ]
This is a special case of equation 1.4.4.31 with L = $, .

(  2 + 2 ] + = 2 + + + ) = ]I 2 + 2 ]
# , .
This is a special case of equation 1.4.4.27 with $ = 3
(  2 + 2 ] + = 2 + + + ) = ]I 2 + 2 ]
This is a special case of equation 1.4.4.32 with $ = #3, .
(  2 + 2 ] + = 2 + + ) = ]I 2 + 2 ]
This is a special case of equation 1.4.4.31 with $ = #3, .

(  2 2 ] + (m 2 + + + ) = ( 2 + 2 (m
This is a special case of equation 1.4.4.29 with =  .

+ =

+ =

+ =

]

+ +

+ s.

+ I +

+ s.

+ x

+ s.

4F .

] 3

 .

+ s.

+ I +

+ s.

+ x

+ s.

+ +

+ s.

+ I +

+ s.

+ x

+ s.

+ s.

+ +

2003 by Chapman & Hall/CRC

Page 153

154
24.

FIRST-ORDER DIFFERENTIAL EQUATIONS


( 

2 ]

+ (m

+ +

+ ) = (

This is a special case of equation 1.4.4.29 with


25.

( 

( 

+(

+ 2 $I + L3

+=

+=

+2 (

( 

[( $ #3, ) H
2

+(

( 

[( $ #3, ) H

( 

+ 2(

+(

[( $ #3, ) H
( 

+2 (

V
( 

+G

+H

+ , ,

where

+ I GI

= ]I

+ (m

+ =

' .

, = , + 2 $I .

+ + ( + ) + s.

+ [(2 #3, + $ )H +  , +  ] + #2H

+2 ]

+ ( ]


+ (m + = )

+ !H +

+ +

+s.

+(

+ ) =
(

+ [2( #3, + $ )H +  , +  ] + #2H

+2 ]

]
 3

+ !H +

) +s.

+ x + ( +

 = ( #3, 2 + $, ) 2 + [2( #3, + $ )H +  , +  ] + #2H 2 + !H + ( .

+ +

+ )

= (

+2 (m

]
 3

) +s.

+ x + ( +

 = , ( $ #3, )

 , ) H + (, ]

+ +

+ ) =
(

+2 ]

+ (  , +  ) + #2H

+m
( 2

+ !H + ( .
) +s.

+ x + ( +


= (

+ )
+ 2 ]

+ [2( #3, + $ )H +  , +  ] + #2H

+ ( ]

(m

= )

+ !H + ( .

+ x

+ s.

= H + ,! leads to a Riccati equation with respect to  =  ( H ):


 , ) H + , ]  = ( #3,

+ +

+ ) =
(

+ 2 $, + L )

+2 ]

+ 2( #3, + $ ) H + #2H

+ ( ]


+ (m + = )

+ !H +  .

+ I +

+s.

= H + ,! leads to a Riccati equation with respect to  =  ( H ):

+

Solution:

+ 

+ p + u

 , ) H + (, ]  = 2 , ( #3, + $ )

The substitution

[  ( 

+ s.

= H + ,! leads to a Riccati equation with respect to  =  ( H ):

+=

[( $ #3, ) H

+(

The substitution

33.

= H + ,! leads to a Riccati equation with respect to  =  ( H ):

+]
 2

+2 (

+ ) =
(

+ +

+m
(

2 ]

 , ) H + (, ]

The substitution

32.

+ I +

= H + ,! leads to a Riccati equation with respect to  =  ( H ):

[( #3, + $ ) H

31.

The substitution

30.

= H + ,! leads to a Riccati equation with respect to  =  ( H ):

The substitution

( 

+ +

m
(

+2 (


+ ,  )  = $

+ (   , ) H + , ] 

= ( #3, 2 + 2 $, + L )

[( $ #3, ) H

29.

=  , .

= (

, ]  = ( #3, 2 + $, + L )

The substitution

28.

+I
  2)

]

= H + ,! leads to a Riccati equation with respect to  =  ( H ):

[(B  , ) H +
( 

+ +

The substitution

27.

2 I +

The substitution = leads to an equation of the form 1.4.3.38:


(#

26.

+=

+ 2(

+ 2 (m

, ]  = ( #3, 2 + 2 $, + L )

+ [2( #3, + $ )H +  , +  ] + #2H

+ !H +

) + ( >S + 2  ) + ( BW +  ) + B + S ]


+ ( (  2 +  + 2 ) + ( (XS +  ) + ( (4W + 2 ) + (4 + W = 0.

+ = 

+>

+  + . = L exp( #

$I ).

2003 by Chapman & Hall/CRC

Page 154

1.4. EQUATIONS CONTAINING POLYNOMIAL FUNCTIONS OF

34.

 2 ( ) = =] 2 + 2 (
+ F 2 2 ( + = ) 2( F + ( ) + 
The transformation  = + , = + leads to a linear equation:

( 

2

[ h#*
35.

(

22

12

 2 

+

(2

2#* + $ ) + 2( h$

11

).

# ).

0.

, ) H 2 + ( $ 2 # 2 , ) H + $ 0 # 0 , ] 
= ( # 22 , 2 + # 12 , + # 11 ) 2 + [(2 # 22 , + # 12 ) H + # 2 , + # 1 ] + # 22 H 2 + # 2 H + # 0 .

(  22 2 +  12 +  11 2 +  2 +  1 +  0 ) = ( 22 2 + ( 12 + ( 11 2 + ( 2 + ( 1 + ( 0 .
Here, #2 , $3 , and # 1 are arbitrary parameters, and the other parameters are defined by the
[( $

22

22

relations:

2 # 22 ,
2
+ # 22 2 # 1 ,
0 = # 11
2 = (2 # 11 $ 12 ) + ( # 12 2 $ 22 ) + # 1 ,
1 = 2 $ 11 $ 12 ,
2
+ ( $ 12 2 # 11 ) + ( $ 22 # 12 )
0 = $ 11
2

#
$
$
$

= #

12

( , are arbitrary parameters).


The transformation  = + ,
[ #

22

3 + ($

where , = 2 #

11

22

+#

#
12

12 )

2 + ($

= +
12

11 )

leads to a linear equation:

+$

11 ]

 = (#

22

2+#

12

+#

11 )

+, ,

+ # 1.

1.4.5. Equations of the Form ( M 3


= N 3 3 + N 2 0 2 + N 1 2 + N
1.

+  2 +  1 +  0 ) = ( 22 2 + (2  22 +  12 2 ( 22 )
+ (  22 2 + ( 22 +  11 ) 2 + ( 2 + (  2 +  1 ( 2 ) + (
= H + ,! leads to a Riccati equation with respect to  =  ( H ):

+ $I + c ]  = ( h#*

+ (2# + L )

The substitution

36.

155

0 2 + M 1 2 + M
3 + Z + Z
0
1
0

3
2
2
3
( +
+ ) = + + .
3+M

3 + Y 1 + Y 0 )

Solution in parametric form (  0):


1

 =L

2.

( 3 2 2 + 3 + )
Solution in parametric form:

 =L

3.

a |a | 2    g + 12  L |a | 2    g ,

sign a 

J g + 1  L |a |  1 J g ,
8

( 3 + 2 2 3 + + ) =
Solution in parametric form (   ):

4a 2
 = a + L |a |  + exp O
,
 +  P

4.

( 3 + 2 2 2 + 2 +
Solution in parametric form:

 =L

1 1 1

=L

J g + 1  L3a 2  1 J g ,
3

=L

a |a | 2    g 12  L |a | 2    g .

sign a 

J g 1  L |a |  1 J g .
8

= a L |a |  + exp O

+ 4
=L

1 1 1

4a 2
.
 +  P

+ 4

J g 2  L3a 2  1 J g .
3

2003 by Chapman & Hall/CRC

Page 155

156
5.

FIRST-ORDER DIFFERENTIAL EQUATIONS


( 3 + 2 5 2 + 3 3 + +
Solution in parametric form:
1

 =L
6.

Jg +

a |a | 

1
32

sign a 

= 3

 L |a |  1 J g ,

= 2

= 2

2
27 a 2
 = a + L |a |   + exp
S ,
Q 2(  +  )

9.

10.

3
2( )

a |a |

g
 

1
16 (

12.

= 3

 = a + L3a 2 exp O
13.

( 3 4 2 + 5 2 2 3 + 2 3
Solution in parametric form:
1

 =L

14.

sign a 

sign a 

2a P

3
32

 L |a |  1 J g .

+ 4

J g 2  L |a |  1 J g .
9

27 a 2
S .
Q 2(  +  )

+ 9

a |a |

  g +

= 3

=L
3

15

=L

3

  ) L |a | 2(  )  g .

4a 3
O | a |  +  +
.
2  +  P

+ 10

sign a 

3
16 (

2a 4 + 1.

+ 20

(2 ) + 3

3
2( )

= a + 2 L3a 2 exp O
= 2

J g + 1  L |a |  1 J g ,
8

+ 3

+ ( ) + 2

= L3a L

14

J g +  L |a |  1 J g ,

( 3 5 2 + 7 2 3 3 + 2
Solution in parametric form:

 =L

4a 3
O | a |  +  +
,
2  +  P

( 3 4 2 + 4 2 +
Solution in parametric form:

J g + 2  L |a |1  1 J g .
3

sign a 

= L3a T

( 3 + 3 2 + 3 2 + 3 + + ) =
Solution in parametric form (  2  ):

 = L3a + L

+ 8

=L

L 2
 = L3aA 2a 4 + 1,


11.

Jg

+ 4

=L

3

  ) L |a | 2(  )  g ,

( 3 + 3 2 + 3 2 + 3 +
Solution in parametric form:

a |a | 

sign a 

= a 2 L |a |   + exp

( 3 + 3 2 2 3 3 + + ) =
Solution in parametric form (   ):

 =L

=L

+ 6

+ 15

+ 2

J g + 1  L |a |  1 J g ,
9

( 3 + 3 2 4 3 + + ) = 2
Solution in parametric form (   ):

=L

J g 1  L |a |1  1 J g ,
3

( 3 3 2 + 2 3 + 2 +
Solution in parametric form:

 =L
8.

( 3 + 2 2 2 2 3 + 2 +
Solution in parametric form:

 =L
7.

sign a 

+ 21

sign a 

+ 2 2

2a 2 P

.
3

J g + 2  L |a |  1 J g .

+ 3 4

+ 2 3

J g + 3  L |a |  1 J g .
8

2003 by Chapman & Hall/CRC

Page 156

1.4. EQUATIONS CONTAINING POLYNOMIAL FUNCTIONS OF

15.

( 3 5 2 + 8 2 4 3 + + ) = 2
Solution in parametric form (   ):

10

2 +
a2
 = a + L |a | +   exp
S ,
Q 2(  +  )

16.

( 3 + 5 2 + 3 2 9 3 + + )
Solution in parametric form (   ):

18.

= 3

15

 = L3a"A
19.

 g
 

24.

1
27 (

=L

1 1 1

a |a | 

12

J g 1  L |a |1  1 J g ,
8

( 2 2 2 + 2 3 + ) =
Solution in parametric form ( , 1):


 = a + L |a | exp
S ,
Q 2( , 1)a 2

a2

Q 2(  +  )

+ 27

+ 18

S .

+ ( 2 ) + 3

32 a 2
.
 +  P
+ (4 + ) 3

3 +
3 +
a |a | 4 +2   g 3(  + 12  ) L |a | 4 +2   g .

+ 24

16

2a 4 + 1.

=L

+ 6

3
1

 g
  .

2
2

 2  ) L |a |

+ (3 + ) 2

2 +
2 +
a |a | 3 +2   g 2(3  + 2  ) L |a | 3 +2   g .

1 1 1

=L

( ) + 2

+ 18

=L

2
27 (

 g
  +

a |a | +2   g .

2
2

a |a |

=L

+ 3

=L

+ 3

+ 3

13

J g + 1  L3a 2  1 J g ,
8

(3 3 + 2 3 2 3 + ) =
Solution in parametric form:
1

+ (3 + ) 2

+ ( + ) .

 g
  ,

(3 3 2 3 2 + 3 + ) =
Solution in parametric form:

= L3a"A

2
2

 2  ) L |a |

11

2a 4 + 1,

2 +
2 +
a |a | 3 +2   g (3  + 2  ) L |a | 3 +2   g ,

 =L

a |a | +2   g + (  + 2  ) L |a | +2   g ,

2
2

a |a |

 =L

23.

2 

= 2

(2 3 9 2 + 13 2 6 3 + + ) = 3
Solution in parametric form (  32  ):

 =L

22.

(2 3 + 3 2 3 2 2 3 + + ) =
Solution in parametric form (  2  ):

 =L

21.


 exp

(2 3 3 2 + 2 + + ) = 3
Solution in parametric form (  2  ):

 =L

20.

= a 3 L |a |   + exp O

3 +
3 +
a |a | 4 +2   g (  + 12  ) L |a | 4 +2   g ,

( 3 6 2 + 12 2 8 3 + )
Solution in parametric form (  > 0):

2 +
+

( 3 6 2 + 11 2 6 3 + + ) = 2
Solution in parametric form (  21  ):

 =L

+ 16

= a + 2 L |a |

3
32 a 2
 = a + L |a |   + exp O
,
 +  P

17.

157

a |a | 

2 2

J g 1  L3a 2  1 J g .
8

J g + 1  L |a |1  1 J g .
8
+


S .
Q 2( , 1)a 2

= a + , L |a | exp

2003 by Chapman & Hall/CRC

Page 157

158
25.

26.

FIRST-ORDER DIFFERENTIAL EQUATIONS


( 3 3 2 + 3 2 2 3 3 + ) =
Solution in parametric form (  > 0, , 1):

+ 3

, 1 2
2a + 1,
2 

 = L3a"AeL
2

27.

1
2

( , 1)3
a 3S ,
( , 3)  2 

( + 2)

sign a 

J g +  L |a |  1 J g ,
( , 1)2

=L

 +
+ +


sign a 

( + 2)

(2 + 1)

+ ( + 1) 2 


S .
Q 2( , 1)2 a 2

= a + , L3a 2 exp

(  3 + 2  2 3 + + ) =
1 b . Solution in parametric form with  0:

( , 1)3
a 3S .
( , 3)  2 

+ ( + 2) 2 2 3 + + ]
= I 3 (2 + 1) 2 + 2 ( + 2) 2
Solution in parametric form (   , , 1):

J g +  , L |a |  1 J g .
( , 1)2

( 4) 2 + 2 ( 2) 3 + ]
= (2 1) 3 (4 1) 2 + 2 (2 + 1)
Solution in parametric form ( , 1):

+ [( + 1) + ] 

|a |

3 2

, 1 2
2a + 1.
2 

= L3a + , L

+ 3

 ( , 3), , 1):

+
( , 1)3 2
 = a + L |a | +   exp
a S ,
Q 2(  +  )

30.

= L3aAe, L


 = a + L3a 2 exp
S ,
Q 2( , 1)2 a 2

29.

+ (2 + 1) 2 3 + 2 ( + 1) ]
= I 3 ( + 2) 2 + (2 + 1) 2
Solution in parametric form ( , 1):

 =L
28.

 +
+ +


|a |

+ )
= I 3 3 2 2 + 3 3 2

Solution in parametric form ( 

 = L3a + L

+ [( + 1) + ] 

 +
( , 1)3 2
+ exp
a S .

Q 2(  +  )

= a + , L |a |
3

a |a | 2   |a + 1|  O 2  14 L |a |  2  |a + 1| 2  O ,

= L 1 a |a | 2   |a + 1|  O 2  + 14  L |a |  2  |a + 1| 2  O .

 =L

2 b . Solution in parametric form with  = 0:

 =L
31.

a |a | O

1
1
2

J g 1  L |a | 12O  1 J g ,
8

=L

a |a | O

1
1
2

J g + 1  L |a | 12O  1 J g .
8

(  3 3 +  2 2 +  1 2 +  0 3 + ) = ( 3 3 + ( 2 2 + ( 1 2 + ( 0 3 + .
This is a special case of equation 1.7.1.13 with + ( , ) =  . The transformation a = 
 ,
=  2 leads to a linear equation:

.
.

P (a ) a O (a )] g = 2 O (a ) + 2  ,
. (a ) = # a 3 + # a 2 + # a + # and . (a ) = $ a 3 + $ a 2 + $ a
3
2
1
0
3
2
1
P
O
[

where

+$

0.

2003 by Chapman & Hall/CRC

Page 158

( , )

1.5. EQUATIONS OF THE FORM

32.

33.

4) 2 (  2) 3 +
= (  2) 3 (  4)
Solution in parametric form:

[ 

[ 

+ 2)

= ( , ) CONTAINING ARBITRARY PARAMETERS

+ (

+ 1)

+ 3( 

(

 = a + L |a |1 O exp O

4(  3) 3 + ]
= (2  3) 3 6(  2)

+ 12

+ 2)

+ (

+ 12

8a 2 P

= a L |a |1 O exp O

8a 2 P

159

+ 8

2 + 2

Solution in parametric form:

 = a + L |a |1 O exp O

18a P
2

= a 2 L |a |1 O exp O

18a 2 P

1.5. Equations of the Form ( , ) = ( , )


Containing Arbitrary Parameters
1.5.1. Equations Containing Power Functions
1.5.1-1. Equations of the form  =  ( , ).
1.

2.

=
+ ( 1 ) 2 .
The substitution = 2 #
+ 2 $C# 2  1 J 2 .

Let # = A 2 

4.

5.

=  [2U2A (U )]2 ,

 (U ) = exp(T"U 2 ) X exp(T"U 2 ) U + L2S


Q
M

where

leads to an Abel equation of the form 1.3.1.33:

*p =

=
+ + .
leads to the Abel equation *  = + 2 
The substitution = 2  1
is discussed in Subsection 1.3.1 (see Table 5).

*p =

=
+ ( 2 .
The substitution = 2 #
+ 2 $C# 2  2 .

Solution: X
6.

leads to an Abel equation of the form 1.3.1.32:

 , $ = T 4  (  > 0). Solution in parametric form:

 =   (U ),
3.



( ' + ( + ), which

  + M 1 +  = ln | | + L , where =  
1 


.

The transformation  = (  )1 J' ,


EmdenFowler equation:

1
1

= ($
# )1 J , leads to the generalized
= (u
H )1 J , where

'  = ", H
$

 1
(  )  ,

which is discussed in Section 2.5 (in the classification table, one should search for the equations
satisfying the condition B + + 1 = 0).

2003 by Chapman & Hall/CRC

Page 159

160
7.
8.

FIRST-ORDER DIFFERENTIAL EQUATIONS


= ( + + ) .
This is a special case of equation 1.7.1.1 with  ( ) =  .

=
Solution:

9.
10.
11.
12.

 1 J 
ln | |,
 M + 1 + L = O hP

= 1

+1

where

=O 

+ ' 1 I +1 .

J

hP

 +

, =

+1

 1 J 
.
 P

This is a generalized homogeneous equation of the form 1.7.1.3 with  ( ) =  + '  .

= 
+ + s .
This is a special case of equation 1.7.1.4 with  ( ) =   , ( ) = ' + , % ( ) = ( , and B = 1
2.

=  1+ + + s 1 .
This is a special case of equation 1.7.1.4 with  ( ) =   , ( ) = ' + , and % ( ) = ( .

= 1 1 ( + )  .
This is a special case of equation 1.7.1.7 with  ( ) =  .

1.5.1-2. Other equations.


13.

14.
15.

2.

The substitution = 
 leads to a separable equation:   = (  1) + 

= + +   .
 =  
The substitution =  leads to a separable equation: p

17.
18.

19.

20.

+( .

(  2+ +   ).

= 1.

Solution:  =    OL + lX


16.

   

M P

+ ) + = 0.
Solution: B  = lL J  +   .

+ ) = V (  1)  W .
This is a special case of equation 1.7.1.16 with  ( ) =  , ( ) = 1, % ( ) = '  , and , = B .

( + 2 + ) = + + ^ 2 .
The transformation a = 
 , H =   2 leads to a linear equation with respect to H = H (a ):
( , a ) H g  = (B 2)(!H +  + (a ).
( + 2 + ) = I + + ^ 2 .
The transformation a = 
 , H =   2 leads to a linear equation with respect to H = H (a ):
a  ( , a ) H g  = (B 2)(a  H +  + (a ).

( + + ) =    +  + .
The transformation a = 
 , H =   1 leads to a linear equation: ( a   + a  + 'a 
(B 1)( 'a  +  ) H + B 1.

+1

a ) H g  =

2003 by Chapman & Hall/CRC

Page 160

( , )

1.5. EQUATIONS OF THE FORM

21.

22.
23.
24.
25.
26.

( + +  2 + ( ) =    +  +  + ( 2 .
a  + 'a 
The transformation a = 
 , H =   2 leads to a linear equation: ( a   +

(B 2)( 'a +  ) H + (B 2)( $Ia + # ).

28.

30.
31.

33.
34.

a ) H g  =

(  +  + ) = ( + + ) .
This is a special case of equation 1.7.1.6 with  ( ) =  .

( + ) = 1 1 .
This is a special case of equation 1.7.1.7 with  ( ) = 1
.

( +
Solution:

+ s)



 

+  = 0.

!Y ( ) + j ( ) + ' 
+1

+1
| |
ln

1,

if
if

= 1,

j ( ) =



 

+  = L ,
+1

, +1
ln | |

if , 1,
if , = 1.

( 2 + + ^  ) = [ + \ + .
This is a Riccati equation with respect to  =  ( ).
( + ) =  +  + .
The transformation a = 
 , H =   + +
(B + 1)(a + H + 1).

+  )

+ (

leads to a linear equation: a + ( 'a   a ) H g  =

+  ) = 0.

( 9  : )P
B
 B
(   )O
+
= L , where # =
, $ =
.
Solution:
#
$
 
 

(   +

+ s) + (
Solution:  ,  + ,! + (

+ 

)  = L .

( +  2 + ( ) =  +
The transformation a = 
 , H =   + +
(B + 2)(a + H + $Ia + # ).

32.

+1

[( + ) + ] = ( + ) .
This is a special case of equation 1.7.1.15 with  ( ) =  , ( ) = 1, and % ( ) = ( + .

29.

161

[( + ) + ] = ( + ) .
This is a special case of equation 1.7.1.14 with  ( ) =  , ( ) = 1, and % ( ) = ( + .

where Y ( ) =
27.

= ( , ) CONTAINING ARBITRARY PARAMETERS

+ s) = 0.

+  + ( 2 .
leads to a linear equation: a + ( 'a   a ) H g  =

( 1 +  ) + 1 + s = 0.
This is a special case of equation 1.7.1.19 with  ( ) =   and ( ) = ( .

( +  ) =  s +  .
This is a Bernoulli equation with respect to  =  ( ) (see Subsection 1.1.5).

( 

+ ) = (    ).
This is a special case of equation 1.7.1.16 with  ( ) =  , ( ) = 1, and % ( ) = '  .

2003 by Chapman & Hall/CRC

Page 161

162
35.
36.

FIRST-ORDER DIFFERENTIAL EQUATIONS

( 

+ ) = (    ).
This is a special case of equation 1.7.1.17 with  ( ) =  , ( ) = 1, and % ( ) = '  .

( +1 1 + ' +1 I 1 ) = s s .
This is a special case of equation 1.7.1.3 with  ( ) = ( (  + '  )1 .

37.

( + )  = 1 1 .
This is a special case of equation 1.7.1.7 with  ( ) = (  .

38.



3
1

3
2

3
1

23

= 0,

where 12 = ( + )2 + 2 , 22 = ( )2 + 2 .
This is the equation of force lines corresponding to the Coulomb law in electricity.
 +
 
+ 2
=L .
Solution:  1
39.
40.
41.

&

= ( 

&

+ ).
This is a special case of equation 1.7.1.24 with  ( ) = J
+  )(

= (  +  )( ).
This is a special case of equation 1.7.1.25 with  ( ) = J

+ = (  +  )( ).
This is a special case of equation 1.7.1.24 with  ( ) =

and ( , ) =   +   .

and ( , ) =   +   .

J 2 and ( , ) = (  +   )1 .

1.5.2. Equations Containing Exponential Functions


1.5.2-1. Equations with exponential functions.
1.

]

+ .
=

Solution:
2.

]

Solution:

= B
  ] +D


= 

ln OL 



hP

ln L  X exp(   )  S .

3.

This is a special case of equation 1.7.1.2 with  ( ) = #  , B = 1, and  = 0.

4.

=  + ] +   .
 and ( ) =     .
This is a special case of equation 1.7.2.5 with  ( ) =  

5.

=   + ] +  ] .




This is a special case of equation 1.7.2.8 with  ( ) =   , ( ) = 0, and % ( ) =    .

6.

=  2 ^ _ ] +  ^ +  ^ _ ] .
This is a special case of equation 1.7.2.9 with  ( ) = + ( .

7.

=  ^ + ] +  _ + ` ] .


 =   9 
The substitution =  leads to a Riccati equation: 3



+   : + (  ; .

2003 by Chapman & Hall/CRC

Page 162

1.5. EQUATIONS OF THE FORM

8.

( 

 )

( , )

= ( , ) CONTAINING ARBITRARY PARAMETERS

= 1.

Solution:  =    ln L  X exp(    )
9.

( 

^ ]

( 

] +_

+ )

( 

 _ ] )

+ )

163

 D

+ ]

^ ]



S.
M

This is a special case of equation 1.7.2.13 with  ( ) = ( , ( ) = 1, and % ( ) =  .

10.
11.

= .

 = + ( (  
The substitution ( ) = +  leads to a separable equation: p
+

 ^ _ ]

(

^ +` ]

+  ) +

 +_ ]

+  )1 .

This is a special case of equation 1.7.2.9 with  ( ) =

12.

+ 

= 0.

This is a special case of equation 1.7.2.15 with  ( ) =  ;  and ( ) =  

 .

1.5.2-2. Equations with power and exponential functions.


13.

]

1 b . Solution in parametric form with B 1:


1
+1

 =U 


' U


U!S .
U ln L
X U  +1 exp O
B +1
Q
B +1
B + 1 P3M

2 b . Solution in parametric form with B = 1,  1:

 = * ,

= ln OL 

*


 * .
 +1 P

3 b . Solution in parametric form with B = 1,  = 1:

14.

 = * ,



= U ln( L U ).

Solution in parametric form:

 = ln( #2
where  = T 2 $
15.

,  = A"#

 + ]

 ]



 ]

 ]

  ]

) TU 2 ,

= $ [2 A exp(T"U 2 )

17.
18.

],

$ , = X exp(T"U 2 ) U + L .
M

This is a special case of equation 1.7.2.5 with  ( ) =  


16.

 and ( ) = '  .


This is a special case of equation 1.7.2.5 with  ( ) =   and ( ) =   .
This is a special case of equation 1.7.2.5 with  ( ) =   and ( ) = ' + .
.
This is a special case of equation 1.7.2.8 with  ( ) =   , ( ) = 0, and % ( ) = ' + .

2003 by Chapman & Hall/CRC

Page 163

164

FIRST-ORDER DIFFERENTIAL EQUATIONS

19.

= ( +   + )   .
This is a special case of equation 1.7.2.10 with  ( ) = (  .

20.

= ( ] +

 )1 ) 

Solution in parametric form:

 = exp "U
21.

23.
24.

where

 (U ) = X

, U
.
M
, (  +    * )1 J' + 1

1V
 (U ) + L2W! ,

where

 (U ) = X

, U
.
M
, (  +    * )1 'J  1

= (  +  )1 )  .
Solution in parametric form:

 =  (U ) + L ,
22.

=  (U ) + L ,

1V
 (U ) + L W ,

= exp U +

= 1   ] + 1   ] .
This is a special case of equation 1.7.2.2 with  ( ) =  

1  ^ ]

 ^ ]

+ ' +

+1

This is a special case of equation 1.7.2.4 with  ( ) =  + ' + .

=   +1 +   .
This is a special case of equation 1.7.2.1 with  ( ) =  

+1

+ .

=  ^ +1 +  ^ +1 .
This is a special case of equation 1.7.2.3 with  ( ) =  + '  .

26.

28.

25.

27.



+1



+1

.
This is a special case of equation 1.7.2.1 with  ( ) =  

+1

+ ' +

=  +  ^  +1  .
This is a special case of equation 1.7.2.7 with  ( ) =   , ( ) =  + ' , and

=  1+

+  

+  1

This is a special case of equation 1.7.1.4 with  ( ) =   , ( ) =  



= 1.

, and % ( ) = ( 

 .

29.

=  1+ +   + 1 .


This is a special case of equation 1.7.1.4 with  ( ) =   , ( ) =    , and % ( ) = ( + .

30.

  1+




This is a special case of equation 1.7.1.4 with  ( ) =   , ( ) =   , and % ( ) = ( + .

31.

=   1+ + +  1 .


This is a special case of equation 1.7.1.4 with  ( ) =   , ( ) = ' + , and % ( ) = (   .

32.

33.

 1+

  1


This is a special case of equation 1.7.1.4 with  ( ) =    , ( ) = ' + , and % ( ) = (  .
= +   ] + +   ] .
This is a special case of equation 1.7.2.6 with  ( ) =  

and ( ) =  + ' + .

2003 by Chapman & Hall/CRC

Page 164

1.5. EQUATIONS OF THE FORM

34.
35.
36.
37.

( , )

= ( , ) CONTAINING ARBITRARY PARAMETERS

165

( + ) =  D +  ] .

This is a special case of equation 1.7.1.15 with  ( ) = , ( ) = 1, and % ( ) = (  .


= ] .
This is a special case of equation 1.7.2.11 with  ( ) =  and = 1.

= ] 2 .
This is a special case of equation 1.7.2.12 with  ( ) =  and = 1.

( +  ) = 1.
1 b . Solution in parametric form with B 1:


 U

U!S ,
U ln L
X U  +1 exp O
B +1
Q
B +1
B + 1 PuM
2 b . Solution in parametric form with B = 1,  1:


 =

 = ln OL  *

 +1

 * ,
P

=U 

1
+1

= * .

3 b . Solution in parametric form with B = 1,  = 1:

38.

( ] +

= 1.

Solution in implicit form:  =


39.

= * .

 = U ln( L 'U ),

( ] + 2 ) = 1.
Solutions in parametric form:

 =

:
1
U (ln )  ,
*
2


L   +
 
1
  (L +  )

= ln O

4  "P

if  1,
if  = 1.

=L

1 0(

U ) + L 2 n 0 (U )

and

:
:
1
U 2
U (ln )  ,
= ln O
,
= L 1 0 (U ) + L 2 ; 0 (U ),
*
2
4  "P
where 0 (U ) and n 0 (U ) are the Bessel functions, and 0 (U ) and ; 0 (U ) are the modified Bessel
functions.
 =

40.

( ] + 1 ) = 1.
A #
$ ,  = T 2#
Let  =

. Solution in parametric form:

 = # V 2UA exp(T U 2 )  (U )W ,

= ln V $ (U )W TU 2 ,

where

 (U ) = X exp(T U 2 ) U + L S
Q
M

41.

( D +  ] + ) = D +  ] .

This is a special case of equation 1.7.1.14 with  ( ) =  , ( ) = 1, and % ( ) = (  .

42.

( D +  ] + ) = D +  ] .

This is a special case of equation 1.7.1.15 with  ( ) =  , ( ) = 1, and % ( ) = (  .

43.

(  ^ + ) = .
This is a special case of equation 1.7.2.3 with  ( ) = (  +  ) 1 .

2003 by Chapman & Hall/CRC

Page 165

166
44.
45.
46.

FIRST-ORDER DIFFERENTIAL EQUATIONS

(  ^ +  ) +   +_ ] +  = 0.

This is a special case of equation 1.7.2.15 with  ( ) =  and ( ) =   .

(  ^ +  ) +  _ ] +  = 0.
This is a special case of equation 1.7.2.15 with  ( ) =  and ( ) = ' + .

(  ^ + ) = (  ^  ).
This is a special case of equation 1.7.2.17 with  ( ) = , ( ) = 1, and % ( ) = '  .

47.

+  ) =  ^ ] .
This is a special case of equation 1.7.2.16 with  ( ) = , ( ) = 1, and % ( ) = ' + .

48.

(  ^ ]
]

  ] )

]

This is a Bernoulli equation with respect to  =  ( ) (see Subsection 1.1.5).

49.
50.
51.
52.
53.
54.
55.
56.

( ] + ) =   ] .
This is a Bernoulli equation with respect to  =  ( ).
( + ] ) =  .
This is a Bernoulli equation with respect to  =  ( ).
( +  ) = ] .
This is a Bernoulli equation with respect to  =  ( ).

( 1 + ) + 1 +  = 0.

This is a special case of equation 1.7.1.19 with  ( ) =  and ( ) = (  .

( 1 +  ] ) + 1 + = 0.
This is a special case of equation 1.7.1.19 with  ( ) =   and ( ) = ( .

( 1 +  ) + 1 +  = 0.

This is a special case of equation 1.7.1.19 with  ( ) =   and ( ) = (  .

[( + ) +  ^ ] = ( + )  ^ .
This is a special case of equation 1.7.2.14 with  ( ) =  , ( ) = 1, and % ( ) = ( + .

[( + ) +  ^ ] ] = ( + )  ^ ] .
This is a special case of equation 1.7.2.13 with  ( ) =  , ( ) = 1, and % ( ) = ( + .

1.5.3. Equations Containing Hyperbolic Functions


1.
2.
3.

= cosh( ) + cosh( b ).
This is a special case of equation 1.7.2.18 with  ( ) = 0, ( ) =  , and % ( ) =  cosh( @! ).

= sinh( ) + sinh( b ).
This is a special case of equation 1.7.2.18 with  ( ) =  , ( ) = 0, and % ( ) =  sinh( @! ).

= cosh( ) + w .
This is a special case of equation 1.7.2.18 with  ( ) = 0, ( ) =   , and % ( ) = ' + .

2003 by Chapman & Hall/CRC

Page 166

1.5. EQUATIONS OF THE FORM

4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.

( , )

= ( , ) CONTAINING ARBITRARY PARAMETERS

= sinh( ) + .
This is a special case of equation 1.7.2.18 with  ( ) =   , ( ) = 0, and % ( ) = ' + .

= 1+ + + sinh( ) 1 .
This is a special case of equation 1.7.1.4 with  ( ) =  , ( ) =  , and % ( ) = ( sinh(  ).

= 1+ + sinh( ) + ^ 1 .
This is a special case of equation 1.7.1.4 with  ( ) =  , ( ) =  sinh(  ), and % ( ) = ( .

= cosh ( sinh 1 + ).
This is a special case of equation 1.7.2.22 with  ( ) =   + ' .

= sinh ( cosh 1 + ).
This is a special case of equation 1.7.2.24 with  ( ) =   + ' .

= ( cosh + ) coth .
This is a special case of equation 1.7.2.25 with  ( ) =  +  .

= ( sinh + ) tanh .
This is a special case of equation 1.7.2.23 with  ( ) =  +  .

( cosh + ) = +1 sinh .
This is a special case of equation 1.7.2.24 with  ( ) = (  +  ) 1 .

( sinh + ) = +1 cosh .
This is a special case of equation 1.7.2.22 with  ( ) = (  +  ) 1 .

( + cosh ) =  .
This is a Bernoulli equation with respect to  =  ( ) (see Subsection 1.1.5).
( + tanh ) =  .
This is a Bernoulli equation with respect to  =  ( ).
( + cosh ) = cosh  ( ).
This is a Bernoulli equation with respect to  =  ( ).

16.

( + tanh ) = tanh ( ).
This is a Bernoulli equation with respect to  =  ( ).

17.

18.

19.

( + ) = cosh ( ).
This is a Bernoulli equation with respect to  =  ( ).

20.

167

+ ) +

+ sinh  (

+ ) +

+ tanh (

) = 0.

This is a special case of equation 1.7.1.19 with  ( ) =  and ( ) = ( sinh  (  ).

) = 0.

This is a special case of equation 1.7.1.19 with  ( ) =  and ( ) = ( tanh  (  ).

( + ) = tanh ( ).
This is a Bernoulli equation with respect to  =  ( ).

2003 by Chapman & Hall/CRC

Page 167

168
21.

FIRST-ORDER DIFFERENTIAL EQUATIONS


( cosh + ) = sinh  ( ).
This is a Bernoulli equation with respect to  =  ( ).

22.

( tanh + ) =  .
This is a Bernoulli equation with respect to  =  ( ).

23.

24.

+ sinh 

+ = 0.

+ tanh

+ = 0.

This is a special case of equation 1.7.1.19 with  ( ) =  sinh 

and ( ) = ( .

This is a special case of equation 1.7.1.19 with  ( ) =  tanh 

and ( ) = ( .

1.5.4. Equations Containing Logarithmic Functions


1.
2.
3.
4.
5.
6.

= (  + ln +  ).
This is a special case of equation 1.7.2.3 with  ( ) = ln + .

= 
1 
+1 ( ln + ln ).
This is a special case of equation 1.7.1.3 with  ( ) =   ln .

= ln2 + ln +  .
This is a special case of equation 1.7.3.1 with  ( ) =   , ( ) = ' + , and % ( ) = (  .

= (  + ln ) +  .
+
This is a special case of equation 1.7.2.4 with  ( ) = ln + .

= ( ln + ln ).
This is a special case of equation 1.7.1.3 with  ( ) = ln .

s  (

ln +

ln ) .

This is a special case of equation 1.7.1.5 with  ( ) =


7.
8.
9.

and ( ) = ln .

( D + ) = D 1 + (ln ln ).
This is a special case of equation 1.7.1.12 with  ( ) =  , ( ) = ( ln , and % ( ) = 1.

(

+  ) = ln + (  ) .
This is a special case of equation 1.7.2.16 with  ( ) = , ( ) = 1, and % ( ) = ln .

+  ) = ( ln + ln  ).
This is a special case of equation 1.7.1.16 with  ( ) =  , ( ) = 1, and % ( ) =  ln .

10.

+ x  ) = ( ln + ln  ).
This is a special case of equation 1.7.1.17 with  ( ) =  , ( ) = 1, and % ( ) =  ln .

11.

12.

( ln +

+ )

+ ln  (

) = 0.

This is a special case of equation 1.7.1.19 with  ( ) =  and ( ) = ( ln  (  ).

= 1.

Solution:  =   




  
M

+L




ln .

Page 168

1.5. EQUATIONS OF THE FORM

13.
14.

( , )

= ( , ) CONTAINING ARBITRARY PARAMETERS

(ln ) = ( '  + ) ln .
This is a special case of equation 1.7.3.7 with  ( ) =   + ' and

169

ln ) = (

ln + ).

= 1.

This is a special case of equation 1.7.3.9 with  ( ) =  , ( ) = 1, and % ( ) = ' + ( .

15.
16.
17.
18.

19.

( + ln ) = ln  ( ).
This is a Bernoulli equation with respect to  =  ( ).

+ ln ) = ( ' I ln ).
This is a special case of equation 1.7.3.10 with  ( ) =  , ( ) = 1, and % ( ) = '  .

' I


ln ) = (

ln ).

This is a special case of equation 1.7.3.9 with  ( ) =  , ( ) = 1, and % ( ) = '  .

+ ln 

= 0.

This is a special case of equation 1.7.1.19 with  ( ) =  ln 

ln

and ( ) = ( .

= ln  ( ).

This is a Bernoulli equation with respect to  =  ( ).

20.

( ln + ln  ) = s .
This is a Bernoulli equation with respect to  =  ( ).

1.5.5. Equations Containing Trigonometric Functions


1.

= sin(

cos( ) +  cos(

).

This is a special case of equation 1.7.4.11 with  ( ) = , ( ) = 0, and % ( ) = cos( ' ).


2.

) cos( ) + cos(

) sin( ).

This is a special case of equation 1.7.1.1 with  ( ) = sin and ( = 0.


3.

= tan( ).
The solution is given by the relation:

exp  a  cos 
 a 
1 2
2

a = L exp   '  ,
1
2

where =

4.

= cos( ) + .
This is a special case of equation 1.7.4.11 with  ( ) = '  , ( ) = 0, and % ( ) = ( + .

5.

6.
7.

sin( ) +

This is a special case of equation 1.7.4.11 with  ( ) = 0, ( ) = '  , and % ( ) = ( + .

= cos ( sin 1 + ).
This is a special case of equation 1.7.4.4 with  ( ) =   + ' .

= sin ( cos 1 + ).
This is a special case of equation 1.7.4.3 with  ( ) =   + ' .

2003 by Chapman & Hall/CRC

Page 169

170
8.

9.
10.
11.

FIRST-ORDER DIFFERENTIAL EQUATIONS


sin2
cos2
+

.
cos2
sin2
This is a special case of equation 1.7.4.14 with  ( ) =  + '

= 1+ + + sin( ) 1 .
This is a special case of equation 1.7.1.4 with  ( ) =  , ( ) =  , and % ( ) = ( sin(  ).
= 1+ + sin( ) + ^ 1 .
This is a special case of equation 1.7.1.4 with  ( ) =  , ( ) =  sin(  ), and % ( ) = ( .

sin(

+ ^ ) = 0.

B = 2: 2 2 

13.

14.

15.
16.
17.
18.
19.
20.
21.
22.
23.

The substitution

12.

=  tan

' + (

2
+ 2( !
 ( 1) '

leads to a Riccati equation of the form 1.2.2.35 with


2

= 0.

= 2 tan( ) + .
 =  tan( ' ).
The substitution =  leads to an equation of the form 1.5.5.3: I

= cos2 + cos sin .


This is a special case of equation 1.7.4.8 with  ( ) = 12 (  +  ).

= sin2 + cos sin .


This is a special case of equation 1.7.4.7 with  ( ) = 12 (  +  ).

= sin  cos2  sin 2 .


This is a special case of equation 1.7.4.18 with  ( ) =  +



and ( ) =  .

(1 + tan2 ) = tan +1 + tan + tan1 .


This is a special case of equation 1.7.4.19 with  ( ) =  , ( ) =  , and % ( ) = (  .

( 1 + ) + 1 + sin  ( ) = 0.
This is a special case of equation 1.7.1.19 with  ( ) =  and ( ) = ( sin  (  ).

( 1 + ) + 1 + tan ( ) = 0.
This is a special case of equation 1.7.1.19 with  ( ) =  and ( ) = ( tan  (  ).

( + ) = cos ( ).
This is a Bernoulli equation with respect to  =  ( ) (see Subsection 1.1.5).
( + ) = tan ( ).
This is a Bernoulli equation with respect to  =  ( ).

( cos + ) = +1 sin .
This is a special case of equation 1.7.4.3 with  ( ) = (  +  ) 1 .

( sin + ) = +1 cos .
This is a special case of equation 1.7.4.4 with  ( ) = (  +  ) 1 .
( + cos ) =  .
This is a Bernoulli equation with respect to  =  ( ).

2003 by Chapman & Hall/CRC

Page 170

1.5. EQUATIONS OF THE FORM

24.
25.
26.
27.
28.
29.
30.

( , )

= ( , ) CONTAINING ARBITRARY PARAMETERS

171

( + cos ) = cos  ( ).
This is a Bernoulli equation with respect to  =  ( ).

( 1 + cos ) + 1 + = 0.
This is a special case of equation 1.7.1.19 with  ( ) =  cos 

( cos + ) = cos  ( ).
This is a Bernoulli equation with respect to  =  ( ).

and ( ) = ( .

( + tan ) =  .
This is a Bernoulli equation with respect to  =  ( ).
( + tan ) = tan ( ).
This is a Bernoulli equation with respect to  =  ( ).

( 1 + tan ) + 1 + = 0.
This is a special case of equation 1.7.1.19 with  ( ) =  tan 

( tan + ) = tan ( ).
This is a Bernoulli equation with respect to  =  ( ).

and ( ) = ( .

1.5.6. Equations Containing Combinations of Exponential,


Hyperbolic, Logarithmic, and Trigonometric Functions
1.
2.
3.

= ] + ln .
+
This is a special case of equation 1.7.2.5 with  ( ) =   and ( ) =  ln  .

= ln ( b )  ] + .
This is a special case of equation 1.7.2.5 with  ( ) =  ln  ( @! ) and ( ) = ' + .

=  ] ( + ln ) .
+
This is a special case of equation 1.7.2.2 with  ( ) =  ln .

4.

=   ( + ln ) .
+
This is a special case of equation 1.7.2.1 with  ( ) =  ln .

5.

= ln2 + ln +  .

This is a special case of equation 1.7.3.1 with  ( ) =  , ( ) =  , and % ( ) = (   .

6.

= ln2 +  ln + ^ .

This is a special case of equation 1.7.3.1 with  ( ) =  , ( ) =   , and % ( ) = ( .

7.

= ] sin + tan .


This is a special case of equation 1.7.5.6 with  ( ) =  +  .

8.

= (  sin + ) tan .
This is a special case of equation 1.7.5.4 with  ( ) =  +  .

9.

=  sin2 +  cos2 .
This is a special case of equation 1.7.5.8 with  ( ) = 12 (  + 
).

2003 by Chapman & Hall/CRC

Page 171

172
10.

11.

12.

13.

14.

15.

FIRST-ORDER DIFFERENTIAL EQUATIONS


cos (

) ]

).

 ]

+ cos (

 ]

+ tan (

This is a special case of equation 1.7.2.5 with  ( ) =  cos  (/ ) and ( ) = ' + .
This is a special case of equation 1.7.2.5 with  ( ) =   and ( ) =  cos + (/ ).

).

This is a special case of equation 1.7.2.5 with  ( ) =   and ( ) =  tan + (/ ).

) ]

tan (

This is a special case of equation 1.7.2.5 with  ( ) =  tan  (/ ) and ( ) = ' + .

B

cos( ) +

(4 

sin( ) +

B-



The substitution = tan( 12  ) leads to a linear equation:  = !$   + !#   .

sin(

) sinh( ) + cos(

) cosh( ).

This is a special case of equation 1.7.2.18 with  ( ) =  sin(/ ), ( ) =  cos(/ ), and % ( ) = 0.


16.

17.

ln2 +

ln + sin (

tan1+

This is a special case of equation 1.7.3.1 with  ( ) =  , ( ) =  , and % ( ) = ( sin  (  ).


(1 + tan2 )

+ tan + 

tan1

This is a special case of equation 1.7.4.19 with  ( ) =  , ( ) =  , and % ( ) = ( ! .


18.

( 

cos + ) = cot .

This is a special case of equation 1.7.5.5 with  ( ) = (  +  ) 1 .


19.

( 

sin + )

= tan .

This is a special case of equation 1.7.5.4 with  ( ) = (  +  ) 1 .


20.

( ] cos + ) = tan .


This is a special case of equation 1.7.5.6 with  ( ) = (  +  ) 1 .

21.

( 

sin + )

= cot .

This is a special case of equation 1.7.5.7 with  ( ) = (  +  ) 1 .


22.

(

 )

 _ ]

ln

23.

(

 )

 _ ]

cos

24.

(

+ 

= 0.

+
This is a special case of equation 1.7.2.15 with  ( ) =  and ( ) =  ln  .



= 0.

This is a special case of equation 1.7.2.15 with  ( ) =  and ( ) =  cos +  .


cos

+  ) +

 _ ]

cos (

) + 

= 0.

This is a special case of equation 1.7.2.15 with  ( ) = cos 

and ( ) =  cos + (  ).

2003 by Chapman & Hall/CRC

Page 172

( , ,

1.6. EQUATIONS OF THE FORM

) = 0 CONTAINING ARBITRARY PARAMETERS

173

1.6. Equations of the Form c ( , , ) = 0 Containing


Arbitrary Parameters
1.6.1. Equations of the Second Degree in 0
1.6.1-1. Equations of the form  ( , )( )2 = ( , ).
1.

( )2 =

See equation 1.6.3.43.


2.

3.

( )2 =

+ .

The substitution = 2
*2 = 4  + 2  .

)2 =

+ 

+ .
3

Solution:  = LA\X ( 
4.

( )2 =

+ ' + ( leads to an Abel equation of the form 1.3.1.2:

+ ( )1 J

+

See equation 1.6.3.26.


5.

6.

( )2 =

+ ,

The substitution  = 2 
*2 =  2  1 J 2 .

( )2 =

+1

The substitution

2(

=2

0.
+   + ( leads to an Abel equation of the form 1.3.1.32:

+1
+ 3)2
+  +

+1

+ .

+1

+ 3)2

2(
2( + 1)

*

 =
 + 2 (
form 1.3.1.10:

( + 3)2
7.

( )

8.

= +

+1

+ S

+ 1)

1 2

leads to an Abel equation of the

+ .

+ .

For 0, the substitution = 2( +  2 + ' + +1 + ( )1 J 2 leads to the Abel equation


*2 = 4  2  + 2  2 ( + 1) + , which is outlined in Subsection 1.3.1 (see Table 5).

Special cases of the original equation are equations 1.6.1.11.6.1.6.


)2 =

+ .

See equation 1.6.3.32.


9.

10.

)2 =

The substitution 
*2 = 2 ( 2 

2 (

)2 =

+ ,
=2 
.

0.
+  + (

leads to an Abel equation of the form 1.3.1.33:

+ .

See equation 1.6.3.34.


11.

2 (

)2 =

2 ) 5

+ .

See equation 1.6.3.28.

2003 by Chapman & Hall/CRC

Page 173

174
12.

FIRST-ORDER DIFFERENTIAL EQUATIONS


(

)(

)2 = 1.

See equation 1.6.3.44.


13.

)(

)2 =

See equation 1.6.3.46.


14.

+ )( )2 = .

See equation 1.6.3.33.


15.

2 (

)2 =

See equation 1.6.3.45.


16.

+ )( )2 =

See equation 1.6.3.35.


17.

)(

)2 = 1.

See equation 1.6.3.27.


18.

2 3 ) 5 + )(

)2 =

See equation 1.6.3.29.


19.

 ]

+ .

( )2 =



( )2 =

)2 =

See equation 1.6.3.3 with , = 2.

20.

See equation 1.6.3.4 with , = 2.

21.



See equation 1.6.3.8 with , = 2.

22.

2 (

)2 =

2  ]

+ .

See equation 1.6.3.9 with , = 2.


23.

( ] +

)( )2 = 1.

See equation 1.6.3.9 with , = 2.

24.

( 

+ )(

)2 =

See equation 1.6.3.8 with , = 2.

25.

( )2 =

+ ln .

See equation 1.6.3.13.


26.

27.

)2 =

ln + ,

The substitution = 2
*2 = 2  2  1 .

( ln +

)(

0.
+  ln  +  leads to an Abel equation of the form 1.3.1.16:

)2 = 1.

See equation 1.6.3.14.

2003 by Chapman & Hall/CRC

Page 174

1.6. EQUATIONS OF THE FORM

( , ,

) = 0 CONTAINING ARBITRARY PARAMETERS

175

1.6.1-2. Equations of the form  ( , )( )2 = ( , )  + % ( , ).


28.

( )2 + + = 0.
Solution in parametric form:

' = 2a  ln a + L ,

29.

30.

= a 2 a .

( )2 + = + .
We differentiate the equation with respect to  , take as the independent variable, and assume
=  to obtain a linear equation with respect to = ( ): ( 2  )  +  + 2 2 = 0.
(

)2 +

= 0.

2
g
The transformation  =  , =  2 leads to an autonomous equation:  g + 2 + 12   =
1 2
1
4  ( ' . Having extracted the root and carried over the terms 2 + 2  from the left-hand

side to the right-hand side, we obtain a separable equation of the form 1.1.2.

31.

32.
33.

34.

35.

36.

= + 2 + ( )2 + ^ + ,
0.
Differentiating with respect to  and changing to new variables a = " and (a ) = 2  , we
arrive at an Abel equation of the form 1.3.1.2: * g  = 4  'a 2!( .

( )2 + ( + ) + = 0,
 L 2 + (
Solutions: = (  +  ) L +

0.
and 4 

( )2 + ( + ) + = 0.
This equation can be factorized: ( + 

= L  and = 21 ' 2 + L .

= 4 ( (  +  )2 .

)(  + ' ) = 0. Therefore, the solutions are:

( )2 + 2 + = 0.
The transformation H = ln  , =  3 leads to an equation independent implicitly of H :
(  )2 + (  + 6 )  + (3  +  + 9 ) = 0. Rewriting the latter equation to solve for  , we obtain
a separable equation of the form 1.1.2.
)2 = 0.
Solution in parametric form:

a
V L +  ln  a + a 2 + 1  W ,
 = 2
a +1

= a


a

)2 + = 0.
1 b . For  1, the solution in parametric form is written as:

 = L3a  +

2 1

2 b . For  = 1, the solution is:


37.

a 2,

a = a 2 +  ,

where

, =

= L3 + 
L .

1
.
 1

)2 + + = 0.
1 b . For  1, the solution in parametric form is written as:

 = L3a |(  + 1)a 2 +  |
There are two singular solutions:

= A"

+2
2( +1)


(  + 1).

 2
(a +  ).
 a

2003 by Chapman & Hall/CRC

Page 175

176

FIRST-ORDER DIFFERENTIAL EQUATIONS


2 b . For  = 1, the solution in parametric form is written as:

 = L3a exp O
38.

)2

a2
,
2 P

= O a +

a P

+ = 0.

Solution in parametric form:

 = L (a  ) exp(a
 ),
= 0.

There is a singular solution:


39.

40.

)2

)2 +

= L3a 2 exp(a
 ).

= 0,

0.

We divide the equation by   and differentiate with respect to  . Passing to the new variables
a =  and (a ) = 2 , we arrive at an Abel equation of the form 1.3.1.33: *pg  = !(a 2 .

= 0.

Solution in parametric form:

a + (  + a 2 ) = 0,

(a 2 +  +  ) + = a  J (

+ )

where

 + 2
.
2(  +  )

There two singular solutions = A"   corresponding to the limit LZD


= 0 is also a singular solution.
41.

)2 + ( )

F . In addition,

= 0.

Solutions: L ( L3 +  ) +  = 0 and (  )2 = 4 ' .


42.

)2 + (

+ )

, L

= 0.

. In addition, there is a singular solution which can be written in


Solution: = L3 +
 L + 
parametric form as:

 =
43.

)2 ( +

) +

45.

)2 + +

,!a
.
a + 

= 0.

L ( +  )
 L 

+ '   )2 4  '

and ( 

= 0.

= 0.

g
The substitution  =  leads to an equation of the form 1.6.1.68: ("g )2 +  ! g +   ( 

2 (

)2 (2

2 (

=  L

 +  L

)2 2

Solutions:

+ )

Solutions:
46.

= a +

Differentiating with respect to  and factorizing, we obtain (2  "  ' +  +  )   = 0.
Equating both factors to zero and integrating, we arrive at the solutions:
= L3 +

44.

,

( a +  )2

+1)

g + = 0.

= 0.
= 41 

and

( 1)

+ 

= L3

1+

= 0.

 , where , =

(  1)
 .

2003 by Chapman & Hall/CRC

Page 176

( , ,

1.6. EQUATIONS OF THE FORM

47.

1)

( )2 + 2

) = 0 CONTAINING ARBITRARY PARAMETERS

177

= 0.

Solution in parametric form:

 = L (a 2 + 1)1 J 2  a + a 2 + 1 
48.

2 (

)2 + (

+ )

= a +  a 2 + 1.

= 0.

The equation can be factorized: ( +  3 )( 2  +  ) = 0. Equating each of the factors to
zero, we obtain the solutions: = A (2  + L )1 J 2 and = 
 + L .

49.

50.

51.

)( )2 2

= 0.

Solving for , differentiating with respect to  , and setting ( ) =  , we obtain a factorized


equation: ( 3 )( 2 2 +  2  2 ) = 0. Equating each of the factors to zero, we arrive
at the solutions:
1 2
2
=
(  L 2 ) and
+  2 =  ( 0).
2L
2 )( )2 + 2

= 0.

The equation can be factorized: (  +   + )(    + ) = 0. Equating each of the


factors to zero, we obtain the solutions: ( +  ) = L and (  ) = L .

+ )(

)2 2

= 0.

Differentiating with respect to  , we obtain a factorized equation: [(


Therefore, the solutions of the original equation are:
=L
52.

3 (

)2 +

=L

)2 (

2
1

where  L

+L

2
2

+  = 0;

'

+

+  )  

]   = 0.

+   = 0.

+ = 0.

Solutions: L3
53.

 + L 2,

 +
2

and 

+ ) +

= 4 .
= 0.

This differential equation represents an equation of curvature lines of a surface defined by the
relation #* 2 + $ 2 + LuH 2 = 1, where  = #2$ ( L $ ),  = #2$ ( # L ), , = L ($ # ).
Solutions:
(  L  )
54.

2 (

)2 + 2

Solutions:

55.

( )

= L (  L  )

+ (1 )
2

( )2 2

+ 

, L

+ ( 1) = 0.

and

 = (  1)( + L )2
+

and

= '

+ 

A 2 , , .

 = 0.

= 0.

Solutions:


56.

)( )2 + 2

Solution: ( L

= L3 + 

+  )2 = 4 

 
L
4

and (   )

'

= (  )  .

= 0.

2003 by Chapman & Hall/CRC

Page 177

178
57.

FIRST-ORDER DIFFERENTIAL EQUATIONS


( 2 2 2 )( )2 + 2 + (1 2 )
Solution in parametric form:

= 0.

L3a
,
 = 2
a +1

58.

( )2 [ 2 ( )2 + 2 ] 2 ( + )2 = 0.
We solve the equation for  ' and differentiate with respect to  . Setting ( ) =   , we
obtain a factorized equation with respect to ( ): (   )[(  2 2 +  2 )3 J 2 A\ , 2 ] = 0.
Equating each of the factors to zero and integrating, we arrive at the solutions:
( ' L )2 + ( 

59.

60.

63.

64.

+1

+ )2 +
=

2 2

B

+ )2

+

+2

67.

' = A/, 2.

( )2 = 0.
2

+ 'B

( + )2 = 2 ( 2 + 2 )  [( )2 + 1].
This equation splits into two equations of the form 1.8.1.4 with  ( ) = A" J 2 :

!   +  = A  (

2 (

) J

 2
( ) + 1.

+ x )2 + 2 ( + )2 = 1.
= L 1   + L 2  + . Here, the constants L
Solution:
2
2
(  L 1 + L 2 )(B )2 = 1.

and L

are related by the constraint

) = ( 2 + 2 )  ( + )2 .
This equation splits into two equations of the form 1.7.1.20 with  ( ) = A  J 2 :
2


) J 2 (!  +  ).

( )2 = 2 ( 2 + 2 )  [( )2 + 1].
This equation splits into two equations of the form 1.8.1.3 with  ( ) = A" J 2 :

   = A  (

66.

and

= 0.

   = A  (

65.

  + L .

 L
= L3  A

Solutions:

62.

L )2 = ,

0.
( + )2 2 + 2 = 0,
2
2
The substitution 2   = leads to the equation "   (  )+  2 = 0. Further assuming
 =  ( ), we obtain ( +  )3 + 2 + 1 = 0. Taking to be the independent variable,
we arrive at a linear equation whose solution is:  = ( 2 + 1)1 J 2 V L  ln  + 2 + 1  W .
Solution:

61.

.
=  L
a 2+1

) J

 2
( ) + 1.

( + + 2 )2 = 4( + 2 + ).
The substitution =  +  2 +  leads to a separable equation:  = A 2 .

( 2 + 2 + 2 )( )2 + ( 1 + 1 + 1 ) + 0 + 0 + 0 = 0.
The Legendre transformation  = g , = a g (  = a ) leads to a linear equation:
[  (a ) + a (a )] g = (a ) + % (a ),

where  (a ) =  2 a 2 +  1 a +  0 , (a ) =  2 a 2 +  1 a +  0 , and % (a ) = ( 2 a 2 ( 1 a ( 0 .

2003 by Chapman & Hall/CRC

Page 178

( , ,

1.6. EQUATIONS OF THE FORM

68.

)2 +

= 0.

+ =

 3 (

)2 .

= 12
 L

Solution:
70.

179


2, solving for   and performing the substitution =  + 2 , we arrive at a
2
 A  2 4   (see Subsection 1.1.2).
separable equation:  = +
2
= 2, solving the original equation for " , we obtain a separable equation:
2 b . With
2


 4    .
2  =  A
1 b . With

69.

 

) = 0 CONTAINING ARBITRARY PARAMETERS

= ( 2 2 )( 

12 L 



).

This is a special case of equation 1.8.1.56 with  ( ) = .


71.

72.

 2 (

+  )2 +

 2_ (

)2 = 1.


= L 1   + L 2  : . Here, the constants L
Solution:
constraint (  L 12 + L 22 )( )2 = 1.

= ( 2 2 ) cosh .

= ( 2 2 ) sinh .

and L

are related by the

This is a special case of equation 1.8.1.58 with  ( ) =  .


73.

This is a special case of equation 1.8.1.59 with  ( ) =  .


74.

cosh sinh )2 + ( sinh cosh )2 = 1.

= L 1 sinh  + L
Solution:
constraint  L 12 + L 22 = 1.

75.

( )2

Solutions: 

76.

cosh  . Here, the constants L

ln( ) = 0.

= exp( L3 L

and L

are related by the

= exp( 41  2 ).

) and 

) cos .

= ( 2 +

= ( 2 + 2 ) sin .

This is a special case of equation 1.8.1.64 with  ( ) =  .


77.

This is a special case of equation 1.8.1.65 with  ( ) = and  = 0.


78.

cos + sin )2 + ( sin cos )2 = 1.

Solution: = L 1 sin  + L
 L 12 + L 22 = 1.
79.

cos  . Here, the constants L

cosh sinh )2 + [(

)2 2 ] +

and L

cos + sin )2 + [( )2 +

Solution:
 L 12 +  ( L

2
1

are related by the constraint

and L

are related by the constraint

= 0.

= L 1 sinh  + L 2 cosh  . Here, the constants L


Solution:
constraint  L 12 +  ( L 12 L 22 ) + ( = 0.
80.

] + = 0.

= L 1 sin  + L 2 cos  . Here, the constants L


+ L 22 ) + ( = 0.

and L

are related by the

2003 by Chapman & Hall/CRC

Page 179

180

FIRST-ORDER DIFFERENTIAL EQUATIONS

1.6.2. Equations of the Third Degree in

1.6.2-1. Equations of the form  ( , )(   )3 = ( , )   + % ( , ).


1.
2.

3.
4.
5.
6.
7.

( )3 + + + = 0.
This is a special case of equation 1.8.1.9 with  ( ) =

( + + )( )3 = 
Dividing both sides by 
of the form 1.7.1.6 with 

9.
10.
11.
12.

+  + .
+  + ( and raising to the power 1/3, we finally arrive at an equation
( ) = 1 J 3 .

)3 + = .
This is a special case of equation 1.8.1.7 with  ( ) = 

(
(

)3 + = .
This is a special case of equation 1.8.1.8 with  ( ) = 

+ .

+ .

)3 + = .
This is a special case of equation 1.8.1.10 with  ( ) = 

)3 + = .
This is a special case of equation 1.8.1.11 with  ( ) =  and ( ) = 

( )3 + 3 = 0,
Solution in parametric form:

0.

 =
8.

a

a 3+1

=L +

4a 3 + 1
.
6 (a 3 + 1)2
2

( )3 + 2 2 = 0.
Differentiating with respect to  and eliminating , we obtain a factorized equation with
respect to ( ) =   : [2(2 )2  2 +  ](9  2 ) = 0. Equating each of the factors to
1
 3 .
zero and integrating, we find the solutions: = 14  L ( L )2 and = 27

)3 + = .
This is a special case of equation 1.8.1.11 with  ( ) = 

3 ) 2 (

)3 + 2 = .
= 2 L  +  L
Solution:

)3 + = .
This is a special case of equation 1.8.1.15 with  ( ) = 

 3 (

and ( ) =  .

)3 + ( + ) + = 0.

= L  + (  L 3 ( ) 
Solution:

1.6.2-2. Equations of the form  ( , )( )3 = ( , )(  )2 + % ( , )  + & ( , ).


13.

)3 + ( )2 = .
This is a special case of equation 1.8.1.7 with  ( ) = 

+

2003 by Chapman & Hall/CRC

Page 180

( , ,

1.6. EQUATIONS OF THE FORM

14.
15.

)3 + ( )2 = .
This is a special case of equation 1.8.1.8 with  ( ) = 

( )3 + ( )2 + + +
Solution in parametric form:

ln(a +  ) + L ,

In addition, there is a singular solution:


16.

17.
18.
19.

20.

21.
22.
23.
24.

25.

26.
27.

=  ' a 3 a 2 .

 = L + 32  a 2 + 2 'a + ( ln |a |,

= a 3 + 'a 2 + (a .

) = .
This is a special case of equation 1.8.1.11 with  ( ) = 
3

+

= 
 .

( ) + ( ) + ^ = + .
Solution in parametric form:
( ) + (

= 0.

2 ' = 3a 2 + 2 a 2 

181

) = 0 CONTAINING ARBITRARY PARAMETERS

)3 + ( )2 = .
This is a special case of equation 1.8.1.11 with  ( ) = 

and ( ) = 

and ( ) = 

( 2 2 )( )3 + ( 2 2 )( )2 + + = 0.
The equation can be factorized: ( + ' )[(  )2 ( 2  2 )+1] = 0, whence we find the solutions:
= 21 ' 2 + L and = A arcsin(
 ) + L .

+ )3 +  3 ( )3 + = 0.

= L 1 L 2  . Here, the constants L
Solution:
 L 13 + L 23 + ( = 0.

and L

are related by the constraint

( )3 + + = 0.
This is a special case of equation 1.8.1.16 with  ( ) = 1, ( ) =  , % ( ) =  , and B = 3.

( )3 + + = 0.
This is a special case of equation 1.8.1.16 with  ( ) = 1, ( ) =  , % ( ) =  , and B = 3.

( )3 + + = 0.
This is a special case of equation 1.8.1.16 with  ( ) = 1, ( ) =  , % ( ) =  , and B = 3.
cosh sinh )3 + ( sinh cosh )3 + = 0.
= L 1 sinh  L 2 cosh  . Here, the constants L 1 and L
Solution:
constraint  L 13 + L 23 + ( = 0.

cosh sinh )3 + ( 2 2 ) + = 0.
= L 1 sinh  + L 2 cosh  . Here, the constants L
Solution:
constraint  L 13 +  ( L 12 L 22 ) + ( = 0.

cosh sinh )3 = ( 2 2 ) cosh .


This is a special case of equation 1.6.4.21 with B = 3 and

sinh cosh )3 = ( 2 2 ) sinh .


This is a special case of equation 1.6.4.22 with B = 3 and

and L

are related by the

are related by the

= 1.

= 1.

2003 by Chapman & Hall/CRC

Page 181

182
28.

29.
30.
31.

FIRST-ORDER DIFFERENTIAL EQUATIONS


cos + sin )3 + ( sin cos )3 + = 0.
Solution: = L 1 sin  L 2 cos  . Here, the constants L 1 and L
 L 13 + L 23 + ( = 0.

are related by the constraint

cos + sin )3 + ( 2 + 2 ) + = 0.
This is a special case of equation 1.6.4.24 with B = 3 and , = 1.

cos + sin )3 = ( 2 + 2 ) cos .


This is a special case of equation 1.6.4.25 with B = 3 and

= 1.

sin cos )3 = ( 2 + 2 ) sin .


This is a special case of equation 1.6.4.26 with B = 3 and

= 1.

1.6.3. Equations of the Form ($ ) d = ( ) + ( )


1.6.3-1. Some transformations.
1 b . In the general case, the equation
(  )  =  ( ) + ( )

(1)

can be reduced with the aid of the transformation a = X [ ( )]1 J'  , = X [  ( )]1 J'
M
M
same form
( g )  =  ( ) + 4 (a ),
where functions 

=  ( ) and 4

to the

= 4 (a ) are defined parametrically by the following formulas:

 ( ) =
4 (a ) =

 ( )

1
,
( )

= X [  ( )]1 J'
a = X [ ( )]1 J'

M
M

 .

2 b . Taking as the independent variable, we obtain from equation (1) an equation of the same class
for  =  ( ):
(  )  = ( ) +  ( ).
3 b . The equation

 = 

+ ( )

 , = 1,  = 

can be reduced with the aid of the substitution ( ) = 2  1


2  2 ( ), which is outlined in Subsection 1.3.1.
4 b . The equation
  = 1 + ( )
( , = 1,  =
is an alternative form of representation of the Abel equation !"
Subsection 1.3.2.

5 b . The equation
  =  + ( )
 , = 1,  = 

to the Abel equation *3

= ( ) + 1, which is outlined in

can be reduced, with the aid of the substitution 

= ( )  followed by raising both sides


M

of the equation to the power of 1 , to an equation of the class in question:


(   )1 J =  + X (  )  .

2003 by Chapman & Hall/CRC

Page 182

( , ,

1.6. EQUATIONS OF THE FORM

6 b . The equation

(   )2 = 

183

) = 0 CONTAINING ARBITRARY PARAMETERS

( , = 2,  =  ,  0)

can be reduced with the aid of the substitution  = 2  + ( ) to an Abel equation of the second
kind:

+ ( )

*  = + Y ( ),

where

Y = 2

  ( ),

which is outlined in Subsection 1.3.1.


7 b . The equation

(   )1 J

=

+ ( )

( , = 1
2,  = 

can be reduced by squaring both sides and performing the substitution H = 


equation:
H   = !H 2 +   .

+ ( ) to the Riccati

For some specific functions = ( ), the solutions of the latter equation are given in Section 1.2.
8 b . The equation

(   )1 J

J + ( )

1 2

=

( , = 1
2,  = 

J )

1 2

can be reduced by squaring both sides and performing the substitution


equation of the second kind:

   =  exp( 21  2  ) ! + 12 exp(  2  )

= exp(  2  )

to an Abel

(see Subsection 1.3.3).


9 b . The equation

(   )1 J

=  ( ) + 

( , = 1
2, =  )

can be reduced by squaring both sides and performing the substitution =  ( ) +  to a Riccati
equation:
 =  2 +   .

For some specific functions  =  ( ), the solutions of the latter equation are given in Section 1.2.
1.6.3-2. Classification tables and exact solutions.
For the sake of convenience, Tables 913 given below list all the equations outlined in Subsection 1.6.3. The five tables classify the equations in which functions  and are of the same form.
The right-most column of a table indicates the numbers of the equations where the corresponding
solutions are given. After the tables follow the equationsthey are arranged into groups so that the
solutions of the equations within each group are expressed in terms of the functions indicated before
the groups as a notation list.
1.

( )  =



Solution:  = X ( #
2.

( )  =

4.

( )  =

+ $ )1 J'

B]

Solution:

(4

 +L .
M

Solution:  = X ( #   + $ )1 J'

( )  =

+L .

= X ( # + $I )1 J'

Solution:
3.

( e

= X ( # + $  )1 J'

+L .

 +L .

2003 by Chapman & Hall/CRC

Page 183

184

FIRST-ORDER DIFFERENTIAL EQUATIONS

TABLE 9
Solvable equations of the form (   )  = #

&

Equation

&

Equation

arbitrary

arbitrary
( , )

,
,

1.6.3.7

1.6.3.23

arbitrary

arbitrary

1.6.3.1

1
2

1.6.3.42

1.6.3.6

1
2

1.6.3.17

1.6.3.2

1
2

1
arbitrary
( 1, 0)

1
arbitrary
(& 1, 0)

1.6.3.38

arbitrary

1 ,
0

1+ ,
arbitrary

arbitrary

1.6.3.5

1
2

1.6.3.46

1
2

1.6.3.37

1.6.3.33

1.6.3.20

1.6.3.29

1
2

1.6.3.27

1.6.3.22

1.6.3.30

1.6.3.11

1.6.3.16

2
5
1
2

1.6.3.35

2
arbitrary
( 0)
arbitrary
( 2, 0)

1.6.3.44

1
2
1
2

1.6.3.10

1
2

1
2

1.6.3.24

1.6.3.15

1
2

1.6.3.41

1.6.3.21

1.6.3.31

2
5

1.6.3.45

1
2

1.6.3.36

1.6.3.34

1.6.3.12

1.6.3.32

1.6.3.40

1
2

1.6.3.25

1
1

1
1

1
2

1
2
1
2

1
2

1.6.3.39

1.6.3.28

1.6.3.26
1.6.3.43

( )  =  + ( .
Solution in parametric form:

 = X ( #*U 1 'J  + $ )1 U + L ,
M
6.

arbitrary
( , 1, 1)

5.

+ $I

( )  =  1  + ( 1+  ,
Solution in parametric form:

 = l X
where # = 

1+

 




+L
M

= U 1 'J  +

U $X ( #*U 1 'J  + $ )1 U $L2 .


M

| | 1.

 $ , $ = 

1+

+1

= / U X

hL
M

= U 1 'J  +





  ( , 1) =" .
 ( , + 1)

2003 by Chapman & Hall/CRC

Page 184

1.6. EQUATIONS OF THE FORM

( , ,

185

) = 0 CONTAINING ARBITRARY PARAMETERS

TABLE 10
Solvable equations of the form

(  )  = #  + $I

TABLE 11
Solvable equations of the form

+$ 
(  )  = #

&

Equation

arbitrary

1.6.3.9

arbitrary

1.6.3.8

arbitrary

1.6.3.3

arbitrary

1.6.3.4

1.5.2.40

arbitrary

1.5.2.37

1.5.2.38

1
2

1.6.3.18

1
2

1.5.2.39

1.5.2.14

1.6.3.19

arbitrary

1.5.2.13

7.

8.

Equation

1.5.2.8

1.5.2.2

( )  =  s + (  s ,

Solution in parametric form:

)
(

 =L 

=L 

+4
( 

=  

Equation

(  )2 = # ln + $I

1.6.3.14

(  )1 = # ln + $I
(  )2 = # + $ ln 

s.

1.5.4.12
1.6.3.13

1

,
O# + $IU   U!

P
1

,
O# + $IU   !U
U exp X

M
1

U ,


U .

.
Solution in parametric form:
1
( # + $   * )1 J' S

 = exp U

( )1 =



Solution:  = 

=X
.

P  O# X

( )  = B] + (  .
Solution in parametric form:

1
( $ + #   * )1 J' + S

+L

U +

U +L ,
M

1
( # + $   * )1 J' S

1
( $ + #   * )1 J' + S

 P 

= exp U +

10.

Form of equation

exp X

 =X

9.

Equation

TABLE 13
Solvable equations containing
logarithmic functions

TABLE 12
Solvable equations of the form

(  )  = #  + $ 

U +L .

2003 by Chapman & Hall/CRC

Page 185

186

FIRST-ORDER DIFFERENTIAL EQUATIONS

In the solutions of equations 1114, the following notation is used:

11.

=  1 ) 2 + ( 1 .
Solution in parametric form:

=  [2U2AC exp(T"U 2 )] ,

( )1 =  1 + ( 1 ) 2 .
Solution in parametric form:

 =  [2U2AC exp(T"U 2 )] ,

13.

=  exp(T U 2 ),

( )2 =  + ( ln .
Solution in parametric form:

where # = 4 

# = A 2

where

# = T 4 , $ = A 2 1 J 2 

 , $ = T 4 # .

( )2 =  ln + ( .
Solution in parametric form:

 = , [2U2AC exp(T"U 2 )] A 4 ln(  ) 4U 2 . ,


where # = T 4 !$ , $

 J , $ = T 4 .

1 1 2

where

= f, [2U2A exp(T U 2 )] A 4 ln( ! ) 4U

 = ! exp(T U 2 ),
14.

 = ! exp(T"U 2 ),

12.

 = X exp(T"U 2 ) U + L2S
Q
M

= 4  

=  exp(T U 2 ),

In the solutions of equations 1519, the following notation is used:

n (U ) for the upper sign,


(U ) + L 2
L 1  (U ) + L 2 ; (U ) for the lower sign,
1

where (U ) and n (U ) are the Bessel functions, and  (U ) and ; (U ) are the modified Bessel
functions.
x3y'z|{}~/
: :
:
The solutions of equations 1519 contain only the ratio 
= (ln )  . Therefore,
*
*
for the sake of symmetric appearance, two arbitrary constants L 1 and L 2 are indicated in the
:
definition of function (instead, we can set, for instance, L 1 = 1 and L 2 = L ).
15.

s 2, s 0.
( )1 =  s + ( 2 ,
Solution in parametric form:

 = U
where @ =
16.

1
+2
, # =T
 
+2
2

2
1

1,
( )1 ) 2 =  + ( e ,
Solution in parametric form:

V U (ln : )  + @W ,

= 'U

, $

+2
2

1 1

0.

U (ln : )  + @2A & + 1 U 2 S ,


*
2&
Q
1
J
2
1
(& + 1) 
& + 1
, # =  1
S , $ =T
 # .
where @ =
& +1
Q
2
2&
 = U 2 ,

= 'U

2003 by Chapman & Hall/CRC

Page 186

1.6. EQUATIONS OF THE FORM

17.

( , ,

s 1, s 0.
( )1 ) 2 =  s + ( ,
Solution in parametric form:

:
 = U 2 U (ln

Q
1
+1
, # =T
  $ ,
where @ =
+1
2

18.

+1 2

U S ,
= 'U 2 ,
*
2
( + 1)  1 J 2
S .
$ =  1
Q
2

)  + @*A

( )1 ) 2 =  + (4 .
Solution in parametric form:

:
=  V U (ln )  A

 = ln( U 2 ),
where @ = 0, # = 
19.

1J 2
 12   , $ = T

1 1
2

( )1 ) 2 = B] + ( .
Solution in parametric form:

where @ = 0, # = T

1
2

 

$ , $ =

1 2
2

U W,

:
 =  V U (ln )  A

187

) = 0 CONTAINING ARBITRARY PARAMETERS

# .

1
2

U 2W ,

= ln( 'U 2 ),

1J 2
 21   .

In the solutions of equations 2035, the following notation is used:

J U ) + L 2 n 1 J 3 (U ) for the upper sign,


1 J 3 ( U ) + L 2 ; 1 J 3 ( U ) for the lower sign,

1 1 3(
1

where 1 J 3 (U ) and n 1 J 3 (U ) are the Bessel functions, and 1 J 3 (U ) and ; 1 J 3 (U ) are the modified Bessel
functions;
:
:
:
:
1 = U  + 13 ,
2 = 12 ACU 2 2 ,
3 = A 23 U 2 3 2 1 2 .

x3y'z|{}~/

The solutions of equations 2035 contain only the ratio 


= (ln )  . Therefore,
*
*
for the sake of symmetric appearance, two arbitrary constants L 1 and L 2 are indicated in the
:
definition of function (instead, we can set, for instance, L 1 = 1 and L 2 = L ).
20.

21.

22.

23.

=  1 + ( 2 .
Solution in parametric form:

J :

 = U

4 3

 = U

2 3

2,

( )1 =  2 + ( 1 .
Solution in parametric form:

J :

J :

2 3

= 'U

1
2

3,

where

# = 2

J :

2,

where

# =T

1
2

1,

= 'U

4 3

J :

2,

where

# =T

2,

= 'U

2 3

J :

1,

where

# = 2

3,

=  1 + ( .
Solution in parametric form:

= 'U

4 3

1 2

2
3

 , $ =T

2
3

  .

  , $ = 2  2 

J :

 = U

2 3

 = U

4 3

( )1 =  + ( 1 .
Solution in parametric form:

J :

2 1 2
,
3

$ = 2

 , $ =T

.

2 2 1
.
3

 

2003 by Chapman & Hall/CRC

Page 187

188
24.

FIRST-ORDER DIFFERENTIAL EQUATIONS

 1 )

1 2

Solution in parametric form:

25.

( )1 =

J :

4 3

 = U



1 2

12 ,

= 'U

( 1)

8 3

J :

22 ,

where

# = 2

J :

8 3

26.

( )2 =



J :

12 ,

where

# =T

22 ,

( 1)

27.

J :

4 3

)2 =

J :

J :

8 3

= 'U

2
2

2
2

4 2
3

8 3

 = U
4
3

28.

( )2 =



J $ , $ = 4  

 

1 2

2
3

J , $ = 2 1J 2 

1 2

 

1 ,

# = 4

where

 , $ =T

J # .

4 1 2
3

Solution in parametric form:

where # = T

12,

 1 )

4 3

= 'U

Solution in parametric form:

 = U

J .

2 1 2
3

Solution in parametric form:

 = U

 J , $ =T

1 1 2

2 5

4
3

1 ),

= 'U

4 3

= 'U

J :

4 3

12 ,

Solution in parametric form:

 = U

)2 =

J 5J 2,
1

5 2

J  $ , $ =

4 2 5 2
3

where # = T
29.

J :

5 3



2 5

J :

2
2

4
3

1 )1 J 2 ,

J 

16 8 5 2
.
25

Solution in parametric form:

 = U

30.

 1 )

16 2 8 5
,
25

where # =

 

J :

4 3

$ =T

2
2

31.

( )1 =



32.

( )2 =

J :

4 3

( 1)



= 'U
2

1
2 ,

1 )1 J 2 ,

= 'U

J :

5 3

J 5J 2,
1

5 2

J :

4 3

2
2

 J

2
2

32 ,

where

# =A

4 1 1 2
,
3

1
2 ,

where

# = 4  , $ = A

$ = 4   .

32 ,

= 'U 4 J

 J 

4 1 2 1
.
3

Solution in parametric form:

:
 = U 4 J 3

1
2 ,

J # .

 

Solution in parametric form:

 = U

4 2 2 5
3

Solution in parametric form:

:
 = U 4 J 3

4
3

= 'U

J :

4 3

2
2
2 ( 3

3
2 ),

where

# =

16 2
9

 , $ = 4  # .

2003 by Chapman & Hall/CRC

Page 188

( , ,

1.6. EQUATIONS OF THE FORM

33.

( )2 =  1 + ( .
Solution in parametric form:

 = U

34.

J :

4 3

1 2
,
2

= 'U

J :

2 3

= 'U 4 J

3
2 ),

J :

2 3

( )2 =  2 + ( 2 .
Solution in parametric form:

 = U
]

2
2
2 ( 3

( )2 =  2 + ( 2 .
Solution in parametric form:

:
 = U 2 J 3

35.

1
2
2 ( 3

= 'U 2 J

1
2
3 1 2
,
2 ( 3 4 2)

1
2 ,

3 1 2
,
2)

1 2
,
2

# = 4 $ , $ =

where

where # = 4 2  2 $ , $





1
2

an L 2 . One of
= A 1), while the

( )1 =  2 + ( 2 .
Solution in parametric form:

where # =

1 Te@
2

  , $ = 21 

= 'U 2 ,

1 1

@ = |1 4 #2$ |,

( )1 ) 2 =  + ( 1 .
Solution in parametric form:

 = U 2 ,

= 'U

O

( )1 ) 2 =  1 + ( .
Solution in parametric form:

 = U
39.

for the upper sign,


L 1 U + L 2 U
L 1 sin( @ ln U ) + L 2 cos( @ ln U ) for the lower sign,
L 1 ln U + L 2
for @ = 0,

(1 + @ ) L 1 U + (1 @ ) L 2 U
for the upper sign,
( L 1 @ L 2 ) sin( @ ln U )+( L 2 + @ L 1 ) cos( @ ln U ) for the lower sign,
L 1 ln U + L 1 + L 2
for @ = 0.

 = U

38.

$ = 4 2  2 # .

 

The expressions of and contain two arbitrary constants L


them can be fixed to set it equal to any nonzero number (for example, we can set L
other constant remains arbitrary.

37.

 

16 4 2
.
9

16 2 4
,
9

# =

where

16
9

In the solutions of equations 3646, the following notation is used:

x3y'z|{}~/

36.

189

) = 0 CONTAINING ARBITRARY PARAMETERS

O

1 Te@
2

1 Te@
2

where

# =

= 'U 2 ,

where

# =

1 Te@
2

=  1 + ( 1 ) 2 .
Solution in parametric form:

1 2

2 rP

, $

 $ , $ = 

1 Te@
2

2 "P

 # .

1 2

 = U 2 2 ,

= 'U ,

where

# = (1 A@ 2 )

2

$ =

J .

1 2

2003 by Chapman & Hall/CRC

Page 189

190
40.

FIRST-ORDER DIFFERENTIAL EQUATIONS


( )1 =  1 ) 2 + ( 1 .
Solution in parametric form:
= 'U 2

 = U ,
41.

= 'U 2

= 'U 2 [

$ = (1 A@ 2 )

where

# = 2(1 Ae@ 2 ) 

= 'U ,

where

# = 4 

(1 Ae@ 2 )

(1 A@ 2 )

( )2 =  2 + ( 1 .
Solution in parametric form:
= 'U [

],

 = U [

(1 A@ 2 )

2 1 2

2

= 'U ,

where

# = (1 Ae@ 2 )  

= 'U 2

.

 , $ = (1 Ae@ 2 ) 

# = 16

2 1 2

$ = 4

$ = 2(1 A@ 2 )  1 J 2 

where

(1 A@ 2 )

( )2 =  1 + ( 2 .
Solution in parametric form:

 J ,

1 1 2

],

( )2 =  2 + ( .
Solution in parametric form:

 = U 2 2 ,

46.

J ,

1 2

# =  

( )2 =  + ( 2 .
Solution in parametric form:

 = U 2 [

45.

( )1 =  + ( 1 ) 2 .
Solution in parametric form:

 = U ,

44.

where

=  1 ) 2 + ( .
Solution in parametric form:

 = U 2

43.

 = U ,

42.

# .

$ , $ = 16  

where

# = (1 Ae@ 2 ) 

1 2

1 2

where

# =  2

= (1 A@ 2 )  2 

, $

 $ , $ =

 .

# .

1.6.4. Other Equations


1.6.4-1. Equations containing algebraic and power functions with respect to  .
1.

2.

= + 2 + + ,
0.

Differentiating the equation with respect to  and changing to new variables a =  and
(a ) = 2  , we arrive at an Abel equation of the form 1.3.1.32: *pg  =  'a 1 J 2 .

= ( + ) ( )2 + 1.

This is a special case of equation 1.8.1.5 with  ( , ) =  + ' .

3.

+ = ( + ) ( )2 + 1.

This is a special case of equation 1.8.1.6 with  ( , ) =  + ' .

4.

+ ( ) .

= 3
L  +  L  . In addition, there is a singular solution:
!#  1 B  = (B 1)  1 , B 1.

Solution:

= #* 

, where

2003 by Chapman & Hall/CRC

Page 190

1.6. EQUATIONS OF THE FORM

5.
6.
7.

8.

( , ,

= + ( ) .
This is a special case of equation 1.8.1.15 with  ( ) =  p+ .

= ( )2 + 2 .
This is a special case of equation 1.8.1.51 with  ( ) =   .

= + 2 + ( )2 + ( ) +1 + ,
0.
Differentiating the equation with respect to  and passing to the new variables a =  and
(a ) = 2  , we arrive at the Abel equation *ug  = 4  'a 2!( ( + 1)a + , whose solvable

case are outlined in Subsection 1.3.1.

( )

+ ( ) = .
1 b . Solution in parametric form with B 1,

 = a  + 'a + ,

 =

( )


a

+
+ 'a ,

B 
a
B 1

 = L +  ln |a | +

11.

B +1

a

+1

( )

+1

+1

a+

a+

+1

+1

1:

1:

= L +  ln |a | +

2 b . Solution in parametric form with B = 1,

10.

B

=L +

+ ( ) = .
1 b . Solution in parametric form with B 1,

 =L +

1:

2 b . Solution in parametric form with B = 1,

9.

191

) = 0 CONTAINING ARBITRARY PARAMETERS

a+

+
= a  + 'a .

1:

a+

= a + 'a + .

+ ( ) + ( )  = 0.
This is a special case of equation 1.8.1.12 with  ( ) =   , ( ) = ' + , and % ( ) = (  .
= ( ) .
g ,
The Legendre transformation  = u
2

+
2


a =
( g ) .

g
=a2

1 b . Solution in parametric form with

B , B 1:

1
1

 + B a O a  + L2 + +  ,
 =O
 P
B +1  P

2 b . Solution in parametric form with


1

 =

= 1, B = 1:

+
+ +

= B , B 1:

a 
B
a 
= L aO
1 exp
aO
.
 P

B + 1 P
B , B = 1:

+
 V
 (1 ) ln |a | + L3W 1 + ,

4 b . Solution with

a 
a 
1
+B
a O
=
L2u
a O
+ L2
1+B
 P
B +1  P

a 
B
a 
 = LO
exp
aO
,
 P

B + 1 P
3 b . Solution in parametric form with

(  = a ) leads to a separable equation:

= L3

= a V  (1

) ln |a | + L3W

1
1

+ .

1 .

2003 by Chapman & Hall/CRC

Page 191

192
12.

FIRST-ORDER DIFFERENTIAL EQUATIONS

  (

+ =

=
 L 

Solution:
13.

14.

1
2


+ ( )

+ (

1
1

+B

constraint  1 L

16.

( + )

= 0.

  1

2B 1 +
1
+  2L 2

and L

22.

23.

and L

2.

are related by the

= 0.

  ( )

  (

  _ (

)  + [(
1
2

= 0.

. Here, the constants L

)2

1
2

= L 1 L 2
L 2 )  + ( = 0.

+ )  +

] +

 ` (

and L

are related by the constraint

= 0.

. Here, the constants L


+  ) + = 0,

and L

L 1 : 
L 2 


 ; . Here, the constants L
=
=
constraint  L 1 + L 2 + ( = 0.

(
(

cosh sinh ) + ( sinh cosh )  +

cosh sinh ) = ( 2

sinh cosh ) = ( 2

and L

are related by the

= 0.
1

and L

are related by the

and L

are related by the

cosh sinh ) + ( 2 2 )  + = 0.

are related by the constraint

= L 1 sinh  + L 2 cosh  . Here, the constants L


Solution:
constraint  L 1 +  ( L 12 L 22 )  + ( = 0.
21.

= 0,

= L 1 sinh  L 2 cosh  . Here, the constants L


Solution:
constraint  L 1 + L 2 + ( = 0.
20.

are related by the constraint

  2 . Here, the constants L

2B
+  = 0.

2 + 2 )

+ (

+ ) +

Solution:

19.

12 L 

The contact transformation m =  +  , n = (  )2 + 2  , n o  = 2  , where n = n (m ),


leads to a separable equation: n o  = 2m  ( n + + ( ).
The inverse contact transformation:  = 12  1 (2m n o  ), = 18  1 [4 n ( n o  )2 ],  = 12 n o  .

Solution:
 L 2 +  ( L
18.

+ )

= L 1L
Solution:
 L 1 + L 2 + ( = 0.
17.

= L 1   + L 2 . Here, the constants L


Solution:

 ( L 1 B ) +  ( L 2 B ) + + ( = 0.

Solution:

15.

)

cosh .

The contact transformation m = ( )2 2 , n =  cosh  sinh  , n o  =


leads to a separable equation: 2 'm + n o  = n  .
)

cos + sin ) + (

Solution: = L 1 sin  L
 L 1 + L 2 + ( = 0.

cosh  (  )1

1
2

sinh  (  )1

sinh .

The contact transformation m = ( )2 2 , n =  sinh  cosh  , n o  =


leads to a separable equation: 2 'm + n o  = n  .

1
2

sin cos )  + = 0.

cos  . Here, the constants L

and L

are related by the constraint

2003 by Chapman & Hall/CRC

Page 192

1.6. EQUATIONS OF THE FORM

24.

25.

26.

( , ,

) = 0 CONTAINING ARBITRARY PARAMETERS

cos + sin ) + ( 2 + 2 )  + = 0.
Solution: = L 1 sin  + L 2 cos  . Here, the constants L
 L 1 +  ( L 12 + L 22 )  + ( = 0.

and L

193

are related by the constraint

cos + sin ) = ( 2 + 2 ) cos .


The contact transformation m = ( )2 + 2 , n =  cos  + sin  , n o  =
to a separable equation: 2 'm + n o  = n  .

sin cos ) = ( 2 + 2 ) sin .


The contact transformation m = ( )2 + 2 , n =  sin  cos  , n o  =
to a separable equation: 2 'm + n o  = n  .

1
2

cos  (  )1 leads

1
2

sin  (  )1 leads

1.6.4-2. Equations containing exponential, logarithmic, and other functions with respect to   .

27.

= exp( ) + exp(
).
This is a special case of equation 1.8.1.7 with  ( ) =  exp( ) +  exp( ).

28.

29.
30.

= exp( ) + exp(
).
This is a special case of equation 1.8.1.8 with  ( ) =  exp( ) +  exp( ).
= + exp( ).
This is a special case of equation 1.8.1.15 with  ( ) =  exp( ).

= exp( ) + exp(
).
This is a special case of equation 1.8.1.11 with  ( ) =  exp( ) and ( ) =  exp( ).

31.

= sinh( ) + sinh(
).
This is a special case of equation 1.8.1.7 with  ( ) =  sinh( ) +  sinh( ).

32.

= sinh( ) + sinh(
).
This is a special case of equation 1.8.1.8 with  ( ) =  sinh( ) +  sinh( ).

33.
34.

= + sinh ( ).
+
This is a special case of equation 1.8.1.15 with  ( ) =  sinh ( ).

= sinh( ) + sinh(
).
This is a special case of equation 1.8.1.11 with  ( ) =  sinh( ) and ( ) =  sinh( ).

35.

= cosh( ) + cosh(
).
This is a special case of equation 1.8.1.7 with  ( ) =  cosh( ) +  cosh( ).

36.

= cosh( ) + cosh(
).
This is a special case of equation 1.8.1.8 with  ( ) =  cosh( ) +  cosh( ).

37.
38.

= + cosh ( ).
+
This is a special case of equation 1.8.1.15 with  ( ) =  cosh ( ).

= cosh( ) + cosh(
).
This is a special case of equation 1.8.1.11 with  ( ) =  cosh( ) and ( ) =  cosh( ).

2003 by Chapman & Hall/CRC

Page 193

194
39.

FIRST-ORDER DIFFERENTIAL EQUATIONS


ln

= 0.

1 b . Solution in parametric form with  0,  1:


1

 =

+ L3a

a

1
= (a + ln a +  ).

1
+1 ,

2 b . Solution in parametric form with  = 0:

 =

ln a + 

= L + (  1) ln a +

1
(ln a )2 .
2

3 b . Solutions with  = 1:

40.
41.

42.
43.

= L3 + ln L + 
+ ln + ( + )  +

and

= ln(1
 ) +  1.

= 0.

This is a special case of equation 1.8.1.41 with  ( , ) = ln +   +  .

+ ln + ,

0.

Differentiating the equation with respect to  and changing to new variables a =  and
(a ) = 2  , we arrive at an Abel equation of the form 1.3.1.16: *pg  = 2  'a 1 .
ln ( ).

+
This is a special case of equation 1.8.1.15 with  ( ) =  ln ( ).

sin( ) + sin(
).

This is a special case of equation 1.8.1.7 with  ( ) =  sin( ) +  sin( ).

44.

= sin( ) + sin(
).
This is a special case of equation 1.8.1.8 with  ( ) =  sin( ) +  sin( ).

45.

46.
47.
48.
49.
50.
51.

sin ( ).

This is a special case of equation 1.8.1.15 with  ( ) =  sin + ( ).


sin( ) + sin(
).

This is a special case of equation 1.8.1.11 with  ( ) =  sin( ) and ( ) =  sin( ).

cos( ) + cos(
).

This is a special case of equation 1.8.1.7 with  ( ) =  cos( ) +  cos( ).

cos( ) + cos(
).

This is a special case of equation 1.8.1.8 with  ( ) =  cos( ) +  cos( ).


+

cos (

).

This is a special case of equation 1.8.1.15 with  ( ) =  cos + ( ).


cos(

) + cos(

).

This is a special case of equation 1.8.1.11 with  ( ) =  cos( ) and ( ) =  cos( ).

tan (

).
This is a special case of equation 1.8.1.15 with  ( ) =  tan + ( ).

2003 by Chapman & Hall/CRC

Page 194

1.7. EQUATIONS OF THE FORM

( , )

= ( , ) CONTAINING ARBITRARY FUNCTIONS

195

1.7. Equations of the Form ( , ) = ( , )


Containing Arbitrary Functions
]

Notation:  , , and % are arbitrary composite functions whose argument, indicated after the
function name, can depend on both  and .

1.7.1. Equations Containing Power Functions


1.

= ! ( + + ).
In the case  = 0, we have an equation of the form 1.1.1. If  0, the substitution ( ) =
 +  + ( leads to a separable equation:  =  ( ) +  .

2.

= ! ( + + ) 1 .
The substitution = +   +  leads to a separable equation:  =  ( ).

3.

! (

).

Generalized homogeneous equation. The substitution H =   +


"H   = BH + H! ( H ).

leads to a separable equation:

4.

5.

= ! ( ) 1+

+ " ( ) + % ( ) 1 .
 = B ( )
The substitution =  leads to a Riccati equation: 3

 ! ( )" (

The substitution H =   +
6.

=I
! O

+ B ( ) + B% ( ).

).

leads to a separable equation: H   =

+ +
.
+  + P

  
 ++
+  ( ) H +

( H ).

= (
( =
, = ( ) +
leads to
1 b . For =   0, the transformation  = +

an equation:
 + '
 =  O
.
+ P
Dividing both the numerator and denominator of the fraction on the right-hand side by , we
obtain a homogeneous equation of the form 1.1.6.
2 b . For = 0 and  0, the substitution ( ) =  +  + ( leads to a separable equation of
the form 1.1.2:
'
  =  +  O
.
+ = (CP
3 b . For = 0 and 0, the substitution ( ) =  + + = also leads to a separable
equation:
' + ( =
  = + h*O
.

7.
8.

= 1 1 ! ( + ).
 =  
   +  + leads to a separable equation: 3
The substitution = 

+ + " ( ) ! ( +1 +
The substitution =  +1 +  

+1

[ B + 

 ( )].

+1

) = 0.
 + (B + 1) ( )  ( ) = 0.
leads to a separable equation: 3

2003 by Chapman & Hall/CRC

Page 195

196
9.

FIRST-ORDER DIFFERENTIAL EQUATIONS

! (

)+

" ( )]

= % ( ).

This is a Bernoulli equation with respect to  =  ( ) (see Subsection 1.1.5).


10.

! (

) + " ( )]

= % ( ).

This is a Riccati equation with respect to  =  ( ) (see Section 1.2).


11.

= [ ! ( ) + " ( )] ( )( ).

The substitution
  ( ) ( ).
12.

= (  )
(  ) leads to a Riccati equation: A 2  = [  ( ) + ( )]

!IO
P

D-%IO S
P

=3
" O

D 1 I
% O

The substitution = a leads to a Bernoulli equation with respect to  =  (a ): [ (a ) a (a )] g =


 (a ) + % (a ) +1 .
13.

h ( , )]
.
Darboux equation. Here,

[g

( ,

)+

i ( ,

)+

h ( ,

).

 and  are homogeneous polynomials of order B , and + is


a homogeneous polynomial of order . Dividing the Darboux equation by   leads to an

equation of the form 1.7.1.12.


14.

[ ! (

" (

)+

)]

= % (

" (

).

  +  leads to a linear equation with respect to 


The substitution a = 
[   (a ) + % (a )] g =  (a ) +  (a ).
15.

[ ! (

" (

)+

 " (

)] = [ % (

 " (

)] = [ % (

The substitution a =  + 


 (a ) + % (a ).
16.

17.

[ ! (

)+

[ ! (

) + x

)]

= % (

) " (

=  (a ):

).

leads to a linear equation with respect to = (a ): [   (a )+ % (a )] g =

 " (

)].

 " (

)].

!

The transformation a =   + , H =   leads to a linear equation with respect to H = H (a ):


a [B (a ) + % (a )]H g  = ,  (a ) H , (a ).

18.

The transformation a =   + , H =  leads to a linear equation with respect to H = H (a ):


a [B (a ) + % (a )]H g  = , % (a ) H + ,!B (a ).

 s

w
 s

19.

[ ! ( ) +

[s ! (

) j" (

)] = [#" (

The transformation a =   + , =  

Solution: X ( )
20.

= ! (

+ " ( ) +

)].

g  = ( ).
leads to a separable equation: a (a )u

+ Xw ( )  +  

= 0.

+ =L .

+ 2 )( + ).

Setting  = & (a ) cos a ,


a = X& 1  (& 2 ) & + L .

^_

= & (a ) sin a and integrating, we obtain a solution in implicit form:

Reference: G. W. Bluman, J. D. Cole (1974, p. 100).

2003 by Chapman & Hall/CRC

Page 196

( , )

1.7. EQUATIONS OF THE FORM

21.

= ( , ) CONTAINING ARBITRARY FUNCTIONS

197

= ! ( 2 2 )( ).
Setting  = & (a ) cosh a , = & (a ) sinh a and integrating, we obtain a solution in implicit form:
a = X & 1  (& 2 ) & + L .

22.

23.

[ ! ( 2 + 2 ) + " ( 2 + 2 ) + % ( 2 + 2 )]( + ) = .
The transformation  = & cos Y , = & sin Y leads to an equation of the form 1.7.4.11 with
respect to Y = Y (& ): Y  =  (& 2 ) cos Y + (& 2 ) sin Y + & 1 % (& 2 ).

[ ! ( 2 2 ) + " ( 2 2 ) + % ( 2 2 )]( ) = .
1 b . For  > , the transformation  = & cosh Y , = & sinh Y leads to an equation of the form
1.7.2.18 with respect to Y = Y (& ): Y  =  (& 2 ) cosh Y (& 2 ) sinh Y & 1 % (& 2 ).
2 b . For  < , the transformation  = H sinh j , = H cosh j leads to an equation of the form
1.7.2.18 with respect to j = j ( H ): j  =  ( H 2) sinh j ( H 2) cosh j H 1 % ( H 2 ).

24.

! 

Setting  = & (a ) cos a ,

! 

 "I

Setting  = & (a ) cosh a ,

! ( , )

+ " ( , ) = 0,

(cos a , sin a )

 (& 2 )
& +L .
& M

( ).

= & (a ) sinh a and integrating, we obtain the solution:

26.

25.

( + ).
+
+ 2

= & (a ) sin a and integrating, we obtain the solution:

 "I

(cosh a , sinh a )
where

k
k

 (& 2 )
& =L .
& M

"

Exact differential equation.




Solution: X  ( 0 , a ) a + X  (a , ) a = L , where 
M
M
0
0

and

are arbitrary numbers.

1.7.2. Equations Containing Exponential and Hyperbolic Functions


1.
2.
3.

4.

=   ! (  ).

The substitution =  

leads to a separable equation:  =  ( ) + .

=  ] ! (  ] ).
The substitution =    leads to a separable equation:   =

 ( ) + .

= ! (  ^ ).
This equation is invariant under translationdilatation transformation. The substitution

H =  9 + leads to a separable equation: H   = hH + H! ( H ).


! ( ^]
1

).

This equation is invariant under dilatationtranslation transformation. The substitution


H =    9  leads to a separable equation: "H   = BH + hH! ( H ).

2003 by Chapman & Hall/CRC

Page 197

198
5.

6.

FIRST-ORDER DIFFERENTIAL EQUATIONS

= ! ( )

+ " ( ).

]

The substitution =    leads to a linear equation:  = ( ) " ( ).

]

+ ! ( )" (

).

The substitution H =    leads to a separable equation: H   =  ( ) H ( H ).


7.

8.
9.
10.

11.

 ! ( )" (  ^

).


The substitution H =  9 + leads to a separable equation:
 ++

H   = exp (1 , ) S  ( ) H +

= ! ( ) ] + " ( ) + % ( )

= ! ( +

]

( H ).

The substitution =  leads to a Riccati equation:  = " ( )

 ^ _ l] ! (  ^

 _ ]

+ ( ) + "% ( ).

).



The substitution =   9 +   :  leads to a separable equation:  =  9 [ ! + 'h ( )].



+ ) 


 =  ( ).
The substitution = +   +  leads to a separable equation: u

! (  ^ ]

The substitution a =    9  leads to a linear equation with respect to


B + s (a ).
12.

! (  ^ ]

[ ! (

[ ! (

 ^ ]m" (

 ^ " (

)+

)+

)+

 ]

a (a ) g =

The substitution a =    9  leads to a Riccati equation:


13.

= (a ):

)]

= % (

)]

= % (

a (a ) g = B

) 

^ ]n" (

) 

^ " (

).

).

+ s (a ).

The transformation a =  +  , H =  9  leads to a linear equation with respect to H = H (a ):


[   (a ) + % (a )] H g  = s% (a ) H + h (a ).

14.

The transformation a =  +  , H =  9


[   (a ) + % (a )] H g  = s (a ) H s (a ).

15.

[

^ ! (

Solution: X
16.

17.

[ ! (  ^ ]

_ ]" (

+

 :   ( )
M

)+




leads to a linear equation with respect to H = H (a ):

= 0.

+ X  9 ( )    9

) + " (  ^ ] )]

= % (  ^ ]

 :
 =L .

) " (

 ^ ]

).

The substitution a =    9  leads to a linear equation with respect to


a [B (a ) + s% (a )] g  = B (a ) + % (a ).
[ ! (

)+

" ( ^

)] = [ % ( 

)

" ( ^

)].

The substitution a =  9 + leads to a linear equation with respect to 
a [ s (a ) + % (a )] g = (a ) +  (a ).

(a ):

=  (a ):

2003 by Chapman & Hall/CRC

Page 198

1.7. EQUATIONS OF THE FORM

18.
19.
20.

21.

22.
23.
24.
25.

( , )

= ( , ) CONTAINING ARBITRARY FUNCTIONS

= ! ( ) sinh( ) + " ( ) cosh( ) + % ( ).

= ! ( ) sinh2 ( ) + " ( ) cosh2 ( ) + % ( ) sinh(2 ) + s( ).

= ! ( ) sinh 

+ " ( )  .

= ! ( ) cosh 

+ " ( )  .

! (

sinh ).

The substitution =   leads to a Riccati equation: 2  = (  + )


The substitution = tanh(
The substitution =
  ( ).
The substitution =
  ( ).
coth

! (

+ 2 "% + ( ).

 = " ( ) cosh +
+ ( ) leads to an equation of the form 1.7.2.18: I

tanh B! ( sinh ).

tanh

 = " ( ) sinh +
+ ( ) leads to an equation of the form 1.7.2.18: I

The transformation a =   , H = sinh

+ 2 "% + (  ).

 = (  + )
) leads to a Riccati equation: p

The transformation a = sinh  , H = +

199

cosh ).

leads to an equation of the form 1.7.1.3: aH g  =

H! (aH ).

leads to an equation of the form 1.7.1.3: B/aH g  = H! (aH ).

The substitution a = cosh  leads to an equation of the form 1.7.1.3: a g =  (a + ).

coth B! ( cosh ).

leads to an equation of the form 1.7.1.3: "H   = H! (  H ).

The substitution H = cosh

1.7.3. Equations Containing Logarithmic Functions


1.

= ! ( ) ln2 + " ( ) ln + % ( ) .

The substitution = ln
2.

+1

! (

leads to a Riccati equation:  =  ( )

ln ).

The substitution a = ln  leads to an equation of the form 1.7.1.3:


3.

1  ! (

ln ).

The substitution H = ln
4.

1  ]l! ( ]

ln ).

6.

 ! (

+ ( ) + % ( ).


g =
a

[a +  (a + )].

H  (  H )
leads to an equation of the form 1.7.1.3: H   =
.
   H

The substitution a = ln  leads to an equation of the form 1.7.2.4:


5.

 1
g = [a    (a   )].
a

ln ).


 ( H )

.
The substitution H = ln leads to an equation of the form 1.7.2.3: H  = H

 H

ln + ! ( )" (w ln ).

The substitution ( ) =   ln

 =    ( ) ( ).
leads to a separable equation: u

2003 by Chapman & Hall/CRC

Page 199

200

FIRST-ORDER DIFFERENTIAL EQUATIONS

7.

! (
ln

The transformation a =   + , H = ln
2
a (a ) H g  = BH +  (a ).

8.

! ( ) .
(ln )2

The transformation a =   + , H = ln

9.

leads to a linear equation with respect to H = H (a ):

leads to a Riccati equation:

[ ! (

a (a ) H g  = BH 2 +  (a ).

) + ln o" ( )] = [ % ( ) ln o" ( )].


The transformation a =   + , H = ln leads to a linear equation with respect to H = H (a ):
a [B (a ) + % (a )]H g  = B (a ) H + % (a ).

10.

[ ! (

) + ln " ( )] = [ % ( ) ln " ( )].


The transformation a =   + , H = ln  leads to a linear equation with respect to H = H (a ):
a [B (a ) + % (a )]H g  = (a ) H +  (a ).

1.7.4. Equations Containing Trigonometric Functions


1.
2.
3.
4.
5.
6.
7.
8.

9.

10.

= +1 sin H ( cos ).
This is an equation of the type 1.7.4.3 with  ( ) =  ( ).

= +1 cos H ( sin ).
This is an equation of the type 1.7.4.4 with  ( ) =  ( ).

= tan ! ( cos ).
The substitution a = cos  leads to an equation of the form 1.7.1.3: a g =  (a + ).

= cot ! ( sin ).
The substitution a = sin  leads to an equation of the form 1.7.1.3: a g =  (a + ).

= 1 tan B! ( sin ).
The transformation a =   , H = sin

= 1 cot B! ( cos ).
The transformation a =   , H = cos

= 1 sin 2B! ( tan ).


The transformation a =   , H = tan

= 1 sin 2B! ( cot ).


The transformation a =   , H = cot

leads to an equation of the form 1.7.1.3: B/aH g  = H! (aH ).


leads to an equation of the form 1.7.1.3: B/aH g  = H! (aH ).
leads to an equation of the form 1.7.1.3: B/aH g  = 2 H! (aH ).
leads to an equation of the form 1.7.1.3: B/aH g  = 2 H! (aH ).

! ( tan ).
sin 2
The substitution a = tan  leads to an equation of the form 1.7.1.3: 2a g =  (a + ).

! ( cot ).
sin 2
The substitution a = cot  leads to an equation of the form 1.7.1.3: 2a g =  (a + ).

2003 by Chapman & Hall/CRC

Page 200

1.7. EQUATIONS OF THE FORM

11.
12.

( , )

= ( , ) CONTAINING ARBITRARY FUNCTIONS

= ! ( ) cos( ) + " ( ) sin( ) + % ( ).


The substitution = tan( 
2) leads to a Riccati equation: 2  =  ( %  ) 2 +2  ! +  (  + % ).

= ! ( ) cos2 ( ) + " ( ) sin2 ( ) + % ( ) sin(2 ) + s( ).


The substitution = tan( ) leads to a Riccati equation:  =  ( + )

+ 2  % +  (  + ).

13.

= ! ( + tan ) tan2 .
The substitution = +  tan  leads to a separable equation:  =  +  ( ).

14.

sin 2
! (tan tan ).
sin 2
The transformation a = tan  , H = tan

15.

16.

17.

18.
19.
20.

21.

22.

23.

201

= cot tan B! (sin sin ).


The transformation a = sin  , H = sin

! (

leads to an equation of the form 1.7.1.3: aH g  = H! (aH ).

" (sin sin ).


cos
The substitution ( ) = sin  sin leads to a separable equation:  = sin 3 ( ) ( ).

= cot tan +

leads to an equation of the form 1.7.1.3: aH g  = H! (aH ).

sin 2
+ cos2 B! ( )" (tan tan ).
sin 2
 = tan 3 ( ) ( ).
The substitution ( ) = tan  tan leads to a separable equation: p

= 1 sin 2 + cos2 B! ( )" ( 2 tan ).


The substitution ( ) =  2  tan leads to a separable equation:  =  2   ( ) ( ).

(1 + tan2 ) = ! ( ) tan +1 + " ( ) tan + % ( ) tan1 .


leads to a Riccati equation:  =  ( )
The substitution = tan +

( ) +

% ( ).

= ! ( ) sin  + " ( )  .
 = " ( ) sin +
The substitution = + ( ) leads to an equation of the form 1.7.4.11: I
  ( ).

= ! ( ) cos  + " ( )  .
 = " ( ) cos +
The substitution = + ( ) leads to an equation of the form 1.7.4.11: I
  ( ).

= ! ( ) sin2  + " ( )  .
 = " ( ) sin2 +
The substitution = + ( ) leads to an equation of the form 1.7.4.12: I
  ( ).

= ! ( ) cos2 p + " ( )  .
 = " ( ) cos2 +
The substitution = + ( ) leads to an equation of the form 1.7.4.12: I
  ( ).

1.7.5. Equations Containing Combinations of Exponential,


Logarithmic, and Trigonometric Functions
1.

= sin 2 + cos2 B! ( )" (  2 tan ).



 =  2   ( ) ( ).
The substitution ( ) =  2 tan leads to a separable equation: u

2003 by Chapman & Hall/CRC

Page 201

202
2.

FIRST-ORDER DIFFERENTIAL EQUATIONS

cos )
.
sin

H (


This is an equation of the type 1.7.5.5 with  ( ) =  ( )


.
3.

 ]

cos

H ( ]

sin ).

This is an equation of the type 1.7.5.7 with  ( ) =  ( ).


4.

= tan B! ( 

sin ).


leads to an equation of the form 1.7.2.3: H   = H! (  H ).

The substitution H = sin


5.

= cot B! ( 

cos ).


leads to an equation of the form 1.7.2.3: H   = H! (  H ).

The substitution H = cos


6.

7.

8.

9.

10.

= tan

! (  ]

cos ).

= cot

! ( ]

sin ).

= sin 2B! (

= sin 2B! (

The substitution a = cos  leads to an equation of the form 1.7.2.4: a g =  (a  ).


The substitution a = sin  leads to an equation of the form 1.7.2.4: a g =  (a  ).

tan ).

The substitution H = tan  leads to an equation of the form 1.7.2.3: H   = 2 H! (  H ).

cot ).


The substitution H = cot  leads to an equation of the form 1.7.2.3: H   = 2 H! (  H ).

H (


sin )
.
cos

This is an equation of the type 1.7.5.4 with  ( ) =  ( )


.
11.

]

sin

H (  ]

cos ).

This is an equation of the type 1.7.5.6 with  ( ) =  ( ).


12.

tan )
.
sin 2
The substitution a = tan  leads to an equation of the form 1.7.2.4: 2a g =  (a  ).

! ( ]

13.

cot )
.
sin 2
The substitution a = cot  leads to an equation of the form 1.7.2.4: 2a g =  (a  ).

14.

! ( ]

  !

+ ln ).

The substitution =  + ln
15.

 ] ! (

+ ln ).

The substitution =

leads to a separable equation:  = 

+ ln  leads to a separable equation:   = 

 ( ) + .

 ( ) + 1.

2003 by Chapman & Hall/CRC

Page 202

1.8. EQUATIONS OF THE FORM

( , ,

) = 0 CONTAINING ARBITRARY FUNCTIONS

203

1.8. Equations of the Form c ( , , ) = 0


Containing Arbitrary Functions
1.8.1. Some Equations
1.8.1-1. Arguments of arbitrary functions depend on  and .
1.

( )2 + [ ! ( ) + " ( )] + ! ( )" ( ) = 0.

The equation can be factorized: [ +  ( )][  + ( )] = 0, i.e., it falls into two simpler
equations   +  ( ) = 0 and   + ( ) = 0. Therefore, the solutions are:

 ( )  = L
M

+X
2.

( )2 + 2 ! + "I

= ("

) exp O 2 X

3.

= ! (



 2  +L

M
P
 M P
L exp O X  
M P


  cosh O X
 2  +L
exp O X


P
M P
M
exp O X



!

Here,  =  ( ), = ( ). Solution:



+ X ( )  = L .

and

+ 2 ) (

sin O X

if >  2 ,
if  2 ,
if <  2 .

)2 + 1.

Raising the equation to the second power and applying the transformation  = & (a ) cos a ,
= & (a ) sin a , one arrives at the relation & 4 =  2 (& 2 )[(& g  )2 + & 2 ]. Solving it for & g  yields a
separable equation:  (& 2 )& g  = A"& & 2  2 (& 2 ).
4.

= ! (

+ ) (

)2 + 1.

Raising the equation to the second power and applying the transformation  = & (a ) cos a ,
= & (a ) sin a , one arrives at the relation & 2 (& g  )2 =  2 (& 2 )[(& g  )2 + & 2 ]. Solving it for & g  yields a
& (& 2 )
.
separable equation: & g  = A
& 2  2 (& 2 )

^_
5.

Reference: G. W. Bluman, J. D. Cole (1974, p. 100).

!e

)2 + 1.

Raising the equation to the second power and applying the transformation  = & (a ) cos a ,
= & (a ) sin a , one arrives at the relation & 2 =  2 (cos a , sin a )[(& g  )2 + & 2 ]. Solving it for & g  yields
a separable equation:  (cos a , sin a )& g  = A"& 1  2 (cos a , sin a ).
6.

+ =

!e

)2 + 1.

Raising the equation to the second power and applying the transformation  = & (a ) cos a ,
= & (a ) sin a , one arrives at the relation & 2 (& g  )2 =  2 (cos a , sin a )[(& g  )2 + & 2 ]. Solving it for & g 
yields a separable equation:
1  2 (cos a , sin a ) & g  = A"& (cos a , sin a ).

2003 by Chapman & Hall/CRC

Page 203

204

FIRST-ORDER DIFFERENTIAL EQUATIONS

1.8.1-2. Argument of arbitrary functions is " .


7.

= ! ( ).
Solution in parametric form:

 =  (a ),
8.

9.

= ! ( ).
Solution in parametric form:

! ( ) +

+ + s = 0.
Solution in parametric form:

= X a  g  (a ) a + L .

a
 = Xv g  (a ) M + L ,
a

=  (a ).

 g  (a ) a
M ,  =   (a ).
 + 'a
In addition, there is a particular solution =  + , where and are determined by solving
 =L X

the system of two algebraic equations:


 +  = 0,

10.

11.

12.

13.

14.

15.

16.

 ( ) + ' + = 0.

+ ! ( ).
= L3 +  ( L ).
The Clairaut equation. Solution:
In addition, there is a singular solution, which may be written in the parametric form as:

 =  g  (a ),
= a g (a ) +  (a ).

= ! ( ) + " ( ).
The LagrangedAlembert equation. For the case  (a ) = a , see equation 1.8.1.10. Having
differentiated with respect to  , we arrive at a linear equation with respect to  =  (a ), where
a =  : [a  (a )] g =  g  (a ) + g  (a ). See also 1.8.1.12.

! (

) + " ( ) + % ( ) = 0.
The Legendre transformation m
[  (m ) + m (m )] n o  (m ) n + %
Inverse transformation:  = n

=  , n =   , n o  = 
(m ) = 0.



o , = mn o n ,  = m .

leads to a linear equation:

= 2 ! ( ) + " ( ) + % ( ).
Having differentiated with respect to  , we arrive at an Abel equation with respect to  =  (a ),
where a =  :
V 2  (a ) + (a ) aW g =  g  (a ) 2 g  (a ) % g (a )
(see Subsection 1.3.4).

= 2 ! ( ) + " ( ) + % ( ).
Having differentiated with respect to  , we arrive at an Abel equation with respect to = (a ),
where a =  :
V 2a (a ) + a (a ) 1W g  = a g  (a ) 2 a g  (a ) a% g (a )
(see Subsection 1.3.4).

= ! ( ) + .
Differentiating with respect to  and denoting a =   , we obtain a Bernoulli equation for
 =  (a ): B (a ) g  g  (a )  2  = 0.

( ) ! ( ) + " ( ) + % ( ) = 0.
The Legendre transformation  = g , = a g
[a (a ) + % (a )] g = (a ) +  (a )  .

(  = a ) leads to a Bernoulli equation:

2003 by Chapman & Hall/CRC

Page 204

1.8. EQUATIONS OF THE FORM

( , ,

) = 0 CONTAINING ARBITRARY FUNCTIONS

205

1.8.1-3. Arguments of arbitrary functions are linear with respect to  .


17.

= (

)2 + ! ( 2

).

1
=  (L ) +
( L )2 . In addition, there is a singular solution, which can be
Solution:
4
represented in parametric form as:

 = a + 2   g  (a ),
18.

= ! ( +

The contact transformation


leads to a linear equation: n
Inverse transformation:
19.

= ! ( +

)( 2 + 2

 ^ ]

= ! (

).

The substitution

22.

23.

24.

25.

m


).

=  +  , n = 12 (  )2 +  , n o  =  , where n = n (m ),


= 2 (m ) n +  (m ).


=  1 (m n o ), = 12  1 [2 n ( n o )2 ],  = n o  .

) + " ( +

)( 2 + 2 )

The contact transformation m =   +  , n = 12 (   )2 +  , n o  =   , where n = n (m ),


leads to a Bernoulli equation: n o  = 2  (m ) n + 2  (m ) n  .
Inverse transformation:  =  1 (m n o  ), = 12  1 [2 n ( n o  )2 ],  = n o  .

20.

21.

)( 2 + 2

) + " ( +

=  (a ) +  [  g  (a )]2 .

= ! ( )" (


 ).
leads to an equation of the form 1.8.1.32:   9 =  ( 2

= ln

).

The Legendre transformation m =  , n =   , n o  =  leads to a separable equation:


n o  =  (m ) ( n ).
Inverse transformation:  = n o  , = mn o  n ,  = m .

! (

) +

" (

= ! (

+ )" (

= ! (

+ )+

)( )  = .

The modified Legendre transformation m =   , n =  , n o  = 1


 leads to a Bernoulli
equation: !n o  =  (m ) n + (m )n  .
Inverse transformation:  = ( n o  )1 , = n ( n o  )1 m ,  = n .

).

The contact transformation m =   + , n =  2  , n o  =  , where n = n (m ), leads to a


separable equation: n o  =  (m ) ( n ).
Inverse transformation:  = n o  , = m n ( n o  )1 ,   = n ( n o  )2 .

" (

+ ).

The contact transformation m =   + , n =  2  , n o  =  , where n = n (m ), leads to a


linear equation: n o  = (m ) n +  (m ).
Inverse transformation:  = n o  , = m n ( n o  )1 ,   = n ( n o  )2 .

 ! (

+ )+

3 ( )2" (

+ )= .

The contact transformation m =   + , n =  2  , n o  =  , where n = n (m ), leads to a


Bernoulli equation: !n o  =  (m ) n + (m )n 2 .
Inverse transformation:  = n o  , = m n ( n o  )1 ,   = n ( n o  )2 .

2003 by Chapman & Hall/CRC

Page 205

206
26.

FIRST-ORDER DIFFERENTIAL EQUATIONS


+ ) = 2 ( 2 + 1).
Setting ( ) = !   +  and differentiating with respect to  , we obtain

! (

 [    ( ) 2 + 2 ] = 0.

(1)

= ( L ) + $ . Substituting
Equating the first factor to zero, after integrating we find
the latter into the original equation yields $ =  ( L ). As a result we obtain the solution:
2
=  ( L ) (  L )2 .
There is also a singular solution that corresponds to equating the second factor of (1) to
zero. This solution in parametric form is written as:
2

27.

28.

 = 12    ( ),

=  ( ) 14 [    ( )]2.

= ! ( ) + 2 ( 2 1)" ( ).
The contact transformation  = nIn o  m , = n [( n o  )2 1]1 J 2 ,  = n o  [( n o  )2 1]1 J 2 ,

= n 2.
where n = n (m ), leads to the equation nCn o  =  (m ) + n 2 (m ), which is linear in





!
2
1
J
2
2
1
Inverse transformation: m =   , n = [(  ) 1] , n o =  [(  ) 1] J 2 .

1
! Oi
2I

+ Oi

" Oi
P3

The contact transformation m =  + 


 , n = 
equation: mn o  = 2 (m ) n + 2mR (m ).
Inverse transformation:
m n o

1
 =A
mn o  n ,
=A
m
2 mn
29.

D 1 I
! Oi

+ Oi

"3Oi
P

For  1, the contact transformation m =


to a linear equation: n o  = (m ) n +  (m
Inverse transformation:
1
1
 = ( n o  ) 1 ,
=
(mn
1
30.

D +1 I
! Oi

= " (

D +1

(  )2

2n
,

o n

, n o  = 2 2  leads to a linear

 =

= .

  
 , n = 
).



o n )( n o ) 1 ,

m 2n o 
.
2(mn o  n )

  , n o  = 
 =

mn o  !n
.
(1  ) n o 

).

For  1, the contact transformation m =  +  


 , n =  +1   , n o  = 
to a separable equation:  (m ) n o  = ( n ).
Inverse transformation:

1
1
mn o  + !n
 =
(mn o  n )( n o  ) +1 ,
 = ( n o  ) +1 ,
=
.
 +1
(  + 1) n o 
31.
32.

leads

 ^

= ! (
 ).
The substitution  = ln a leads to an equation of the form 1.8.1.32: a9

+1

leads

 =  (a g
 ).

= ! (  ).
We pass to a new variable ( ) =   
 , divide both sides of the equation by   + , and
differentiate with respect to  . As a result we arrive at a separable equation:   ?  ( )2 =
( + B )  ( ).
Solution in parametric form:
 ?  ( )
+ L ,   + =  ( ).
ln | | = X
M
( + B )  ( )

In addition, there are singular solutions


equation # +  (B
) = 0.

 J + , where #  are roots of the algebraic


= #  

2003 by Chapman & Hall/CRC

Page 206

( , ,

1.8. EQUATIONS OF THE FORM

33.

34.

35.

36.

=  ! ( ) .

The contact transformation m =   , n =  + , n
separable equation: nCn o  =  (m ).
m n
Inverse transformation:  = ln n o  , = n 

 ! (

) " ( + ) = 0.

The contact transformation m =  
separable equation: ( n ) n o  =  (m
Inverse transformation:  = ln n

! (

) + " (  )( +
The contact transformation
linear equation: !n o  = (m
Inverse transformation:



o =  , where n = n (m ), leads to a
 

o ,  = mn o .


=  + , n o  =  , where n


o ,

) " ( + ) = .
This equation can be rewritten in the form 1.8.1.34:



o =  , where n = n (m ), leads to a
 

o ,  = mn o .

! ( 

where

 ( ) = ln  ( ),

sin cos ) = 
! ( cos + sin ).
The contact transformation

  cos  + sin 
(  )2 + 2 ,
2
leads to the homogeneous equation: n o  =  ( n
m ).

H (

+1

1
= 
(  )2 +

H ( 2

n o  =  (   sin  cos  )


n =

Solution:
 ( L 1 B , L

39.

( ) = ln 4 ( ).

38.

= n (m ), leads to a

= n mn o  ,  = mn o  .

) =  .

m =   , n =  + , n
) n +  (m ).
 = ln n o  , = n 
m n



  (    ) 4 (   + ) = 0,

37.

, n
).

207

) = 0 CONTAINING ARBITRARY FUNCTIONS

+ ) = 0.
= L 1   + L 2 . Here, the constants L
2 B ) = 0.

and L

are related by the constraint

+ 2 , 3 + 2 ) = 0.
= L 1  1 + L 2  2 . Here, the constants L 1 and L 2 are related by the constraint
Solution:
 ( L 1 , L 2 ) = 0.
The singular solution can be represented in parametric form as:

  ( , ) + " ( , ) = 0,

 ( , ) = 0,

=  2 a + 2 ,

where

=  3a + 

The subscripts and denote the respective partial derivatives, and a is the parameter.
40.

+ , +1 + ) = 0.
= L 1   + L 2  + . Here, the constants L
Solution:
  L 1 ( B ), L 2 (B )  = 0.

H (

+1

41.

42.

and L

are related by the constraint

H (  ,

+ ) = 0.

= L 1 + L
Solution:
 ( L 1 , L 2 ) = 0.

H (  ^

2.

Here, the constants L

and L

are related by the constraint

+ J ^ ,  _ + q _ ) = 0.


Solution: = L 1  9 + L 2  : . Here, the constants L
  L 1 ( ), L 2 ( )  = 0.

and L

are related by the constraint

2003 by Chapman & Hall/CRC

Page 207

208
43.

FIRST-ORDER DIFFERENTIAL EQUATIONS

H (

cosh sinh ,

= L 1 sinh  + L
Solution:
constraint  ( L 1 , L 2 ) = 0.
44.

H ( ,

45.

HwO

, ln

Solution:
 ( L 2 , ln L
46.

HwO

1)

47.

H (

ln  + L

r P

= L
Solution:
 ( L 1 , L 2 ) = 0.
49.

are related by the

and L

are related by the constraint

 ). Here, the constants L

and L

are related by the constraint

ln

= 0.

and L

cos  . Here, the constants L

= 0,

r (

 ( L 1 , L 2 ) = 0.

and L

are related by the constraint

).

Y ( ) + L 2 . Here, the constants L

are related by the constraint

sin cos ) = 0.

r  r
s r , r

r P
s = L Y ( s ) + L s j ( ).
s
Solution:
1
2
HwOr s

and L

k 2 . Here, the constants L

= L 1
1 ) = 0.

cos + sin ,

HwO r

Here, the constants L

2.

Solution: = L 1 sin  + L
 ( L 1 , L 2 ) = 0.
48.

cosh  . Here, the constants L

= 0.

= L 1 exp( L
= 0.

, ln

Solution:
 ( L 2 , ln L

ln ) = 0.

= L
Solution:
 ( L 1 , L 2 ) = 0.

sinh cosh ) = 0.

= 0,

and L

r (

),

Here, the constants L

are related by the constraint

and L

s
2

( ).
are related by the constraint

The singular solution can be represented in parametric form as:

 ( , ) = 0,

j  ( , ) + Ys ( , ) = 0,

where =

j
Y j
h



aj
,
  jY 

Y
Y j
h



aY
.
  jY 

The subscripts and denote the respective partial derivatives, and a is the parameter.
50.

H r

r , r
]

(r

r ) 
]
Lt , Y  = Lt

Here, Y = Y ( , ), Y  =

= 0.

(Y  + Y

. Differentiating with respect to  , we obtain

  )  ( 

"

) = 0,

where   =  and  = are partial derivatives of function  ( , ). Equating the first



 solution:
factor to zero, we find the

Y ( , ) = L3 + # ,

where

It remains to be checked whether the equation 


which of them satisfy the original equation.

 ( L , # ) = 0.

"

= 0 possesses any solutions and

2003 by Chapman & Hall/CRC

Page 208

1.8. EQUATIONS OF THE FORM

( , ,

209

) = 0 CONTAINING ARBITRARY FUNCTIONS

1.8.1-4. Arguments of arbitrary functions are nonlinear with respect to  .


51.
52.

= ! ( 2) + 2 .
Solution: [  ( L )]2 = 4 L3 .

= 2 ( )3 + ! ( 3 2 ).
This is a special case of equation 1.8.1.72 with B = 3.
 L 3J 2
. In addition, there is the following singular solution
Solution: =  ( L ) + 2 rO
3 P
written in parametric form:

 = a + 3  [  g  (a )]2 ,

53.

54.

55.

56.

=  (a ) + 2 [  g (a )]3 .

= ! ( 2 ) + (2 3 3 )" ( 2 ),
0.


2
The contact transformation m =  (  ) ' , n = 2  (  )3 3  , n o  = 3  leads to a linear
equation: n o  = 3 (m ) n + 3  (m ).
1 1
 [2  ( n o  )3 27 n ],  = 13 n o  .
Inverse transformation:  = 19  1 [  ( n o  )2 9m ], = 81

= ! ( +

)"

( 21 2 + ).

The contact transformation m =  +  ,


leads to a separable equation: n o  =  (m )
Inverse transformation:  =  1 (m n

) = ! ( 2 2 ).

The contact transformation m = ( )2


(separable) equation: 2  (m ) n o  = n .

= ! ( 2 2 )(  + 
The contact transformation

n = 12 (  )2 +  , n o  =  , where n = n (m ),
( n ).




o ), = 12  1 [2 n ( n o )2 ],  = n o .



, n =  (  ), n o  = 12  (   )1 leads to a linear

).





n =   (   +   ) (     ),
leads to a separable equation: 2  (m ) n o  = 1.
m

57.

58.

59.

60.

= (   )2

=  ! ( 2 2 )" (   ).
The contact transformation m = ( )2
separable equation: 2  (m ) ( n ) n o  = 1.

, n



n o  = 12 (   +   )(   )1

=  (  ), n o  =

1
2


 (  )1 leads to a

= ! ( 2 2 ) cosh .
The contact transformation m = (   )2 2 , n =   cosh  sinh  , n o  =
leads to a separable equation: 2  (m ) n o  = 1.

= ! ( 2

1
2

cosh  (   )1

1
2

sinh  (   )1

) sinh .

The contact transformation m = ( )2 2 , n =  sinh  cosh  , n o  =


leads to a separable equation: 2  (m ) n o  = 1.

= ! ( 2 2 )( cosh + sinh ).
The contact transformation

m = (   )2 2 ,

n =   (  cosh  +  sinh  ) (  sinh  +  cosh  ),

leads to a separable equation: 2  (m ) n o  = 1.

 cosh  +  sinh 
n o =
2 

2003 by Chapman & Hall/CRC

Page 209

210
61.

62.

63.

64.
65.

FIRST-ORDER DIFFERENTIAL EQUATIONS


= ! ( 2 2 )" ( cosh sinh ) cosh .
The contact transformation m = ( )2 2 , n =  cosh  sinh  , n o  =
leads to a separable equation: 2  (m ) ( n ) n o  = 1.

= ! ( 2 2 )" ( sinh cosh ) sinh .


The contact transformation m = ( )2 2 , n =  sinh  cosh  , n o  =
leads to a separable equation: 2  (m ) ( n ) n o  = 1.

) + 2 cosh sinh = cosh .


The contact transformation m = ( )2 2 , n =  cosh  sinh  , n o  =
leads to a linear equation: 2 n o  = !
 n +  (m ).

 ! ( 2

68.

1
2

cosh  (  )1

= ! ( 2 + 2 )( cos + sin ).
The contact transformation

n o  = 12 (  cos  +  sin  )(   )1

= ! ( 2 + 2 )" ( cos + sin ) cos .


The contact transformation m = ( )2 + 2 , n =  cos  + sin  , n o  =
to a separable equation: 2  (m ) (n ) n o  = 1.

) + 2 cos + sin = cos .


The contact transformation m = ( )2 + 2 , n =  cos  + sin  , n o  =
to a linear equation: 2 n o  = !n +  (m ).

 ! ( 2 +

= ! ( 2 + 2 )" ( sin cos ) sin .


The contact transformation m = ( )2 + 2 , n =  sin  cos  , n o  =
to a separable equation: 2  (m ) (n ) n o  = 1.

1
2

cos  (  )1 leads

1
2

cos  (  )1 leads

1
2

sin  (  )1 leads

P " ( 2 2

=I
! Oi +

).

The contact transformation m =  + 


 , n =  2 (  )2
equation: n o  = 2  (m ) ( n ).
Inverse transformation:

 =A

71.

sinh  (  )1

= ! ( 2 + 2 ) cos .
This is a special case of equation 1.8.1.65 with  = 1 and  = 0.

69.

70.

1
2

n =   (  cos  +  sin  )+ (  sin   cos  ),


leads to a separable equation: 2  (m ) n o  = 1.

67.

cosh  (  )1

m = (   )2 + 2 ,

66.

1
2

mn o  n ,

=A

mn o  2 n
,
2 mn o  n

, n o  = 2 2  leads to a separable

 =

m 2n o 
.
2(mn o  n )

= ! ( 2 )" (2 3 3 ),
0.

2
The contact transformation m =  (  ) ' , n = 2  (  )3 3  , n o  = 3  leads to a separable
equation: n o  = 3  (m ) ( n ).
1 1
 [2  ( n o  )3 27 n ],  = 13 n o  .
Inverse transformation:  = 19  1 [  ( n o  )2 9m ], = 81

= ! (

Solution:

)+

 1
BL
=  (L  ) +
  .
B 1

2003 by Chapman & Hall/CRC

Page 210

1.8. EQUATIONS OF THE FORM

72.

( , ,

) = 0 CONTAINING ARBITRARY FUNCTIONS

211

= ( 1)( ) + !  ( ) 1  .
Differentiating with respect to  , we obtain a factorized equation:

V 1 B (B 1)(   ) 

 B (  )
where a =  

    W V    g  (a )W = 0,

(1)

. Equate the first factor to zero and integrate the obtained equation.
Substituting the expression obtained into the original equation, we find the solution:

 L

B P

=  ( L ) +  (B 1) O

Equating the second factor in (1) to zero, we have another solution that can be written in
parametric form as:

 = a + B [  g  (a )] 

73.

=  (a ) +  (B 1)[  g  (a )]  .

= !  ( )   "  ( )  +1 ( + 1)  .
The contact transformation (   0, , 1)

=  (   )  ' ,

n =  , (   ) 

+1

n o  = ( , + 1)  

 ( , + 1) ,

leads to a separable equation: n o  = ( , + 1)  (m ) ( n ).


Inverse transformation:

 (n o  ) 
m
,

 ( , + 1)


 =
74.

 , ( n o  ) 
 ( , + 1) 

+1
+2

 ( , + 1)

= !u ( )   +  ( )  +1 ( + 1)  *
"  (
The contact transformation (   0, , 1)

)


=  (  )  ' ,

n =  , (   ) 

+1

 ( , + 1) ,

leads to a linear equation: n o  = ( , + 1) (m ) n + ( , + 1)  (m ).


Inverse transformation:

 (n o  ) 
m
,
 ( , + 1) 


 =
75.

76.

H 

+
Solution:

H ( 

78.

 , ( n o  ) 
 ( , + 1) 

+1
+2

 ( , + 1)

2 +

 = 0.
=  + 2 L 1  + L 2 . Here, the constants L

, 2


= L 1 + L
Solution:
 (2 L 2 , 4 L 1 L 2 ) = 0.

77.

and L

  L 1 , L 12 +  L 2  = 0 if
  L 1 , L 12 +  L 2  = 0 if

n o
.

 =

, +1

.

n o  = ( , + 1)  

n o
.

 =

, +1

are related by the constraint

> 0,
< 0.

) = 0.

. Here, the constants L
2

and L

cosh sinh , 2 2 ) = 0.
= L 1 sinh  + L 2 cosh  . Here, the constants L
Solution:
constraint  ( L 1 , L 12 L 22 ) = 0.

are related by the constraint

H (

cos + sin , 2 + 2 ) = 0.
Solution: = L 1 sin  + L 2 cos  . Here, the constants L
 ( L 1 , L 12 + L 22 ) = 0.

H (

and L

and L

are related by the constraint

are related by the

2003 by Chapman & Hall/CRC

Page 211

212

FIRST-ORDER DIFFERENTIAL EQUATIONS

1.8.2. Some Transformations


1.

2.

= ! ( , ).
Substituting a =   and differentiating both sides of the equation with respect to  , we obtain
an equation with respect to = (a ):

where g = g ,  = .
[1 a ( , a )] g  = ag ( , a ),




If = (a ) is the solution of the latter equation, the solution of the original equation can be
represented in parametric form as:
 =  ( (a ), a ),
= (a ).

= ! ( , ).
Differentiating with respect to  and setting a = " , we obtain an equation with respect to
 =  (a ):

where g = g ,   =  .
[a   ( , a )] g = g ( , a ),

If  =  (a ) is the solution of the latter equation, the solution of the original equation can be
represented in parametric form as:
 =  (a ),
=  ( (a ), a ).

3.

=I
! O 

Set H =  

and

P
 

.
. Divide both sides of the equation by   +

and differentiate

with respect to  . As a result we arrive at the following equation with respect to = ( H ):

H ( + , )(  +  ?  ) = ( + B )  , where  =  ( H , ),

which is usually simpler than the original equation, since it is readily solved for the derivative.
If = ( H ) is the solution of the equation obtained, the solution of the original equation is
written in parametric form as:
  = H ,   + =  ( H , ( H )).
4.
5.
6.

7.

8.

 ^ ]

= ! (
The substitution

 ^

 _ ] ,

).
= ln leads to an equation of the form 1.8.2.3:   9 =    + : ,  
 .

= ! (  _ ,  ).
The substitution  = ln a leads to an equation of the form 1.8.2.3: a 9

! ( ,

 =   a : + , a g 
  .

, ) = 0.
The Legendre transformation  = g , = a g (  = a ), where = (a ), leads to the
equation  ( g , , a ) = 0. Inverse transformation: a =   , =    , g =  .

( )2 = + ! ( ).
For 0, the transformation = 2 +  ( ) leads to an Abel equation of the second
kind,
*  = + Y ( ), where Y = 2 2    ( ),
which is outlined in Subsection 1.3.1 for specific functions Y .

= + 2 + ! ( ),
0.
Differentiating the equation with respect to  and changing to new variables a =   and
= 2  , we arrive at an Abel equation of the second kind,
* g  = + Y (a ), where Y = 2   g  (a ),
which is outlined in Subsection 1.3.1 for specific functions Y .

For information about contact transformations, see Subsection 0.1.8.

2003 by Chapman & Hall/CRC

Page 212

Chapter 2

Second-Order Differential Equations


2.1. Linear Equations
2.1.1. Representation of the General Solution Through a Particular
Solution
1 b . A homogeneous linear equation of the second order has the general form

 2 ( )    +  1 ( )   +  0 ( ) = 0.
Let 0 = 0 ( ) be a nontrivial particular solution (
solution of equation (1) can be found from the formula:
=

OL

+L

2
0

 ,
M P

(1)

1 0) of this equation. Then the general

where

 =X


1
2

 .

(2)

For specific equations described below in 2.1.22.1.9, often only particular solutions are given,
while the general solutions can be obtained with formula (2) (see also Paragraph 0.2.1-1, Item 3 b ).
x3y'z|{}~/
Only homogeneous equations are considered in Subsections 2.1.2 through 2.1.8; the
solutions of the corresponding nonhomogeneous equations can be obtained using relations (7) and (8)
of Subsection 0.2.1.
2 b . Suppose a particular solution of a homogeneous linear equation is obtained in the closed form
= [  ( )] , with this formula valid for  ( ) 0. If the equation makes sense in a range of  where
 ( ) < 0, then the function = |  ( )| will be a particular solution of the equation in that range.

3 b . Suppose = L 1  ( )[ ( )] + L 2  ( )[ ( )]  is the general solution of the homogeneous linear


equation with   , where  and  are free parameters. Then the function = L 1  ( )[ ( )] +
L 2  ( )[ ( )] ln ( ) will be the general solution of this equation with  =  .

2.1.2. Equations Containing Power Functions


2.1.2-1. Equations of the form   +  ( ) = 0.
1.

2.

+ = 0.
Equation of free oscillations.
  L 1 sinh( | | ) + L 2 cosh( | | ) if  < 0,
= L 1 + L 2
Solution:
if  = 0,
L 1 sin(  ) + L 2 cos(  )
if  > 0.

( + ) = 0,
0.
2 J 3
(  +  ) leads to the Airy equation:
The substitution = 

 

= 0,

(1)

213
2003 by Chapman & Hall/CRC

Page 213

214

SECOND-ORDER DIFFERENTIAL EQUATIONS


which often arises in various applications. The solution of equation (1) can be written as:
=L

Ai( ) + L

Bi( ),

where Ai( ) and Bi( ) are the Airy functions of the first and second kind, respectively.
The Airy functions admit the following integral representation:
1

Ai( ) =

X cos  13 a 3 + a  a ,
0
M

Bi( ) =

X exp  31 a 3 + a  + sin  13 a 3 + a  S a .
0
Q
M

The Airy functions can be expressed in terms of the Bessel functions and the modified Bessel
functions of order 1
3 by the relations:

V
Bi( ) = 13 V

Ai( ) =

1
3

H ) 1 J 3 ( H )W , Ai( ) = 13 V

1 V

1 J 3 ( H ) + 1 J 3 ( H )W , Bi( ) =
3
1 3 (

J H ) + 1 J 3 ( H )W ,

1 3 (

J H ) 1 J 3 ( H )W ,

1 3 (

where H = 23 3 J 2 .
For large values of , the leading terms of the asymptotic expansions of the Airy functions
are:
1 1 J 4
1 1 J 4

exp( H ), Ai( ) =
sin OH +  ,
Ai( ) =
4P
2


1 1 J 4
1 1 J 4

exp( H ),
Bi( ) =
cos OH +  .
Bi( ) =
4P

The Airy equation (1) is a special case of equation 2.1.2.7 with  = B = 1.


3.

+ ) = 0.

Particular solution:
4.

= exp 

1
2

 2  .

+ ) = 0.

The Weber equation (two canonical forms of the equation correspond to  = A

1
4 ).

1 b . The transformation H =  2  , =  J 2 leads to the degenerate hypergeometric equation


1 
1
+ 1 = 0.
2.1.2.70: H '  + O H  O
2
4
P
P

2 b . For  = ,

> 0,  = (2B + 1) , , where B = 1, 2,  , there is a solution of the form:


= exp  12 ,! 2  5   ,  ,

, > 0,

where 5  ( H ) = (1)  exp  H 2  M  exp  H 2  is the Hermite polynomial of order B .


H
See also Subsection S.2.10.M

^_
5.

References: H. Bateman and A. Erdelyi (1953, Vol. 2), M. Abramowitz and I. A. Stegun (1964).

(2

Particular solution:
6.

= 0.
0

= exp  12  2 

+   .

+ ) = 0.

The substitution =  +

2

leads to an equation of the form 2.1.2.4:

  O 2 + ( 

4 P

= 0.

2003 by Chapman & Hall/CRC

Page 214

215

2.1. LINEAR EQUATIONS

7.

= 0.
1 b . For B = 2, this is the Euler equation 2.1.2.123, while for B = 4, this is the equation
2.1.2.211 (in both cases the solution is expressed in terms of elementary function).

2 b . Assume 2
(B + 2) = 2



12



 (

12

(

+ 1, where

& c )+

+1

) + L

&c

L

exp

is an integer. Then the solution is:

exp

 &

 &

+L

+L

exp

exp

 &

 &

0,

if

if

< 0,

B +2

1
=
.
2
2 +1

M
3 b . For any B , the solution is expressed in terms of the Bessel functions and modified Bessel
functions of the first or second kind (see 2.1.2.126 and 2.1.2.127):

where c

where
8.

= M ,



L 1 

10.

1
2

 &

&

+L

2n

1
2

 &

&


 & + L 2  ; 1
 &
L 1  1


2
2
&
&



if  < 0,
if  > 0,

= 12 (B + 2).

Particular solution:
9.

1
0

) = 0.

= exp O

B +1

 

+1

2 ( + + 1) = 0.
Particular solution: 0 =  exp   
B  .

+ ( 2 + 1 ) = 0.
The substitution =   +1 leads to a linear equation of the form 2.1.2.108: (B + 1) 2  +
B (B + 1)  + (  +  ) = 0.

2.1.2-2. Equations of the form   +  ( )  + ( ) = 0.


11.

+ + = 0.
Second-order constant coefficient linear equation. In physics this equation is called an
equation of damped  vibrations.
V
1
1
1
2
2
 exp  2   L 1 exp  2   + L 2 exp  2   W if =  4  > 0,
Solution:

12.



exp  12   V L

  +L
1
    L 1 + L 2 
exp  2 
1

sin 

1
2

cos 

1
2

 W

+ + ( + ) = 0.
1 b . Solution with  0:

if

= 4 

if 

= 4 .

> 0,

= exp  12  
where

VL

 3 J 2  + L 2 n
2
 3J 2  W ,
J 
1J 3  3 "

2
1 1 3 3

= +

4( 
4

J H ) and n 1 J 3 ( H ) are the Bessel functions.

1 3(

2 b . For  = 0, see equation 2.1.2.11.

2003 by Chapman & Hall/CRC

Page 215

216
13.

14.

SECOND-ORDER DIFFERENTIAL EQUATIONS

+ (

+ ) = 0.

= exp  12  2   leads to a linear equation of the form 2.1.2.108:


The substitution
2   + (2 " +  )2 + (  " ( +  ) = 0.

Particular solution:
15.

Particular solution:
16.

17.

+ 2) = 0.

= exp  13 ' 3  .

= exp O

= exp O

B +1

= M  exp  21  2  L

Q

+ 2 = 0,

 

) = 0.

B +1

+ ( + 1) = 0,

Solution:
19.

+ (

Particular solution:
18.

= exp  12 ' 2  .

Particular solution:

+ 1) = 0.

+ (

+1

) = 0.

 

+1



= 1, 2, 3,
1

+L

= 1, 2, 3,

uuu

X exp  12  2   S .
M

uuu

Solution:
= exp( 2 ) M  exp( 2 ) L 1 + L 2 X exp( 2 )  S" .

Q
M
For L 1 = (1)  andM L 2 = 0, this solution defines the Hermite polynomials.
20.

= 0.

= L 1 `  12  1  , 12 , 21  2  + L 2 i  12  1  , 12 , 21  2  , where ` (  ,  ;  ) and


Solution:
i (  ,  ;  ) are the degenerate hypergeometric functions (see equation 2.1.2.70 and Subsection S.2.7).
21.

= 0.


=   J V L 1 `  12  3  2 , 12 , 12  2  + L 2 i  12  3  2 , 12 , 21  2  W , =  2  2  ,
Solution:
where ` (  ,  ;  ) and i (  ,  ;  ) are the degenerate hypergeometric functions (see equation

2.1.2.70 and Subsection S.2.7).

22.
23.
24.

+ + ( + ) = 0.
This is a special case of equation 2.1.2.108 with 

26.

=

= 0 and 

+ 2 + ( 4 + 2 2 + + ) = 0.
This is a special case of equation 2.1.2.49 with B = 1 and

+ (

+ )

Particular solution:

25.

+ = 0.
0

= exp  12 

+ ( + ) = 0.
Particular solution: 0 =  +  .

= 1.

= 2.

'  .

+ ( + ) + ( + ) = 0.

Particular solution: 0 =  v .

2003 by Chapman & Hall/CRC

Page 216

217

2.1. LINEAR EQUATIONS

27.

Particular solution:
28.
29.

31.

32.

34.
35.

37.

= 

2 )

= 0.

+ ) + [( )

+ (

= exp 

(  .
2

12

= 1.

+ 2( + ) + ( 2 2 + 2 + ) = 0.
The substitution = exp  12  2 + '  leads to a constant coefficient linear equation of the
form 2.1.2.1:   + ( (   2 ) = 0.

+ ( + ) + (  2 +  + ) = 0.
The substitution = exp( 2 ), where is a root of the quadratic equation 4 2 + 2 ! + = 0,
leads to an equation of the form 2.1.2.108:   + [(  + 4 ) +  ]  + [( + 2  ) + = + 2 ] = 0.

+ (

+ ) + (
0

= exp O

+1

B +1

 

+1

+ ( 2 2 ) ( + ) = 0.
Particular solution: 0 =   .

) = 0.

+ ( 2 + ) + (
Particular solution: 0 = 

+ (

+ 2 )

+ ) = 0.
.

= 

+ (
0

v

2 )

= 0.

+ (2 2 + ) + ( 4 + 2 + 2 + ) = 0.
The substitution = exp  13  3  leads to a constant coefficient linear equation of the form
2.1.2.11:    +   + ' = 0.

+ ( 2 + ) + (  2 +  + ) = 0.
1 b . This is a special case of equation 2.1.2.146 with B = 1.

38.

+ ( 2 + + 2 ) + 2 ( 2 + 1) = 0.

Particular solution: 0 = (  + 1)  .

39.

=  exp  31 

12 '  2  .

+ (

+ ) + (

Particular solution:

+ (

+ (

13

= exp 

+ ) + (

Particular solution:
41.

+ 1] = 0.

2 b . Let = 0, = 3  , = = 2  . Particular solution:

40.

= 0 and 

Particular solution:
36.

+ ( + ) + ( + ) = 0.
This is a special case of equation 2.1.2.108 with 

Particular solution:
33.

Particular solution:
30.

+ 2 ) + (

+ (

= exp 

= 

+ 2 ) + (

Particular solution:

12

+
3


3

'

+ 2 ) = 0.

(  .

+ ) = 0.

(   .

2 )

= 0.

2003 by Chapman & Hall/CRC

Page 217

218
42.

SECOND-ORDER DIFFERENTIAL EQUATIONS

+ (

Particular solution:
43.

+ (

+ 2 )

+ ) = 0.

12 '  2  .

+ 1) = 0.

=  exp  14 

+ 2(


.
0 = (  + 1) 

Particular solution:
44.

+ 2(2

= 0.

This equation is encountered in the theory of diffusion boundary layer.


=L

Solution:
45.

46.

+ 2

+L

48.

+ (

+ (

+ 2

+
2

+ (

+ (

+ 2 )

+ (

+ (

+ 2

) = 0.


 
B +1

+1

) = 0.

+ (

 

 +

+1

) = 0.

leads to a linear equation of the form 2.1.2.10:

+ (

^ ]

= +( .
= .

2 )

= 0.


= (  + 2)  .

) = 0.

) = 0.

+ 2 ) + 2 (

+1


.
0 = (  + 1) 

Particular solution:

+1

B +1

= 

) = 0.

+ (

+ ( + ) +

Particular solution:
55.

exp O

Particular solution:
54.

v
0 = 

Particular solution:
53.

+ ) + (

Particular solution:

Particular solution:

52.

= exp O

+ (

   + ( ' 2 + + ( +

51.

+1

The substitution =   +1 leads to a linear equation of the form 2.1.2.108: (B + 1) 2   +


(B + 1)(  + B )  + ( ' + ( ) = 0.

The substitution

50.

=  exp O

Particular solution:
49.

= 0.

For B = 1, we obtain the Euler equation 2.1.2.123. For B 1, the substitution H =  


'

leads to an equation of the form 2.1.2.108: (B + 1) 2 H   + (B + 1)(!H + B )  +  = 0.

Particular solution:
47.

  +1
 .
B + 1 P3M

X exp O

+ 1) = 0.

= 0.

+ ) = 0.

) = 0.

2003 by Chapman & Hall/CRC

Page 218

219

2.1. LINEAR EQUATIONS

56.
57.

+ (

+ (

+ (

+ [ ( + 1)

+ (

+ (

+ (

+ (

v
0 = 

Particular solution:
60.

=  exp O

Particular solution:
59.

) = 0.

Integrating yields first-order linear equation: " + (   + ' + ) = L .

Particular solution:
58.

+ )

Particular solution:

 

+1

 +

+1

+ 1)

+ (

+1

] = 0.
.

) = 0.

+ (

= exp O

B +1
+

=  exp O

+ [

+ (
+1

+


B +1

 

 +
+
+1

+ 1)
+1

] = 0.

^
(

) = 0.

2.1.2-3. Equations of the form (  +  )  +  ( )  + ( ) = 0.


61.

62.

1
2

= 0.

cos 4  + L 2 sin 4 


if  > 0,
L 1 cosh 4|  | + L 2 sinh 4|  | if   < 0.

Solution:
+

= 0.

1 b . The solution is expressed in terms of Bessel functions:


1
2

=

V
L

 2 '  + L 2
n  2 '  W ,

@ = |1  |.

where

2 b . For  = 12 (2B + 1), where B = 0, 1,  , the solution is:

63.



if  > 0,
M  cos 4  + L 2 M  sin 4 



M 
M
L 1 M  cosh 4|  | + L 2 M  sinh 4|  | if   < 0.



M
M
L



= 0.

1 b . The solution is expressed in terms of Bessel functions:


1
2

=

V
L

 "
   + L 2 n     W ,

where

@ = 12 |1  |.

2 b . For  = 2B , where B = 1, 2,  , the solution is:


=

64.





3P

cos     + L


3P

sin    

M1


L 1O
cosh     + L 2 O
sinh    
M
M
 P
 3P
 3
M
M

+ (

1 M

+ ) = 0.

This is a special case of equation 2.1.2.108 with 

= 1 and 

=

if  > 0,
if  < 0.

= 0.

2003 by Chapman & Hall/CRC

Page 219

220
65.

SECOND-ORDER DIFFERENTIAL EQUATIONS


+ + 12 = 0.
For B = 1, this is the Euler equation 2.1.2.123. For B 1, the solution is:

66.
67.




70.

sin

exp

B 1





B 1

+L

+ (1 3 ) 2 2 2 1 = 0.
= L 1 (   + 1) exp(   ) + L
Solution:

cos
exp


B 1


if  > 0,

+L

2 (

  + 1) exp(  ).

B 1

if  < 0.

+ + = 0.
If B = 1 and  = 0, we have the Euler equation 2.1.2.123. If B 1 and  0, the solution is
expressed in terms of Bessel functions:

1
2

L

hO

2  

B +1

+ + (

Particular solution:
69.

=
68.



+1

+L

n O

2  

B +1

+1
2

where

@ =

|1  |
.
B +1

+ ) = 0.

= exp O

+ + = 0.

Particular solution: 0 =   .

+1
2


 
B +1

+1

+ ( ) = 0.
The degenerate hypergeometric equation.
1 b . If  0, 1, 2, 3,  , Kummers series is a particular solution:

( )   
,
` ( ,  ;  ) = 1 +
 =1 (  )  , !
where (  )  =  (  + 1)  ( + , 1), (  )0 = 1. If  >  > 0, this solution can be written in
terms of a definite integral:

1  g
( )
` ( ,  ;  ) =
w
X  a 1 (1 a ) 
( ) (   ) 0
w
w

g
1
a is the gamma function.
where ( H ) = X  a
w
M
0
If  is not an integer, then the general solution has the form:

=L

` ( ,  ;  ) + L 2 

 ` (

a,

 + 1, 2  ;  ).

Table 14 gives some special cases where ` is expressed in terms of simpler functions.
The function ` possesses the properties:


` (  ,  ;  ) =  ` (   ,  ;  );


( )
M  ` (  ,  ;  ) =  ` (  + B ,  + B ;  ).

( )
M

The following asymptotic relations hold:

` ( ,  ;  ) = w
` ( ,  ;  ) =

( ) 
1
S
   1 + [O
( )
| | P
Q
( )
1
( ) 1 + [O
S
w
(  )
| | P
Q

if ED

+F ,

if ED

F .

2003 by Chapman & Hall/CRC

Page 220

221

2.1. LINEAR EQUATIONS

TABLE 14
Special cases of Kummers function ` (  ,  ; H )

H


2

 +1

1
2

3
2

1
2

3
2

Conventional notation




1 

 sinh 

Incomplete gamma function

 = (  ,  )


2

= ( ,  ) = X

2
2

ga

Error function

2
X exp(a 2 ) a
erf  =

erf 

1 
B !
O
5 2  ( )
(2B )! 2 P
1 
B !
O
5 2  +1 ( )
(2B + 1)! 2 P

Hermite polynomials

2 
2
5  ( ) = (1)  
M    ,

M 
B = 0, 1, 2, 3,
Laguerre polynomials

@ +

1
2

B +1

2@ + 1

2

2B + 2




  9
M      + 9  ,
B ! 
=  M 1,
(  )  =  (  + 1)  (  + B 1)

0 ( 9 ) (

B ! ( 1)
0  ( )
( ) 

(1 + @ ) 

2

OB +


O

2P

3  
 O
2P
2P

 (

12

)
+ 12 (

)=

Modified Bessel functions


 ( )

 )

2 b . The following function is a solution of the degenerate hypergeometric equation:

i ( ,  ;  ) =

(1  )
w
` ( ,  ;  ) +
(   + 1)

(  1)

( )

 ` (

 + 1, 2  ;  ).

Calculate the limit as DB (B is an integer) to obtain

i ( , B ;  ) =

(1)  1
` ( , B +1;  ) ln 
B ! ( B )

=0

( )
V j (  + & ) j (1 + & ) j (1 + B + & )W 
(B + 1)
&!


(B 1)!
( )

1
=0

( B )  
,
(1 B ) & !

where B = 0, 1, 2,  (the last sum is omitted for B = 0), j ( H ) = [ln ( H )]  is the logarithmic
w
derivative of the gamma function:

j (1) = = ,

j ( B ) = = +




= = 0.5772  is the Euler constant.

=1

2003 by Chapman & Hall/CRC

Page 221

222

SECOND-ORDER DIFFERENTIAL EQUATIONS


If  is a negative number, then the function i can be expressed in terms of the one with
positive second argument using the relation

 i (

i ( ,  ;  ) = 

 + 1, 2  ;  ),

which holds for any value of  .


3 b . For  0, 1, 2, 3,  , the general solution of the degenerate hypergeometric equation
can be written in the form:
=L

` (  ,  ;  ) + L 2 i (  ,  ;  ),

while for  = 0, 1, 2, 3,  , it can be represented as:


1

=

 V L 1 ` (

 + 1, 2  ;  ) + L

i (   + 1, 2  ;  )W .

The functions ` and i are described in Subsection S.2.7 in more detail; see also the
books by Abramowitz & Stegun (1964) and Bateman & Erdelyi (1953, Vol. 1).
71.

+ ) + [( ) + ] = 0.

+ (

Particular solution:
72.

+ (2

+ )
=

Solution:
73.

=

+ (


  (L

+ ) = 0.

1
 ) if  1,
ln
|

|) if  = 1.
2

+L
1+L
1

(L

+ [( + ) + +

Here, B and

v

] + (

are positive integers;   or B

 +
= L 1 M +

M

74.

+ ( + ) + ( + ) = 0.
This is a special case of equation 2.1.2.108.

75.

Particular solution:
76.

(2

+ (


= L
Q

78.

(2

Solution:

79.

+ (

= exp 

+ 1) +

+ ) = 0.

12

(  .
2

cos 

= 0.

5) + 2 2 ( 2)

+ ( 2 + )
Particular solution:

= ( 

1
2

 2   S .

+ 1) = 0.

= L 1 exp V 12   +  2    2 W + L

Particular solution:
80.

  +L
1
1 sin  2 

Particular solution:

+ ) = 0.

+ ) +

= 0.

= exp  12 ' 2  .

+ 1) + (

Solution:
77.

2(

+ 1)

 1
V    (  ) W + L 2    M
V  +  (  ) W .

1

M

Solution:

exp V

1
2

   2    2W .

= 0.
2

+ 1) exp(  ).

[ 2 + ( +
v .
0 =  

+ 2 ) +

+ 2 ] = 0.

2003 by Chapman & Hall/CRC

Page 222

223

2.1. LINEAR EQUATIONS

81.
82.
83.

+ ( 2 + + 2) + = 0.
Particular solution: 0 =  + 
 .

+ ( 2 + + ) + (2 + ) = 0.
Integrating, we obtain a first-order linear equation:  " + ( 

+ (

v.
0 = 

+ ' + ( 1) = L .

+ ) = 0.

Particular solution:
84.

+ ) + ( 1)(

+ ( 2 + + ) + (  2 + ( + = ) = 0.
1 b . Let # =  , , $ = , (  , ), L  = (, , where , is an arbitrary number.
Particular solution: 0 =   .

2 b . Let # =  (  + , ), $ =  ( ( + 1) , (  + , ), L = (, .
,  .
Particular solution: 0 = exp  12  2 + !
3 b . Let # =  (  + , ), $ = 2  , , 2 , L =  ( ( 1) + , ( ( 2).
Particular solution: 0 =  1 v exp  12  2 + ,!  .
4 b . Let # =  , , $ =  ( ( 1) , (   + , ), L
Particular solution: 0 =  1 v   .

85.

86.

+ ( 2 + + 2) + ( 2 + + ) = 0.
   + (  +  )2
 +
The substitution =  leads to a linear equation of the form 2.1.2.108: I

( ( +  ) = 0.

+ (

+ ) + ( 1)

+ (

.
0 = 

89.
90.

+ ( 3 +
Particular solution:

+ (

+ (

Particular solution:
93.

= 0.

+ ) = 0.

=  exp  13 

'  .

1) + 2

= (  + 1)  .

= 

+
=

+ ) + ( 1)(

x
0 = 

+ (

+ 2)

+
0

Particular solution:
92.

+ ( 3 + 2 + 2) + = 0.
Particular solution: 0 =  + 
 .

Particular solution:
91.

+ ) + (3

Particular solution:
88.

Particular solution:
87.

=  (( 1) + , ( ( 2).

+ ) = 0.

1
1

= 0.

+ 2 ) = 0.

.
= 0.

+ ( + 1 ) + 2 1 = 0.
1 b . For  14 , the solution has the form: = L 1 exp  1    + L
are roots of the quadratic equation: B 2 2 + B/ +  = 0.

2 b . For  = 14 , the solution has the form:

= (L

+L

2 exp

  ) exp  12 B

 2    . Here,
1

and

  .

2003 by Chapman & Hall/CRC

Page 223

224
94.

SECOND-ORDER DIFFERENTIAL EQUATIONS

+ (

Particular solution:
95.

+ (

+ (

+ (

+ (

+ )

+ (

=

.

+ (

v 
.
0 = 

= 0.

= 0.

+ ) = 0.

3 + 1) +

2 1

   + 1) exp(  ).
0 = (

Particular solution:
99.

exp(  
B ).

  
B ).
0 = exp( 

Particular solution:
98.

=

= 0.

+ ) + ( + 1)w

Particular solution:
97.

+ ) + ( 1)

Particular solution:
96.

+ ) +

+ ) + (

= 0.

) = 0.

This is a special case of equation 2.1.2.146 with = = 0.


100.

+ (

+ (

+ (

+ (

+ (

+ 0 ) + (

1) +

=

v.

+ ) +

Particular solution:
106. (

) = 0.

= 0.

+ 1 2 ) +

   + 1) exp(  ).
0 = (

Particular solution:
105. ( + ) + (

+ ) + ( 1)(

Particular solution:
104.

.
0 = (  + 1) 

+ (

Particular solution:
103.

= 0.

  
B ).
0 =  exp( 

Particular solution:
102.

0 =  +  .

Particular solution:
101.

+ 2) +

0 )

) = 0.

2 +

= 0.

= 0.

' + ( 1
 .
 + ?M P

= exp O X
+

= 0.

= 1, 2, 3,  , a polynomial of order


in  is a particular solution of the equation,

+
1 
, +  + 1 [( 1  +  0)c 2 +  0 c ] .   + , where
O
which can be represented as: 0 =
If

+1

=0

with @ 1.
c = M ,  =

@ +1
M

107. (

+ ) + s(

Particular solution:

 1P

+ ) s [( + ) +
0

=

+ ] = 0.

2003 by Chapman & Hall/CRC

Page 224

225

2.1. LINEAR EQUATIONS

TABLE 15
Solutions of equation 2.1.2.108 for different values of the determining parameters



=   ( H ), where H =

Solution:

Constraints

 2 0,
2
 1 4 0 

 2 = 0,
 10
 2 0,
 = 4 0 
2
1

c 
2 2
 0

 1

 2 =  1 = 0,
 00

2

2

2 2 , + 

2
1

2 2 , + 

Notation: c

2
2

2
1 4 0 2,

=

Parameters

 = $ ( , )
(2  2 , +  1 ),
 = (  2  1  1  2 )  2
2

 = $ ( , ) (2  1),
= 
(2  )

( ,  ; H )

  , 12 ; H 2 

:
H J2 /
 H 

 21 4  0 
4 0  2

 

@ = 1 (2  2 , +  1 )  1
2 ,
= 2 $ (, )

1J 2
2 
= O 0
3 2P

:
H 1 J 2 1 J 3  H 3 J 2 
see also 2.1.2.12

$ (, ) =  2 , 2 +  1 , + 

108. ( 2 + 2 ) + ( 1 + 1 ) + ( 0 + 0 ) = 0.
Let the function (  ,  ;  ) be an arbitrary solution of the degenerate hypergeometric equation
:
   + (   )  y  = 0 (see 2.1.2.70), and the function ( ) be an arbitrary solution of



the Bessel equation  2  +   + ( 2 @ 2 ) = 0 (see 2.1.2.126). The results of solving the
original equation are presented in Table 15.
109. ( + )

+ (

Particular solution:

+ )

= exp O X

+ ( 1 + 1 ) = 0.
  + ' + + ( 1
 .
 +
M P

2.1.2-4. Equations of the form  2   +  ( )  + ( ) = 0.


110.

+ = 0.
g
This is a special case of equation 2.1.2.123. The substitution  =  leads to a constant



coefficient linear equation: g g g +  = 0.

111.

+ ( + ) = 0.
This is a special case of equation 2.1.2.132.

112.

+ [

Solution:
113.

2
2

( + 1)] = 0,
= 0, 1, 2, uuu
L
cos



+
L 2 sin 
, where c
  +1 = ( 3 c )  O 1
2  1

[ 2

Solution:
114.

2 2

+ ( + 1)] = 0,

= 0, 1, 2,

uuu

= M .



L 1  + L 2 

+1
3

= ( c ) O
, where c = M .


 2  1
P
M

2 2

+ 2

Particular solution:

) = 0.


=   .

2003 by Chapman & Hall/CRC

Page 225

226
115.

116.

SECOND-ORDER DIFFERENTIAL EQUATIONS


+ ( 2 + + ) = 0.
The substitution =   , where is a root of the quadratic equation 2 + ( = 0, leads to
an equation of the form 2.1.2.108:     + 2  + (  +  ) = 0.
For  = 14 ,  = , , and ( = 14 2 , the original equation is referred to as Whittakers

equation.

5
16

Particular solution:
117.

2 4

= 0.
0

J exp  2 r
3J 2 
.
3  

1 4

=

+ (2 1)

0 =  

exp  21  2  .

Particular solution:

+ ( + 1)] = 0.

118.

+ ( + ) = 0.
This is a special case of equation 2.1.2.132.

119.

2 2

+ (2 + 1) w + ( 1)] = 0.
exp(  
B ).

=

Particular solution:
120.

+ ( 2 + + ) = 0.
This is a special case of equation 2.1.2.146.

121.

+ 

1
4

1
4

The transformation =   +  ,


2  + ( B )2 = 0.

= 0.

122.

+ V

2 (

+ )

1
4

1
4

The transformation = '  + ( ,


2  +  ( 'B )2 +2 = 0.
123.

= 

1
2

leads to an equation of the form 2.1.2.7:

= 0.

= 

1
2

leads to an equation of the form 2.1.2.7:

+ + = 0.
The Euler equation. Solution:






| |

1
2

| |

1
2

| |

1
2

L 1 | |  + L 2 | |  

if (1  )2 > 4  ,

(L

if (1  )2 = 4  ,

V
L

+L

sin( ln | |) + L

ln | |)
2

cos( ln | |)W

if (1  )2 < 4  ,

where = 12 |(1  )2 4  |1 J 2 .
124.

125.

+ + V 2  + 12  W = 0,
= 0, 1, 2, uuu
This is a special case of equation 2.1.2.126.

1
sin 
cos 
OL 1
S .
=   +1 J 2 O
+L 2
Solution:
M

3P

Q 

+ V 2 +  + 12  W = 0,
= 0, 1, 2,
This is a special case of equation 2.1.2.127.


1



+1 J 2
OL 1 + L 2
O
S .
=
Solution:
M

Q 

3P

uuu

P

2003 by Chapman & Hall/CRC

Page 226

227

2.1. LINEAR EQUATIONS

126.

+ + ( 2 b
The Bessel equation.

) = 0.

1 b . Let @ be an arbitrary noninteger. Then the general solution is given by:


=L

( ) + L 2
n ( ),

(1)

where ( ) and n ( ) are the Bessel functions of the first and second kind:

(1)  (
2) +2 
( ) cos  @ ( ) .
, n ( ) =
( ) =
,
! ( @ + , + 1)
sin @
 =0 w

:
Solution (1) is denoted by = ( ) which is referred to as the cylindrical function.

(2)

The cylindrical functions possess the following properties:

2@

( ) =  [

 ) + +1 ( )],
:
:
M [ ( )] = 


1 (

: ( )] =  : ( ),
M [
1
+1 (  ).

M ( ) and n ( ) can be expressed
M
in terms of definite integrals (with  > 0):
The functions




( ) = X

n ( ) = X

cos( sin @

sin( sin @ )

M
M

sin @ X exp( sinh a @!a ) a ,


 0
M

X (
0

g + g cos @ ) 



sinh

g a.
M

2 b . In the case @ = B + 12 , where B = 0, 1, 2,  , the Bessel functions are expressed in terms
of elementary functions:

+ 12 (

 )=

2 


+ 12

 sin 
2 

O M
,  1 ( ) =
2
 3P


M
n  + 12 ( ) = (1)  +1  12 ( ).
1

+ 12

O


 cos 
,
3P


3 b . Let @ = B be an arbitrary integer. The following relations hold:

n  ( ) = (1)  n  ( ).
The solution is given by formula (1) in which the function  ( ) is obtained by substituting
@ = B into formula (2), while n  ( ) is found by taking the limit as @*D B and for nonnegative B

 ( ) = (1)    ( ),

becomes

n  ( ) =



where j (1) = z , j (B ) = z +

( ) ln

(1)  O
7


=0
1
=1

2P


1



+2

=0

(B , 1)! 2 
O
, !
2P

 j ( , + 1) + j (B + , + 1)
,
, ! (B + , )!

, z = 0.5772  is the Euler constant, j ( ) = [ln ( )] 

is the logarithmic derivative of the gamma function.

For nonnegative integer B and large  , we can write

 2  ( ) = (1)  (cos  + sin  ) + [ ( 2 ),



 2  +1 ( ) = (1)  +1 (cos  sin  ) + [ ( 2 ).

The function !
 ( ) can be expressed in terms of a definite integral:
1 z
B = 0, 1, 2, 
  ( ) = X 0 cos( sin a B/a ) M a ;

The Bessel functions are described in Subsection S.2.5 in more detail; see also the books

by Abramowitz & Stegun (1964) and Bateman & Erdelyi (1953, Vol. 2).

2003 by Chapman & Hall/CRC

Page 227

228
127.

SECOND-ORDER DIFFERENTIAL EQUATIONS

+ ( 2 + b 2 ) = 0.
The modified Bessel equation. It can be reduced to equation 2.1.2.126 by means of the
substitution  =  ( 2 = 1).
Solution:
= L 1  ( ) + L 2 ; ( ),
where  ( ) and ; ( ) are the modified Bessel functions of the first and second kind:

( ) =

=0

(
2)2  +
,
, ! ( @ + , + 1)

; (

)=

 )  (
2 (sin
 @

The modified Bessel function  ( ) can be expressed in terms of the Bessel function:

(  z J 2 ), 2 = 1.
The case @ = B + 12 , where B = 0, 1, 2,  , is given in 2.1.2.125.
If @ = B is a nonnegative integer, we have
 1
1
 2 +  (B 1)!


+1
+
(

)
=
(1)

(

)
ln
(1) + O

;|

2 2+
2P
!
=0

  +2 + j (B + + 1) + j ( + 1)
1
,

+ (1)  O
2
! (B + )!
+ =0 2 P

where j ( H ) is the logarithmic derivative of the gamma function (see 2.1.2.126, Item 3 b ); for
B = 0, the first sum is omitted.
As D + F , the leading terms of the asymptotic expansion are:




 ( )
, ; ( )   .
2 
2

( ) = 

The modified Bessel functions are described in Subsection S.2.6 in more detail; see also
the books by Abramowitz & Stegun (1964) and Bateman & Erdelyi (1953, Vol. 2).
128.
129.

+ 2
Solution: 

Solution:
130.

Solution:
131.

132.

( 2 2 + 2) = 0.


= L 1 (  1)  + L 2 (  + 1)  .

+ [ 2 2 + ( + 1)] = 0.
| | ( L 1 sin ' + L 2 cos ' ) if  0,
L 1 | | + L 2 | | +1
if  = 0.

+ [

2 2

+ ( + 1)] = 0.



| | ( L 1   + L 2   ) if  0,
L 1 | | + L 2 | | +1
if  = 0.

+ + ( 2 + + ) = 0.
The substitution =   , where , is a root of the quadratic equation , 2 + ( 1) , + ( = 0,
leads to an equation of the form 2.1.2.108:    + ( + 2 , )  + (  +  ) = 0.

+ + ( + ) = 0,
0.
The case  = 0 corresponds to the Euler equation 2.1.2.123.
For  0, the solution is:


1
2
2
=  2 L 1 O " 2 + L 2 n O  2 S ,

where @ =  1 (1  )2 4 ( ; ( H ) and n ( H ) are the Bessel functions of the first and second

kind.

2003 by Chapman & Hall/CRC

Page 228

229

2.1. LINEAR EQUATIONS

133.

134.

135.

136.
137.

+ + ( + ) = 0.
The substitution =   leads to an equation of the form 2.1.2.108: B
( ' + ( ) = 0.

+ ( + ) + ^ = 0.

= H   , where , is a root of the quadratic equation


The transformation  = H 1 ,
, 2 + (1  ) , + ( = 0, leads to an equation of the form 2.1.2.108:

H '  + [(2  )H + 2 , + 2  ]  + [(1  )H + 2 , + 2  , ] = 0.

+ 2 + ( 2 + + ) = 0.
= exp  12   leads to a linear equation of the form 2.1.2.115:
The substitution


1
2 2
2
2
 + [( 4  +  ) + ( + ] = 0.


+ (

+ (

+ ) + [( )

Particular solution: 0 =  v .
+

+ (

140.
141.

=


 

+ [ ( )

  
.

0 = 

2 2

+ ] = 0.

= 0.

Particular solution:

139.

 + B (B 1 +  )  +

Particular solution:
138.

+ ( +

2 I ) + ( 1)] = 0.

+ ( 1 2 + 1 ) + ( 0 2 + 0 + 0 ) = 0.
The substitution =   , where , is a root of the quadratic equation  2 , 2 +(  1  2 ) , + ( 0 = 0,
leads to an equation of the form 2.1.2.108:  2  2   +(  1  +2  2 , +  1 )2 +(  0  +  1 , +  0 ) = 0.
+ [ 2 + ( 1) + ] + 2

Particular solution: 0 = (  + 1)  .

= 0.

2 ( 2 ) + {2 2 + [(1) 1] } = 0.
For B = 0, 1, 2,  , particular solutions are polynomials:

2
0

=  ( ), where

0(

 ) = 1,

1(

 )= ,

2(

 ) = 2 2 1 2  ,

3(

 ) = 2 3 (3 + 2  ) , 

The polynomials contain only even powers of  for even B and only odd powers of  for
odd B .
142.

+ ( 2 + + ) + (  3 + ( 2 + = + F ) = 0.
1 b . The substitution =   , where , is a root of the quadratic equation ,
leads to an equation of the form 2.1.2.84 (see also 2.1.2.802.1.2.83):

    + ( 

+ ' + ( + 2 , )  + [ #*

+( ( 1) , + c = 0,

+ ( $ +  , ) + L + , ] = 0.

2 b . Let and & be arbitrary parameters.


For # = & , $ = ! + '& & 2 , L =  + (& 2& , c = ( ( 1), a particular solution is:


.

0 = 
For # =  (  & ), $ =  ( ( + 1) + & (  & ), L =  + (& 2& , c = ( ( 1), a particular
solution is: 0 =  exp( 21  2 & ).
143.

Particular solution:

v 
0 =   

+ 2 ^

) = 0.

2003 by Chapman & Hall/CRC

Page 229

230
144.

SECOND-ORDER DIFFERENTIAL EQUATIONS

Particular solution:
145.

+ (

147.

148.

+ )

2 4

=  + exp O

Particular solution:

146.

+2

+ (

+ (

+1

 2+

+1

) = 0.

1) = 0.

.

=

+2

+ ( + ) + (  2 +  + ) = 0.
The transformation H =   , = H  , where , is a root of the quadratic equation
B 2 , 2 + B (  1) , + = = 0, leads to a linear equation of the form 2.1.2.108: B 2 H 2'  +
[B!H + 2 ,!B 2 + B (B 1 +  )]3 + ( hH + ,!B + ) = 0.

+ (2 + ) + [ 2 2 + ( + 1)w +  2 +  + ] = 0.
The substitution = exp( 
B ) leads to a linear equation of the form 2.1.2.146:
 2 2   + ' 2 + (  2 + +  + + = ) = 0.

2
2

+ (

+2

+ ) + (

Particular solution:

= exp O

2.1.2-5. Equations of the form ( 

B +1

+1

 

+1

'

^ )

= 0.

+ ' + ( )   +  ( )  + ( ) = 0.

149. (1 2 ) + ( 1) = 0,
= 0, 1, 2, uuu
This equation is encountered in hydrodynamics when describing axially symmetric Stokes
flows.
1 b . For B 2, the solution is given by:
=L

 ( ) + L 2
{  ( ),

where  ( ) and { ( ) are the Gegenbauer functions which can be expressed in terms of the
y functions of the first and second kind (see 2.1.2.153) as follows:
Legendre

 ( ) =

 )  ( )
,
2B 1

2 (

2 b . For B = 0 and B = 1, the solution is:

150. (

2 )

6 = 0.

Particular solution:

=L

{ (
1

)=

+L

 .

 )  ( )
.
2B 1

2 (

= (4  )| +  | 2  |  | 2  .
2 +

151. ( 2 1) + + = 0.
1 b . For  = , 2 > 0, the solution is:
=

cos( , arccosh | |) + L 2 sin( , arccosh | |) if | | > 1,


L 1 exp( , arccos  ) + L 2 exp( , arccos  )
if | | < 1,

where arccosh  = ln   + 
2 b . For  = ,

1 .

< 0, the solution is:


=

exp( , arccosh | |) + L 2 exp( , arccosh | |) if | | > 1,


L 1 cos( , arccos  ) + L 2 sin( , arccos  )
if | | < 1.
1

3 b . For  = B 2 , where B is a nonnegative integer, particular solutions are the Chebyshev


{
polynomials:  ( ) = cos(B arccos  ).

2003 by Chapman & Hall/CRC

Page 230

231

2.1. LINEAR EQUATIONS

= 0, 1, 2, uuu
152. (1 2 ) + 2 = 0,
This is a special case of equation 2.1.2.151 with  = B
[ J
B

2 +

{
where

2]

(1) +

=0

. Particular solution:

(1) 
1
2   12  

=  ( ) = cos(B arccos  ) =

(B 1)!
(2 ) 

! (B 2 )!


M  V (1  2 ) 


21

+ ,

 ( ) is the Chebyshev polynomial of the first kind, (  )  =  (  + 1)  ( + B 1), and


[  ] stands for the integer part of a number  .

153. (1 2 ) 2 + ( + 1) = 0,
The Legendre equation.
The solution is given by:
=L

= 0, 1, 2,

uuu

 ( ) + L 2  ( ),

where the Legendre polynomials  ( ) and the Legendre functions of the second kind  ( )
are given by the formulas:

 ( ) =

1
B ! 2

.
The functions
.

0(

 ) = 1,


M  (


 1.
1.
1+
 ( ) =
(

)
ln

+

2
1
+ =1

1)  ,

1 (

 )  + ( ).

 =  ( ) can be conveniently calculated by the recurrence relations:


1(

 )= ,

2(

1
2

 ) = (3 2 1),  ,

+1(

 )=

2B +1

B +1

B .
 ( ) B +1 

1(

 ).

Three leading functions  =  ( ) are:

0(

 )=

1 1+
ln
,
2 1

1(

 )=

.
zeros of the polynomial

ln

1+
1,
1

2(

 )=

3

3
1 1+
ln
 .
4
1
2

All B
 ( ) are real and lie on the interval 1 <  < 1; the
.
functions  ( ) form an orthogonal system on the closed interval 1  1, with the
following relations taking place:

1
1

 (

.
)

+ ( )  =
M

2
2B + 1

if B
if B =

154. (1 2 ) 2 + b ( b + 1) = 0.
The Legendre equation; @ is an arbitrary number. The case @ = B where B is a nonnegative
integer is considered in 2.1.2.153.
The substitution H =  2 leads to the hypergeometric equation. Therefore, with | | < 1 the
solution can be written as:
=L

@ 1+@ 1
, ; 
O ,
2

+L

"O

1@
@ 3
, 1+ , ;  ,
2
2 2 P

where  ( , , = ;  ) is the hypergeometric series (see 2.1.2.171).


The Legendre equation is discussed in the books by Abramowitz & Stegun (1964),
Bateman & Erdelyi (1953, Vol. 1), and Kamke (1977) in more detail (see also Subsection
S.2.9).

2003 by Chapman & Hall/CRC

Page 231

232

SECOND-ORDER DIFFERENTIAL EQUATIONS

155. (1 2 ) 3 + ( + 2) = 0,
Particular solution:
0

=  ( ) =

J

2]

sin[(B + 1) arccos  ]

1

(B )!
(2 ) 

! (B 2 )!

(1) +

=0

=
2

= 1, 2, 3,

uuu

(1)  (B + 1)
1
1
2 +1  1 
2  +1


M  V (1  2 ) 


+ 21

+ ,

where  ( ) is the Chebyshev polynomial of the second kind, (  )  =  (  + 1)  ( + B 1),


and [  ] stands for the integer part of a number  .

156. (

1)

+ 2( + 1)

= M 


Solution:
2.1.2.154.
157. (

( b + + 1)( b ) = 0,

uuu

= 1, 2, 3,

( ), where ( ) is the general solution of the Legendre equation

1) 2( 1)

( b + 1)( b + ) = 0,

Solution:
= | 2 1|  M 

M
equation 2.1.2.154.
158. ( 2 1) + (2 + 1)
1 b . Particular solution:
0

uuu

= 1, 2, 3,

( ), where ( ) is the general solution of the Legendre

(2 + ) = 0.

(2  +  )
  2  +  ,  ,  + 12 ;
(  + 1) (2 )

1
2

12   ,

(1)

where  ( , , = ; H ) is the hypergeometric function (see equation 2.1.2.171 and Subsection


S.2.8).
2 b . For  = B , where B = 0, 1,  , the right-hand side of (1) defines the Gegenbauer polynomials,


(2  + B )
(  + , ) (2  + B + , )( 1) 
w
w
.
L  ( ) ( ) = w
  2  + B , B ,  + 12 ; 12 12   =
(B + 1) (2 )
, ! (B , )! 2  (  ) (2  + 2 , )
=0

w
w
w w

159. (1 2 ) + (2 3)
Particular solution:
2
0 (  ) = (1  )

+ ( + 1)( + 2 1) = 0,

J L ( ) ( ) = (1  2 )


1 2

= 0, 1, 2,

uuu


(  + , ) (2  + B + , )( 1) 
J
,
w
w 
 =0 , ! (B , )! 2 w (  ) w (2  + 2 , )

1 2

where L  ( ) ( ) are the Gegenbauer polynomials.

160. (1 2 ) + V   (  +  + 2) WN + ( +
Particular solution:

+  + 1) = 0,

= 0, 1, 2,

uuu


.
(1) 
(1  ) 9 (1 +  ): M  (1  ) 9 +  (1 +  ): +  S
 ) =  9 ,: (  ) = 
 Q
2 B !
M


+
+
+
+
= 2 
L  + L   +: ( 1)  ( + 1) ,
9
+ =0
.
where  9 ,: ( ) are the Jacobi polynomials and L are binomial coefficients.

0(

2003 by Chapman & Hall/CRC

Page 232

233

2.1. LINEAR EQUATIONS

161. (1

) + V   + (  +  2) WU + ( + 1)( +  +  ) = 0,

Particular solution: 0 ( ) = (1  ) 9 (1 +  ):

mials (see 2.1.2.160).

162. (

+ )

The substitution H = X
163. (

2 ) + 2
=L

Solution:
165. (

2)

166. (

+ )

leads to a constant coefficient linear equation: ' + (

+

= |

= 0.

+  |1  .

+ ( 1) = 0.

+ ( 1) = 0.

  |1  + L 2 | +  |1  .

+ 2
=L

Solution:

1|

uuu

+ 2( 1) = 0.

Particular solution:
164. (

= 0, 1, 2,

= 0.

M

+ ) + 2

. ,:
,:
 9 ( ), where  9 ( ) are the Jacobi polyno-



+  2

1
2

sin Y + L

+ (2 + 1)



1
2

+  2

= 0,

cos Y , where Y = (1  ) arctan( 


 ).

uuu

= 1, 2, 3,

This equation can be obtained by B -fold differentiation of an equation of the form 2.1.2.162:
(  2 +  )   +   + ( ( B 2 ) = 0.
= (  ) .
Solution:
167. (1

+ (2

+ ) = 0.

This is an algebraic form of the Mathieu equation. The substitution  = cos H leads to the
'
Mathieu equation 2.1.6.29:   + (  +  +  cos 2 H ) = 0.
168. (1

+ (

+ )

= 0.

1 b . The substitution 2 H = 1 +  leads to the hypergeometric equation 2.1.2.171: H (1 H )/' +

[ !H + 12 (   )]  + ( = 0.
2 b . For  = 2

3,  = 0, and ( = , the Gegenbauer functions are solutions of the equation.

3 b . In the special case  = 2,  = , and ( = B (B + + + 1), solutions of the


equation are the Jacobi polynomials:


9 : ( ) = 2 
L  + + L   +: + ( 1)  + ( + 1) + ,
9
+ =0

( , )

where L are binomial coefficients (see Paragraph S.2.1-2).


169. (

+ ) + (

Particular solution:
170. (

+ )

+ ) + [(  )
0

 ] = 0.

=  .

+ [ ( + )

Particular solution:

+ ( ) + 2  ]

+ ) = 0.

= (  + 1)   .

2003 by Chapman & Hall/CRC

Page 233

234
171.

SECOND-ORDER DIFFERENTIAL EQUATIONS

1) + [(  +  + 1) ] +  = 0.
The Gaussian hypergeometric equation. For = 0, 1, 2, 3,  , a solution can be
expressed in terms of the hypergeometric series:

( )  ( )   
, ( )  = ( + 1)  ( + , 1),
 ( , , = ;  ) = 1 +
, !
 =1 (= ) 
which, a fortiori, is convergent for | | < 1.
For = > > 0, this solution can be expressed in terms of a definite integral:
 ( , , = ;  ) =

(= )
X
( ) (= )

where ( ) is the gamma function.

1
0

a:

(1 a );

(1 a ) 9

a,
M

If = is not an integer, the general solution of the hypergeometric equation has the form:
=L

 ( , , = ;  ) + L 2 

;  ( = + 1, = + 1, 2 = ;  ).

In the degenerate cases = = 0, 1, 2, 3,  , a particular solution of the hypergeometric


equation corresponds to L 1 = 0 and L 2 = 1. If = is a positive integer, another particular solution
corresponds to L 1 = 1 and L 2 = 0. In both these cases, the general solution can be constructed
by means of the formula given in 2.1.1.
Table 16 presents some special cases where  is expressed in terms of elementary
functions.
Table 17 gives the general solutions of the hypergeometric equation for some values of
the determining parameters.
The function  possesses the following properties:

 ( , , = ;  ) =  ( , , = ;  ),
 ( , , = ;  ) = (1  ) ; 9 :  (= , = , = ;  ),

),
 ( , , = ;  ) = (1  ) 9  ( , = , = ;
 1

( )  ( ) 
 ( + B , + B , = + B ;  ).
M   ( , , = ;  ) =

(= ) 
M

The hypergeometric functions are discussed in the books by Abramowitz & Stegun (1964)
and Bateman & Erdelyi (1953, Vol. 1) in more detail; see also Subsection S.2.8.

172.

+ ) + ( + ) + I = 0.
The substitution  = !H leads to the hypergeometric equation 2.1.2.171: H (1 H ) ' +
= 0.
[( (
 ) H ] 

173. 2 ( 1) + (2 1) + ( + ) = 0.
The substitution  = cos2 leads to the Mathieu equation 2.1.6.29:  (  +2  +  cos 2 ) = 0.
174. (

+ 2

Solution:

+ ) + ( + )
= L 1  + + 

= 0.

+ 2  +  

+L

 + + 

+ 2  +  

175. ( 2 + + ) + ( + ) + ( 2 ) = 0.
Integrating yields a first-order linear equation: (  2 + ' + ( )  + [( 2  ) + ,  ] = L .
176.

+ ) + (

Particular solution:

+ )
= ,! + .

= 0.

2003 by Chapman & Hall/CRC

Page 234

235

2.1. LINEAR EQUATIONS

TABLE 16
Some special cases in which the hypergeometric function
 ( , , = ; H ) is expressed in terms of elementary functions




+ 12

1
2

+ 12

3
2

1
2

1
2

12

2 1

1
2

sin 

3
2

sin2 

3
2

sin2 

1
2

sin2 

+ 12

1
2

tan2 

+1

+ 12

+ 12
1
2

=0

=0

(B )  ( )   
,
(B )  , !

1
2

(B )  ( )   
,
(= ) 
, !

2 + 1

where B = 1, 2, 3, 

(1  ) 9

V (1 +  )2 9 + (1  )2 9 W

(1 +  )12 9 (1  )12 9
2 (1 2 )

V  1 +  2 +   29 +  1 +  2   29 W
2 1
2 1
 1+ 2 +  9 +  1+ 2   9
2 1 +  2

1
2

22 9

1
2

1
2

where B = 1, 2, 3, 

 1+ 1 

22

cos(2  )
sin[(2 1)
( 1) sin(2
sin[(2 2)
( 1) sin(2
cos[(2 1)
cos 

]
)
]
)
]

cos2 9  cos(2  )
(1 +  )(1  ) 9

2
 12 + 12 1   9
1 J 2 1
 1 
 2 + 12 1  

1
2

3
2

3
2

1
2

3
2

1
2

1
2

3
2

B +1

B + +1

B + + , +2




arcsin 

arctan 

ln( + 1)

12

1+
1
ln
2
1

2
2

ln   + 1 + 

 
(1) + (B + + , + 1)!  + +

M  + + (1  ) + +  M 

B ! , ! (B + )! ( + , )! 
1
M
M
 =  ln(1  ), B , , , = 0, 1, 2, 3, 

2003 by Chapman & Hall/CRC

Page 235

236

SECOND-ORDER DIFFERENTIAL EQUATIONS

TABLE 17
General solutions of the hypergeometric equation for some values of the determining parameters

+ 12

2 + 1

12

1
2

+ 12

3
2

 QL

+1

+ 2

The substitution = X
form 2.1.2.1:   +

178. (

+ 2

+ ) + (



= 0.

+ ) + 3(

 1+  

9 + L 2

12

 1+  
1

| | 9 OL

+ ) +

X | |; | 1|; :

 1+ 1 

| 1|: OL

| |1; | 1|; :

1+

177. (

= ( )

Solution:

OL

1+
1+

9 +L

12

1+

2
2

12

 1  

X | |;

9  1 + 1   29

12

| 1|: ;

X | | 9 | 1|:

X | | 9

 1  

9 +L
2

| 1|:

9 S

M P


M P


M P

= 0.
leads to a constant coefficient linear equation of the

+ 2 ' + (

+ ) + I = 0.

|  2 + 2 ' + ( | leads to a linear equation of the form 2.1.2.177:


The substitution =

(  2 + 2 ' + ( )2   + (  +  )2 + (  ) = 0.


M

179. (

Let

+ 2 ) + (

and

1 b . For

1 )

= 0.

are roots of the quadratic equation  2

2 , the substitution H =

H (1 H ) ' ( #* + $ )  L


2

1
1

+  2 + (

= 0.

leads to the hypergeometric equation 2.1.2.171:

= 0, where # =

1
2

, $

 1 1 + ( 1
(0
, L =
.
 2( 2 1)
 2
=  , where , is a root of the

2 b . For 1 = 2 = , the transformation  = + 1 ,


quadratic equation  2 , 2 + (  2  1 ) , + ( 0 = 0, leads to a linear equation of the form 2.1.2.108:
 2    [( ( 1 + " 1 ) +  1 2  2 ( , + 1)]  , ( ( 1 + " 1 ) = 0.
3 b . Let ( 0 =  2 B (B 1)  1 B , where B is a positive integer. Then, among solutions there
exists a polynomial of degree B .

180. (

+ ) (

Particular solution:
181. (

+ ) + (

Particular solution:

2)

+ ( + ) = 0.

= , .
3

3 )

2 )

= 0.

= +, .

Page 236

237

2.1. LINEAR EQUATIONS

2.1.2-6. Equations of the form (  3 


182.
183.

184.
185.
186.

188.

+ ( 2 + ) + = 0.
'

The substitution  = 1
H leads to an equation of the form 2.1.2.139: H 2   + H (2  H )  +
( = 0.

+ ( 2 + )
Particular solution:

+ = 0.
=  2+
 .

+ ( 2 + ) + ^ = 0.
The substitution  = 1
H leads to an equation of the form 2.1.2.108: H ' +(2  H )  + (

= 0.

+ ( 2 + ) + ( + ) = 0.
1 b . The substitution =   , where , =
 , leads to a linear equation of the form
M
2.1.2.134:  2   + [(  + 2 , ) +  ]  + [ , (  + , 1) + ( ] = 0.

=  (  2), a particular solution is:

+ ( 3 2 + + ) + 2

Particular solution: 0 = (  + 1)  .

+ ( + ) (
Particular solution: 0 =  exp( 
 ).

3
(

+ ) + 2(
Particular solution:

190.

+ ) 2
=  + 
 .

2
0

2
+ ) + (


= J .

= 0.
+ ) = 0.

= 0.

+ ) + s = 0.
'
The substitution !H =  2 leads to the hypergeometric equation 2.1.2.171: H (1 H )"  +


1 V
1
1
(1 +  ) HW
4 = 0.
2 1 + (

2(

191.

+ ) + [
Particular solution:

192.

2(
2(

+ ) 2 (

Solution:

L 1

+ (2 +  ) +

 ).
0 = exp(

+ 2 ) + 2(

+ L 2
.
 + 
2




2(

 ]

( +

L 1 | |  + L 2 | | + +1
 + 
| |  ( L 1 + L 2 ln | |)
 + 

) ] + [

( 1) +

+ 1)] = 0.

B 1,

if
if

+ ( 2 ) = 0.

+ 3 ) = 0.

+ ) + [ (2

Solution:

194.

+  1  +  0 )   +  ( )  + ( ) = 0.

+ ( + ) = 0.
This is a special case of equation 2.1.2.132 with B = 1.

189.

193.

+  2

2 b . If ( = 0 and
187.

= B 1.

+ 2 ) + ( 1 + 1 ) + ( 0 + 0 ) = 0.
The substitution =   , where , is a root of the quadratic equation  2 , 2 + , (  1  2 )+  0 = 0,
leads to a linear equation of the form 2.1.2.172:  ( +  2 )   + [(2 , +  1 ) + 2 ,  2 +  1 ]  +
[ , 2 + , (  1 1) +  0 ] = 0.

2003 by Chapman & Hall/CRC

Page 237

238

SECOND-ORDER DIFFERENTIAL EQUATIONS

195. (

Particular solution:

196. (

Particular solution:

197. (

Particular solution:

198. (

Particular solution:
199. (

Particular solution:

+ (
0

+

=

 .

= ( +

200. ( 3 + 2 + ) + (
With the constraint

)

a particular solution has the form:

The substitution = X O
2  +

203.

204.

+ ) + (

= 0.



+ + 1) + ( 
The substitution =  1;
2

+ 2(

+ 1) = 0.

+ ( ( + , )(2 , ).

+ ) + ( 1)[( ) +

+ (2 ^

+ (2  + 4 (

1) ] + (2

= 0.

+ 1) = 0.

1)( + ( ).

+ ) + [2( + ) + 1] = 0.

+ 1)[

= (2  + B )

201. ( 3 + 2 + ) + 2 ( 2 )

Particular solution: 0 = (  + 2)  .

+ 2 ) ( 

2(2  + B )(( + , ) + (2  + 2

202. 2 (

( + 1) + ]

+ [(
0

= (  , +  1)

+ (
0

+ 2  ) = 0.

( + (  )(2  )
=
 + 2(  ) + , where =
.


2

+ [2
0

+ 2 ) + ( 2 ) = 0.

= 2  + (2  )

+ (
0

+

) +

1 2

+ ' + ( P

+ 1 + 2 , (  + B )] = 0,
+ (2  + 2

= 0.

= 0.

+ 1)( , ).

 leads to a constant coefficient linear equation:

+  + ) + ( + ) = 0.
leads to an equation of the same form:

+ ' + 1)    + [( + 2  2 = ) 2 + ( + 2  2 = ) + 2 = ] 


+ , [B + (1 = )( = )] + + (1 = )( = ) . = 0.

 ( 

1)( ) + , ( 

+  + 1)

[  +  + 1 + ( + S ) S ] + /
.

+ ( 

= 0.

Heuns equation.
1 b . For |  | 1 and = 0, 1, 2, 3,  , a solution can be represented as the power series:

 ( ,  ; , , = , > ,  ) = (    ,
 =0
where the coefficients are determined by the recurrence formulas:

( 0 = 1,

=/( 1 =  ,

 (B + 1)(= + B )( 

+1

=  (= + > + B 1) + + > + B +

SB( 

V (B 1)(B 2) + (B 1)( + + 1) + W ( 

1 .

A fortiori, the series is convergent for | | 1.

2003 by Chapman & Hall/CRC

Page 238

239

2.1. LINEAR EQUATIONS

TABLE 18
Some transformations preserving the form of Heuns equation
No
1*
2
3
4
5
6
7
8
9
10
11
12

Parameters of transformed equation for = ( )

New variables




= , =

=1 , =


1
 1 = +1 =

=  , = | |; 1

 2 = +1 =
1
= , = | |; 1



1
1
3

=
= , = | | 9



1

= , =




1

=1 , =

1




= , = | | 9

= +1 +=

1
 1


=
=
, = | | 9
1



 ( 1)


, = | | 9
=
=
 1
 (  1)



, = | 1| 9
=

>
 1
 1
 (  1)
1


, = | 1| 9
=
1
>
 ( 1)

Notation:  1 =  + ( = + 1)( = + 1) + > (= 1),
 3 =  1 + ( = + 1) + h 1 (> ) !>

=
>
=
>
+1

2=

>

+1

2=

+ =
> +1

+1 +1

>

+ =
> +1

+ =
> +1

+1

>

+1

+1

>

+ =
> +1

1 +1 + =
> +1

+1

+1

+1

+ =
> +1

 2 =  1 + !> (1 =
.

* This row corresponds to the original equation, while the others refer to the transformed equation for

),

 (~ ).

2 b . If = is not an integer, the general solution of Heuns equation can be presented as follows:
=L

 (  ,  ; , , = , > ,  ) + L 2 | |1;  (  ,  1 ; = + 1, = + 1, 2 = , > ,  ),

where  1 =  + ( = + 1)( = + 1) + > (= 1).


Table 18 lists some transformations preserving the form of Heuns equation. (Whenever at
least one of the indicated equations is integrable by quadrature with some values of parameters,
all the other equations are also integrable for those values of the parameters.)

^_

References: H. Bateman and A. Erdelyi (1955, Vol. 3), E. Kamke (1977), S. Yu. Slavyanov, W. Lay, and A. Seeger
(1955), A. Ronveaux (1995).

205. ( 3 + 2 + + ) ( 2
Particular solution: 0 =  .
206. 2(

+ ) + (3

The substitution = X
2  +

= 0.



+ '


2

+ 2

+ ( + ) = 0.

+ ( +

+ ) +

= 0.

leads to a constant coefficient linear equation:

2003 by Chapman & Hall/CRC

Page 239

240

SECOND-ORDER DIFFERENTIAL EQUATIONS

207. 2( 3 + 2 + + ) + 3(3 2 + 2 + ) + (6 + 2 + ) = 0.
This equation is obtained by differentiating the equation 2.1.2.206.

208. ( 3 + 2 + + ) + [ 
Particular solution: 0 =  + = .

+ ( | +  ) + C ] ( 

209. ( 3 + 2 + + ) + ( 3 +
Particular solution: 0 =  + .

210. 2 (

+ [ (2 )

+ )

) (

+ (1 ) ]

+  ) = 0.

) = 0.
+

 +1

= 0.

The substitution = X  J 2 (  2 + ' + ( )1 J 2  leads to a constant coefficient linear equation:


M
2  + = 0.

2.1.2-7. Equations of the form (  4 


211.

212.

( + )
=

+  1  +  0 )   +  ( )  + ( ) = 0.

leads to a constant coefficient linear equation of the

+ ( 2 + + ) = 0.
The transformation H = 1
 , = 

H 2 '  + ( (H 2 + H +  ) = 0.

leads to a linear equation of the form 2.1.2.115:

+ [( + ) + ] = 0.



L 1   J + L 2    J

( L 1  + L 2 ) J

if   ,
if  =  .

+ 2 2 ( + ) + = 0.
The substitution H = 1
 leads to a constant coefficient linear equation: ' 2   + 

Particular solution:

218.

+  2

215.

217.

+ = 0.
The transformation H = 1
 , = 

form 2.1.2.1: '  +  = 0.

Solution:

216.

+  3

213.

214.

2(

J 
.
0 =   

)2 + = 0.
Solution: = L 1 | | + |  |1 + + L
( 1)  2 =  .

2(

)2
Solution:

2 (

2|

2 )

= 0.

= 0.

 |1 + |  | + , where

is a root of the quadratic equation

)2 .

= | | + |  |1 + OL

X | |1 + |  | + 
 (2 1)
M P
(
+
+
1 +
1 +
+ | | |  | OL 2
 ,
X | | |  |
 (2 1)
M P

where is a root of the quadratic equation ( 1)  2 =  .

2 ( 1)2 + ( 2 + + ) = 0.
Let  and  be roots of the quadratic equations
D ( 1) + = 0,  (
M

1) +  + ( + = 0.
M

The substitution =  K ( 1) & leads to the hypergeometric equation of the form 2.1.2.171:
 ( 1)2   + 2  [( +  )  ]3 + (2 D ( 2 ) = 0.
M
2003 by Chapman & Hall/CRC

Page 240

241

2.1. LINEAR EQUATIONS

219.

220.

2(

+ ) + (

+ )

= 0.

The substitution =  leads to a linear equation of the form 2.1.2.194: 4 2 ( +  )  +


= 0.
2 [(  + 1) +  + ( ]  +
(

+ 1)

= 0.

The Halm equation. Solution:


=
221. (

+1 VL
2+1 VL
 2 + 1 (L






1)2 +

Solution:
=




cos( arctan  ) + L 2 sin( arctan  )W


1 cosh( arctan  ) + L 2 sinh( arctan  )W
1 + L 2 arctan  )

if  + 1 = 2 > 0,
if  + 1 = 2 < 0,
if  = 1.

= 0.

| 2 1| V L 1 cos  ln | H |  + L 2 sin  ln | H |  W
( + 1) L 1 | H |(2: 1) J 2 + L 2 | H |(2: +1) J 2 
| 2 1|  L 1 + L 2 ln | H | 

if  1 = 4 2 > 0,
if  1 = 4 2 < 0,
if  = 1,

where H = ( + 1)
( 1).
222. (

A 2 )2 +

= 0.

This is the equation of bending of a double-walled compressed bar with a parabolic crosssection.
1 b . For the upper sign (constricted bar), the solution is as follows:
= 

+

(L

cos + L

sin ),

where =

1 + (   ) arctan(  ).

2 b . For the lower sign (bar with salients), the solution is given by:
= 
223. 4(

+ 1)2

(L

+ (

cos + L

224. (

+ )2

1
2

+ 2

The substitution = X
225. (

(
(

2
2

|  1| ln   +

1)2 + 2 (



sin ),

where =

 2
2

ln

 +
; | | <  .
 

+ 3) = 0.

Solution:

where =

M2

+ 1)1 J 4 ( L
+ 1)1 J 4 ( L
|

+ )

1
1

cos + L 2 sin )
if  > 1,
cosh + L 2 sinh ) if  < 1,

+ 1|  .
+ ^ = 0.

leads to a constant coefficient linear equation:  + (

+

1) [ b ( b + 1)(

1) +

= 0.

] = 0.

Here, @ is an arbitrary number and B is a nonnegative integer. This is a special case of equation
2.1.2.226.
If B = 0, this equation coincides with the Legendre equation 2.1.2.154. Denote its general
solution by ( ). If B = 1, 2, 3,  , the general solution of the original equation is given
by the formula:

= |

1| J


M 


( ).

2003 by Chapman & Hall/CRC

Page 241

242

SECOND-ORDER DIFFERENTIAL EQUATIONS

226. (1 2 )2 2 (1 2 ) + [ b ( b + 1)(1 2 )
2 ] = 0.
The Legendre equation, @ and are arbitrary parameters.
The transformation  = 1 2 , = | 2 1| J 2 leads to the hypergeometric equation
2.1.2.171:
( 1)   + ( + 1)(1 2 )  + ( @ )( @ + + 1) = 0
with parameters = @ , = + @ + 1, = = + 1.
In particular, the original equation is integrable by quadrature if @ = or @ = 1.
In Subsection S.2.9, the Legendre equation are discussed in more detail. See also the
books by Abramowitz & Stegun (1964) and Bateman & Erdelyi (1953, Vol. 1).
227.

1)2 + ( 2 1) + ( 2 + +  ) = 0.
The transformation = 12 ( + 1), = | + 1|K | 1| & , where  and  are parameters that
are determined by solving the second-order algebraic system

4  (  1) + 2 U + ( + +  = 0,

(  )[2 ( +  1) +  ] = ,

leads to the hypergeometric equation 2.1.2.171 with respect to = ( ).

228. (

+ )2 + (2

The substitution = X



+ )2 + (

Particular solution:
230. (

+ )2

Particular solution:

231. (

+ )2 +

(
(

Particular solution:

M2

+ ) +

= 0.

+ ) + 2( ^ )
( 2 +
= exp O X
M  .
 2 + 3M P

+ )(

+ )


0 =
0

leads to a constant coefficient linear equation of the form

+
= 0.

2.1.2.11:   + (  + ,

229. (

+ )(

= (

+ .

+ )
2

+
+  ) J 2.

+1

+ ) = 0.

+1

+ (

1)

= 0.

+ ] = 0.

.
232. ( )2 ( )2 ^ = 0,
 
, = (  ) leads to a constant coefficient linear equation:
The transformation = ln
 





the solution is as follows:


(   )2 ( ) ( = 0. Therefore,
=L
where

1|

  |(1+  ) J 2 |  |(1  ) J 2 + L 2 |  |(1  ) J 2 |  |(1+  ) J 2 ,

= 4 ( (   )2 + 1 0.

233. ( )2 ( )2 + ( )( )(2 + ) +
= 0.
Let , 1 and , 2 be roots of the quadratic equation (   )2 , 2 + (   )(  +  + ) , + = 0.
Solution:
L |H |  1 + L 2 |H |  2
if , 1 , 2 ,
= 1
| H |  L 1 + L 2 ln | H |  if , 1 = , 2 = , ,
where H = (  )
(  ).

234. (

+ )2 +



= 0.

leads to a constant coefficient


, =
+ ' + (
|  2 + ' + ( |
 ( 1  2 ) = 0.
linear equation of the form 2.1.2.1: u  + ( # + !
4
The transformation = X



2003 by Chapman & Hall/CRC

Page 242

243

2.1. LINEAR EQUATIONS

235. (

1)2 + 2 (

1) + [(

1)(

2 2

] = 0.

= 0, 1,  It arises when separating


Equation for prolate spheroidal wave functions,

variables in the wave equation written in the system of prolate spheroidal coordinates.
1 b . In applications, one usually looks for eigenvalues = +  and eigenfunctions = +  ( )
that assume finite values at  = A 1. The following functions are solutions of the eigenvalue
problem:

(1)
+  ( ,  ) =

(2)
+  ( ,  ) =

+  ( ). +

M
=0,1

+  ( ) +

+
M

(
+

(

(prolate angular functions of the first kind),


) (prolate angular functions of the second kind),

where  + ( ) and + ( ) are the associated Legendre


. functions of the first and second kind.
.
For 1  1, we have  + ( ) = (1  2 ) + J 2 x x   ( ). The summation is performed over
either even or odd values of & , depending on whether |B | is even or odd, respectively.

2 b . The following recurrence relations for the coefficients  = +  (  ) hold:


M
M

 
M
where

3 b . For eD

+2

+ (  +  )  + = 

M 

= 0,

 2 (2 + , + 1)(2 + , + 2)
,
 =

(2 + 2 , + 3)(2 + 2 , + 5)

2( + , )( + , + 1) 2 2 1
,
 = ( + , )( + , + 1) +  2

(2 + 2 , 1)(2 + 2 , + 3)

 2 , ( , 1)
.
=  =
(2 + 2 , 3)(2 + 2 , 1)

0, the eigenvalues are defined by:


1

1)(2

(2

+ 1)

+  = B (B + 1) + 1

S 2 + [   4  .
2Q
(2B 1)(2B + 3)
4 b . For eDGF , we have:

+  =  +

^_
236. (

18 ( 

+ 5)

1
64

 ( 2 + 11 32

)

+[  

,

= 2(B

) + 1.

References: H. Bateman and A. Erdelyi (1955, Vol. 3), M. Abramowitz and I. A. Stegun (1964).
2

+ 1)2 + 2 (

+ 1) + [(

+ 1)(

2 2

)+

] = 0.

Equation
of oblate spheroidal wave functions, = 0, 1,  The transformations  = A"  ,
 = T  lead to equation 2.1.2.235.
See the books by Bateman & Erdelyi (1955, Vol. 3) and Abramowitz & Stegun (1964)
for more information on this equation.
237. (

+ )2

+ (2

+ )(

+ )

+ x = 0.


leads to a constant coefficient linear equation of the
M
2 + ' + (



= 0.
form 2.1.2.11:   + ( ,  )  +

The substitution = X

2003 by Chapman & Hall/CRC

Page 243

244

SECOND-ORDER DIFFERENTIAL EQUATIONS

2.1.2-8. Other equations.


238.

239.

= 0.

The transformation =  2 , = 
form 2.1.2.1: 43  +  = 0.

+ (3

+ )

The substitution = 
240.

+
where

73


(1 

=1

leads to a constant coefficient linear equation of the

= 0.

leads to a constant coefficient linear equation: 4 2   + 

( 1

1 )( 2

R R
  1
3 ) =1

2 )( 3

(  +  ) = 1, |   | + |   | > 0,  =    

=1

+1

 

= 0.

+1

  0,  

+3

=  ,  

+3

= 0,

=  .

Riemanns equation. Denote this equation by:

2
2

3
3

= 0.

(1)

For  1 =  2 = 0,  3 =  3 = 1, 1 = 3 = 0, 2 = , 1 = 1 = , 2 = , and
equation (1) transforms into the hypergeometric equation 2.1.2.171.

The transformation

$L

3
3

1
1

+ &
+&

2
2

where #  = #2  + $  , $  = Lu  + c  .


In (2), assume & = 1 , = 3 , # =  1

to obtain the hypergeometric equation (3).

+ )

+ (

242.


,
 + 

 = 0.

Particular solution:
243.

+ (

+ (

+ (2

+ 2
= 

$ =  1

= 0,

(3)

3,

L =  2

2,

and c

=  2

 + 

leads to an equation of the form 2.1.2.7:

+ ) = 0.

Particular solution:

=  +  .

Particular solution:
245.

3,

3
3

+ ) = 0.

Particular solution:
244.

&
&

= 0.

The transformation =

2  + (

(2)

0, brings the original equation into an equation of similar form:

241.

== ,

= | 1   1| | 3  3| ,
|  2   2 | +

#  +$
*
=
,
L3 + c
where #2c

=

2
0

+ ( 1) = 0.

+
= +
 .

= 0.

2003 by Chapman & Hall/CRC

Page 244

245

2.1. LINEAR EQUATIONS

246.
247.
248.

+ ( + )
Particular solution:

+ (
Particular solution:

+ (

+ )

+ + ) +

.
0 = (  + 1) 

+ 1)
0

+ (

+ ] = 0.

Particular solution:
249. (

+ [( )

=  v .

= exp O
+ )

Particular solution:

=  .

(1 +

+1

 +

+1

= 0.
) = 0.

+ [(  )

 ]

= 0.

250. ( + + ) = ( 1) 2 .
Particular solution: 0 =   + ' + ( .
251.

+ 1) + [( ) + ] + (1 )

Particular solution:
252.

( 2

+ ) + (

254.

2 2 1

+ )

J 
= L 1    +  2 +    + L

Solution:
253.

 + 1)  J  .
0 = (

= 0.

= 0.

 +  2  +    J  .
2 

2 ( 2 2

1) + [ 2 ( + 1) 2 + 1] b ( b + 1) 2 2 2 = 0.
= (   ), where ( ) is the general solution of the Legendre equation
Solution:
2.1.2.154.

2 (

1) + ( ' + ) + ( / + s) = 0.
Find the roots # 1 , # 2 and $ 1 , $ 2 of the quadratic equations
2

(  + 1) # = 0,

( 1) $ + & = 0

and define parameters ( , , , and = by the relations

= = 1 + ( # 1 # 2 )B 1 .
Then the solution of the original equation has the form =  v (   ), where = ( H ) is the
general solution of the hypergeometric equation 2.1.2.171: H ( H 1) '  + [( + + 1) H = ]  +
= 0.
( = # 1,

255. ( + )2

= (#

Particular solution:
256. (

+ )2 + (

Particular solution:
257. (

+ )2

+ )2 +

1)

= (#

+$

2)

+ ) + ( ^ )
(  +
= exp O X
M  .
  + 3M P
+ )(

+ )

 +  )1 J .
0 = (

= |  +  |  J  .

Particular solution:

+$

[( 1) + ( 1)] = 0.

Particular solution:
258. (

+ ) + (

= exp O X

+1

= 0.

+ ) = 0.

1) = 0.

.
M
  +  P

2003 by Chapman & Hall/CRC

Page 245

246
259.

SECOND-ORDER DIFFERENTIAL EQUATIONS

2 (

+ )2 + ( + 1) ( 2 2 2 ) + ^ = 0.
1
  
ln O
leads to a constant coefficient linear equation of the
The substitution =
B
   +  P


form 2.1.2.11:  (B + 2) + ( = 0.

260. (

+1

+ )

+ 

+( 

Particular solution:

+ ) +[ (  )w

= exp X

( B +  )  + ( 'B ) 
  +1 + '  + (

+( 1)( ) w

261. ( + + ) + ( ) + = 0.
Particular solution: 0 =  .

] = 0.

 S .

262. ( + + ) + ( 2 2 ) + ( + ) = 0.
Particular solution: 0 =  .

263. 2(

+ )

264. (

= 0.

+ )

+1

265.

+ (

Particular solution:


  +M ' + + (

The substitution = X
2  +

+ (

+ )

+ I = 0.

leads to a constant coefficient linear equation:

= 0.

S .
M
(   +  ) +

= exp X

) + ( 2 + ) i 2 ( )] + i 2 ( ) = 0.
 2 ( ) are arbitrary polynomials of degrees B and B 2, respectively.

Particular solution: 0 =  + 
 .

g ( ) + [2 g (
. ( ) and
Here,

2.1.3. Equations Containing Exponential Functions


2.1.3-1. Equations with exponential functions.
1.

2.

3.

+ 
Solution:
functions.

= 0,
=L

0.
0 ( H ) + L 2 n 0 ( H ), where H = 2

+ (  ) = 0.

= L 1 22  2   J 2  + L
Solution:

Bessel functions.

+ ( 



Particular solution:

= 0.

= exp O  

n 2

2 2

   J 2 ;

0(

H ) and n 0 ( H ) are the Bessel


 2   J 2  , where

( H ) and n ( H ) are the

4.

[ 2  2 + (2 + 1) + 2 ] = 0.

Particular solution: 0 = exp(   + '  ).

5.

(  2  +  + ) = 0.

The transformation H =  , = H  , where , = (
, leads to an equation of the form
'



+ 2 (2 , + 1)2 ( !H +  ) = 0.
2.1.2.108: 2 H

Page 246

247

2.1. LINEAR EQUATIONS

6.

7.

+ ( 

 3

+ 

9.

10.

1
4

) = 0.


The transformation =  , =   J
2   + 2 (  2 + ' + ( ) = 0.

+ [ 

(  

+ )

The transformation =  
2  +  (  )2  = 0.
8.

 2D

= 0.

 2D

= 0.

1
4

leads to a linear equation of the form 2.1.2.6:

] = 0.


=  J

+(,

leads to an equation of the form 2.1.2.7:



The transformation =  , =  leads to a constant coefficient linear equation of the


2
form 2.1.2.1: +  = 0.

The substitution =  leads to a constant coefficient linear equation of the form 2.1.2.1:
 +  2 = 0.

+ ( 

+ ) = 0.



n  2
Solution: =  J 2 V L 1 / 2 1    J 2  + L 2
(H ) and n (H ) are the Bessel functions.

11.

+ 

 3

 2

The substitution H = 
H 2 ' +  !H 3  + H 2  +
12.

+ V 

+ 2 

1
4

    J 2  W , where @ =

 2 4( ;

= 0.

leads to a second-order linear equation of the form 2.1.2.121:


14 2  = 0.

1
4

(  

1
4

+ ) +

1
4

1
4

W = 0.

leads to a second-order linear equation of the form 2.1.2.122:


The substitution H = 
H 2 ' + V !H 2  ( H  + ( )  + 14 14 2 W = 0.
13.

+ ( + ) 

 

+ 2 

2   + (   2 

= exp O  

+ ( 

17.

+  

( + 2 D

Particular solution:

+L

= exp O
2

exp O

+ ( ) = 0.

+
0

2  2D

= exp O



+
) = 0.


  .
P

 
,

 ).

+ ) = 0.

 (  

The substitution

(L

 (  

Particular solution:
16.

+ ) = 0.

Particular solution:
15.

= exp O  

Solution:
14.

  (  

2  2


 


C

+ ) = 0.

leads to a linear equation of the form 2.1.3.5:

= 0.

 
.

P

2003 by Chapman & Hall/CRC

Page 247

248
18.

SECOND-ORDER DIFFERENTIAL EQUATIONS

+ ( 

+ )

Particular solution:
19.


0

=  

= 0.

+   .

+ (  ) +  2  = 0.

The substitution =  leads to a constant coefficient linear equation: 2  +   + 

20.

+ (  + ) + (  +

Particular solution: 0 =  v .

21.

 2 )

+ ( +

+ (

Particular solution:
22.

+ (

=  

+   .

( )

= (  

) = 0.



3 ) +

Particular solution:

= 0.
2

= 0.

+ 1) exp(    ).

23.

+ (2  ) + ( 2  2  +   ) = 0.
This is a special case of equation 2.1.3.28 with  = , = 0.

24.

+ (2 

+ )

+ [

 

2 2

= 0.

+ ( + )

+ ] = 0.

 
leads to a constant coefficient linear equation of the form
The substitution = exp O  

P
2



2.1.2.11:  +   + ( = 0.

25.

+ (  + 2 ) + (  2 

The transformation =  
,
2  +  2  + ( = 0.

26.

+ (  + ) + [ ( )  2 + ( +

Particular solution: 0 = exp( (  ,! ).

27.

+ (  + ) + ( q 2  + J + ) = 0.



The substitution =  leads to an equation of the form 2.1.2.146: 2  + (   +  + 1)  +
( 2  +  + = ) = 0.

28.

30.

+ (2 

) + (

 

2 2

+ 2 ) 

+ ( )] = 0.

 2

 

+ ) = 0.



leads to a linear equation of the form 2.1.3.5:
 

2   +    2   + (    + , 1 2  = 0. 2 P
4

+ (2 

+ (

exp O

  )

+ (

+   +  2 +  + ) = 0.
 

 leads to an equation of the form 2.1.3.5:
The substitution = exp O   +

2   + V (  2   +    + , 1 (  )2 W = 20. P
4

Particular solution:
31.

The substitution

29.

+  + 2  ) = 0.

=  
leads to a constant coefficient linear equation:

+ 

 

2 2

(   


= exp O  

+ ) = 0.
.

+   (  2 + ) +
[   ( 


Particular solution: 0 =    +    .

)
] = 0.

2003 by Chapman & Hall/CRC

Page 248

249

2.1. LINEAR EQUATIONS

32.

+ ( 



Particular solution:
33.

+ ( 

 

+ ( 

+ 2 

+ ( 

= exp O  

+ ) + [ 

 

( + )


C )

) + [ 

 

= 0.


  (

+ 

  ( .
0 = exp O 

P


Particular solution:
34.

+ )

( + )


C  ]

2  2

 2

= 0.

+ (
)

= 0.

1 b . If = 0, the equation transforms into 2.1.3.24, and if = 0, into 2.1.3.25.


1 
 
leads to a constant coefficient
2 b . For 0, the transformation =   , = exp O  
linear equation: 3  +  2  + ( = 0.

35.

+ [ 

 

( + )



Particular solution:
36.

v
0 = 

( 

+ ) 

Particular solution:
38.

 

2 2

2( 

+ )

=L

  

2   (2 + )

= 0.

) ] = 0.

= 0.

=  

2 ] +

+ ) 

Solution:
39.



+ 1) exp(    ).

+ [ exp(

Particular solution:
37.

= (  

exp(

 

+ .

 

2 2 2

= 0.



+  2  2 +   + L

 

  

+  2  2

+ 

= 0.

The substitution = X (    +  )1 J 2  leads to a constant coefficient linear equation of the


M
form 2.1.2.1: 2   + ( = 0.

40.

( 

+ )

+ ( 

( 

42.

(

+ )

+ ( 

+ [( ) 

+ )

+ (J

 .
0 = 

Particular solution:
41.

+ )

= 0.

For the case  = 0, see equation 2.1.3.27. For  0, the transformation =  ! , =  ,
where , is a root of the quadratic equation  2 , 2 + ", + = 0, leads to an equation of the form

M
2.1.2.172:  2 ( +  )2  + [(2  , +  + ( ) +  (2 , +  + )]3  +(  , 2 2 + (, + B ) = 0.

+ ) + ( 

+  

( 

+ )2 + 

Particular solution:
44.

) = 0.

+ )
+ (

Integrating yields a first-order linear equation: ( 


43.

( 

= 0.




+ , )  + (   +     + ( ) = L .

= 0.

+ ) + 

= (  

(
+  )  , where , = .


= (  

+ )2 +  ( 

Particular solution:



+ 
C 

(



(
+  )  , where , = .


= 0.

2003 by Chapman & Hall/CRC

Page 249

250
45.

SECOND-ORDER DIFFERENTIAL EQUATIONS


( 

+ )2

+ ( 

The substitution = X
2.1.2.11:   + (  +

46.

(  

+ )

4( 

+ )

M

  

+ x = 0.

+ 

+ )

(  



(
+  )  , where , = .


( 

+ )

+ )

= (  

leads to a constant coefficient linear equation of the form

+
= 0.

+ [ 

+ )( 

 (  

Particular solution:
47.

( 

= 0.

+ ) ] = 0.


  + 
  J 2
,
=
leads to an equation of the form 2.1.2.7:
The transformation =


(  +
 +
42  + , ( e )2  = 0,(  where
M =  ( . M
M

2.1.3-2. Equations with power and exponential functions.


48.





+ (

Particular solution:
49.

+ 2 

+ (

The substitution

2   + ( ' 2  + ( 

+ ( + ) 
Particular solution:

51.

+ (



53.

55.

+ ( + ) exp(
Particular solution:

56.

(2

+1

( 
= .

= exp O

w )

B +1

 

= .

+1

exp(2

57.

 

Particular solution:

2 )

 
0

) = 0.

(1 +

)


=  exp O  

2 )

= 0.

= 0.
1

) = 0.

Solution: = exp    L 1 ( H ) + L


2
Q
n ( H ) are the Bessel functions. 
1
2

exp(

  )

exp(
=  +  .





+ (

leads to a linear equation of the form 2.1.2.10:



= 

+ 1) + 4

) = 0.

exp( )

Particular solution:

 

 

 = 0.
0 =  +  .

+ ( +  )

Particular solution:
54.

exp O

+ 2 ) + (

+ (  + 
Particular solution:

B +1

+ 

) = 0.

Particular solution:
52.

 

2 2

50.

= exp O

= 0.
= 0.

= 0.

n ( H )S , where H =
2


 exp   2  ; ( H ) and

= 0.

2003 by Chapman & Hall/CRC

Page 250

251

2.1. LINEAR EQUATIONS

58.

 

[ (

Particular solution:
59.



+ (

+ [ (

62.
63.

65.
66.

+ ( +  
Particular solution:

+ ( 
 

70.

+ [( + 1) 
Particular solution:

73.

= 0.



= 0.

+ [ ( 1) 
=  exp( ' 
B ).

+ + 1 2 ] +
= (   + 1) exp(   ).

] = 0.

2 2 1  

= 0.

+ (   + 
C ) = 0.


Integrating, we obtain a first-order linear equation:  " + (   +    1) = L .

+ [

( + )

+ [

exp(

+ ( 

) + ] + ( 1)

v.
0 = 

+ )
0

exp(

 

= 0.

1 (  
 +

= exp O X

)+1

exp(

= 0.

 .
M P

] = 0.

= 

1
2

leads to a linear equation of the form 2.1.3.1:

+ 2 + [( 2  2 } b 2 ) 2 2 + ( 1)] = 0.


Solution: =  V L 1 (   v )+ L 2 n (   v )W , where ( H ) and n ( H ) are the Bessel functions.

2
2

 

+ ( 

+ ( 
0

=

.

+ 2 ) + [ (

Particular solution:
72.



+ 1 (   + 1) = 0.
= exp(  
B ).

+  )

Particular solution:
71.

= '  ,
The transformation
42  +  ( 'B )2  = 0.
69.

+ 2)] = 0.

= 0.

1] +

= ( ' + 1) 

+ )

+ ( 

.

+ [( 2 + )  + 2] +
Particular solution: 0 =  + 
 .

Particular solution:
68.

+ 1)-

Particular solution:

67.

=

Particular solution:

64.

+ (

+ ) + ( 1) 

Particular solution:
61.

+ 1)

= 

Particular solution:
60.

1) = 0.

v 
.
0 =   

+ (  2 ) b 2 ) = 0.

Solution: =  V L 1 (  1 J ) + L

+ ) 

+ ( 1)] = 0.


n (  1 J )W , where ( H ) and n ( H ) are the Bessel functions.

+ [ exp(2 # ) + exp( # ) + ] = 0.
= 1
 , = 
 leads to a linear equation of the form 2.1.3.5:
The transformation
2  + (   2  +   + ( ) = 0.

2003 by Chapman & Hall/CRC

Page 251

252
74.
75.

SECOND-ORDER DIFFERENTIAL EQUATIONS

+ 2   + [ ( ) 
Particular solution: 0 =  exp( 
 ).

 (

+ )2 +

( 

+ )2 + (

+ )

79.

= exp O X
2

 


=  

+ ] 

Particular solution:

+ ) 

Particular solution:

 
[( + ) -

 +  )1 J .
0 = (

Particular solution:
78.

( + )2 + ( + ) 
Particular solution:

77.

+ .

=(  

= 0.



+ ) (


0 =

Particular solution:

76.

2 ]

+ ) = 0.



+ ( 

) = 0.
1) = 0.

.
M
  +  P
= 0.

+ ' + ( .
= 0.

+  +  .

2.1.4. Equations Containing Hyperbolic Functions


2.1.4-1. Equations with hyperbolic sine.
1.

+ ( sinh2 + ) = 0.

Applying the formula sinh2  = 12 cosh 2 12 , we obtain the modified Mathieu equation
2.1.4.9:    +   12  + 12  cosh 2  = 0.
2.

sinh(

Particular solution:
3.

+ [ sinh (

5.

0 =  

) ] = 0.

) + ] + sinh (
0

=

+ ( + ) sinh ( ) sinh (
Particular solution: 0 =  +  .

+ ( sinh

Particular solution:
6.

Particular solution:
4.

+ [ sinh(

sinh2 (

+1

= exp O

= 0.

+1

 +

= 0.

= 0.

sinh
+1

) = 0.

( H > 0) leads to a linear equation of the form 2.1.2.190:


The substitution  = A ln
H 2+1
'




2
2
2
H ( H + 1) + (3 H + 1) 4  H = 0.
7.

sinh2 [

Solution:
8.

sinh2 + ( 1)] = 0,


1
M
= sinh   O
(L 1  + L
sinh   P

0;


 ).

= 1, 2, 3,

uuu

[ sinh( ) + + ]  2 sinh( ) = 0.
Particular solution: 0 =  sinh(  ) + ' + ( .

2003 by Chapman & Hall/CRC

Page 252

253

2.1. LINEAR EQUATIONS

2.1.4-2. Equations with hyperbolic cosine.


9.

( 2 cosh 2 ) = 0.
The modified Mathieu equation. The substitution  = leads to the Mathieu equation
2.1.6.29:
  + (  2  cos 2 ) = 0.

For eigenvalues  =   (  ) and  =   (  ), the corresponding solutions of the modified Mathieu


equation are:

# 22  ++K cosh[(2 , +  ) ],
K
 =0

Se2  + ( ,  ) = se2  + ( ,  ) = $ 22  ++K sinh[(2 , +  ) ],


K
K
K
 =0
where  can be either 0 or 1, and the coefficients # 22  ++K and $ 22  ++K are specified in 2.1.6.29.
K
K
The modified Mathieu equation is discussed in the books by Abramowitz & Stegun (1964),
Ce2 

( ,  ) = ce2 

K ( ,  ) =

Bateman & Erdelyi (1955, vol. 3), and McLachlan (1947) in more detail.

10.

+ ( cosh2 + ) = 0.

Applying the formula cosh 2 = 2 cosh2  1, we obtain the modified Mathieu equation
2.1.4.9:    +  12  +  + 12  cosh 2  = 0.
11.

cosh(

+ [ cosh(

Particular solution:
12.

+ [ cosh (

14.
15.

18.

+ cosh (
Particular solution:

cosh (

+ [

cosh (

= .

[ (

= 

= 0.

cosh (

= 0.

+ 1) cosh (

= 0.

) + (

) + ] + ( 1) cosh (
=

cosh (

.

+ [ cosh (

+ 2)] = 0.

+ [( 2 + ) cosh ( ) + 2] + cosh (
Particular solution: 0 =  + 
 .

Particular solution:
19.

+ ( + ) cosh ( ) cosh (
Particular solution: 0 =  +  .

Particular solution:
17.

) ] = 0.

Particular solution:
16.

=

) + ] + cos (

Particular solution:
13.

=

.

=

=   .

)
)

= 0.

= 0.

1] = 0.

( cosh + ) + ( cosh + ) + [(  ) cosh +  ] = 0.


Particular solution:

2003 by Chapman & Hall/CRC

Page 253

254

SECOND-ORDER DIFFERENTIAL EQUATIONS

20.

cosh2 (

21.

[ cosh( ) + + ]  2 cosh( ) = 0.
Particular solution: 0 =  cosh(  ) + ' + ( .

= 0.
1
H
ln
(0 < H < 1) leads to the hypergeometric equation 2.1.2.171:
The substitution  =
2

1

H
'




H ( H 1) + (2 H 1) +  2  = 0.

2.1.4-3. Equations with hyperbolic tangent.

22.

+ [ tanh(

) + ] = 0.
1 tanh(  )
,
= H J  , where , is a root of the quadratic
The transformation H =
1 + tanh(  )
equation 4 , 2 +   = 0, leads to a linear equation of the form 2.1.2.172: 4 2 H ( H + 1)2'  +

4 (2 , + )(H + 1)3 + (4 , 2 +  +  ) = 0.

23.

24.

tanh2 (3

Particular solution:

= sinh(3  )[cosh(3  )]1 J 3 .

)]

+ [  ( + ) tanh2 (

Particular solution:
25.

= 0.

+ [3 

= 0.

= [cosh(  )] J  .

( + ) tanh2 (

Particular solution:

)]

= sinh(  )[cosh(  )] J  .

26.

+ [ + tanh( )] = 0.
Particular solution: 0 = cosh(  ).

27.

+ 2 tanh

Solution:
28.

30.
31.

32.

+ = 0.

cosh  =

tanh(

Particular solution:
29.

= 0.

cos( ' ) + L 2 sin( ' )


if  1 =  2 > 0,
L 1 cosh( ' ) + L 2 sinh( ' ) if  1 =  2 < 0.
1

+ [ tanh(
0

=

) ] = 0.

+ 2 tanh + ( 2 + + ) = 0.
The substitution = cosh  leads to a second-order linear equation of the form 2.1.2.6:
2   + (  2 + ' + ( 1) = 0.

+ 2 tanh + ( + 1) = 0.
   +   = 0.
The substitution = cosh  leads to a linear equation of the form 2.1.2.7: I

+ 2 tanh + ( 2 + 1 + 1) = 0.
The substitution = cosh  leads to a second-order linear equation of the form 2.1.2.10:
2   + (  2  + '  1 ) = 0.

+ (2 tanh + ) + ( tanh + ) = 0.
 +
   +  2
The substitution = cosh  leads to a constant coefficient linear equation: I

(  1) = 0.

2003 by Chapman & Hall/CRC

Page 254

255

2.1. LINEAR EQUATIONS

33.

+ tanh ( ) [ + tanh
Particular solution: 0 = cosh(  ).

34.

+ [ tanh (

) + ] +

Particular solution:
35.
36.
37.

40.

+ tanh (
Particular solution:

tanh (

+ [

tanh (

+ ( tanh

=

tanh (

= 0.

+ 1) tanh (

= 0.

) + (

+ ( 1) tanh (

.

+1

= exp O

+1

 +

.

=

=   .

tanh

+ [ tanh (

+ 2)] = 0.

) + ]

tanh (

+1

= 0.

= 0.

) = 0.

.
) 1] = 0.

( tanh + ) + ( tanh + ) + [(  ) tanh +  ] = 0.


Particular solution:

43.

[ (

= 0.

= .
= 

= 0.

+ [( 2 + ) tanh ( ) + 2] + tanh (
Particular solution: 0 =  + 
 .

Particular solution:
42.

Particular solution:
41.

+ ( + ) tanh ( ) tanh (
Particular solution: 0 =  +  .

Particular solution:
39.

tan (

)]

Particular solution:
38.

=

+1

) + ] + [ tanh( ) + ] + [ tanh( ) + ] = 0.
1 + tanh(  )
, = H J  , where , is a root of the quadratic equation
The transformation H =
1 tanh(  )
4(   ) , 2 + 2( ( ) , + B = 0, leads to a linear equation of the form 2.1.2.172:

[ tanh(

4 2 H [(  +  ) H +   ] '  + 2 {[2(2 , + )(  +  ) + ( + ] H + 2(2 , + )(   ) + ( } 


M
M
+ [4(  +  ) , 2 + 2( ( + ) , + B + ] = 0.

2.1.4-4. Equations with hyperbolic cotangent.


44.

45.

+ [ coth(

) + ] = 0.
1 tanh(
The transformation H =
1 + tanh(
equation 4 , 2 +   = 0, leads to
4 (2 , + )(H 1)3 + (4 , 2 +  +  )
coth2 (3

Particular solution:

 )
,
 )

= H  J  , where , is a root of the quadratic

'

an equation of the form 2.1.2.172: 4 2 H ( H 1)2  +


= 0.

= 0.

= cosh(3 )[sinh(3  )]1 J 3 .

2003 by Chapman & Hall/CRC

Page 255

256
46.

SECOND-ORDER DIFFERENTIAL EQUATIONS

Particular solution:
47.
48.

)]

+ [  ( + ) coth2 (
+ [3 

= [sinh(  )] J  .

( + ) coth2 (

+ [ coth(

coth(

Particular solution:

= 0.

=

) ] = 0.

coth (

= 0.

50.

+ ( + ) coth ( ) coth (
Particular solution: 0 =  +  .

= 0.

51.

49.

+ [ coth (

)]

= cosh(  )[sinh(  )] J  .

Particular solution:
+

= 0.

) + ] +

Particular solution:

+ 2 coth

Solution:
52.

=

+ (

1
= O
sinh 

M
M

)]

2 ) = 0,

3P

uuu

= 1, 2, 3,



( L 1  + L 2  ).

+ [ + coth( )] + [ + coth( )] = 0.
[ + coth(
Multiply this equation by tanh(  ) to obtain equation 2.1.4.43.

2.1.4-5. Equations containing combinations of hyperbolic functions.


53.

[ cosh2 (

) + sinh(

Particular solution:
54.

[ sinh2 (

Q 

) + cosh(

Particular solution:
55.

= exp

= exp

Q 

)]

= 0.

sinh( ' )S .

)]

= 0.

cosh( ' )S .

+ ( cosh2 + sinh2 + ) = 0.

Apply the formulas 2 sinh2  = cosh(2 ) 1 and 2 cosh2  = cosh(2 ) + 1 to obtain an


 
 +
+( +
cosh(2 )S = 0.
equation of the form 2.1.4.9:    +
Q 2
2
56.

+ sinh( ) [ + cosh(
Particular solution: 0 = sinh(  ).

57.

+ cosh( )
Particular solution:

58.

tanh(

Solution:
59.

tanh

Solution:

=L

[ + sinh(
0 = cosh(  ).

)
1

exp

= sinh 

)]

= 0.

)]

= 0.

cosh2 (

sinh(  )S + L

= 0.
2

coth2 (sinh )2

n ( H ) are the Bessel functions.

exp

+
sinh 

sinh(  )S .

= 0.
+L

sinh 

S , where ( H ) and

2003 by Chapman & Hall/CRC

Page 256

257

2.1. LINEAR EQUATIONS

60.

61.

sinh (

+ [ cosh

2  + 

cosh (

2  + 

2 

= 0.

The transformation = tanh(  ),

 = 0.

+ [ sinh

sinh (

cosh (

The transformation = coth(  ),

= 0.

leads to an equation of the form 2.1.2.7:

cosh(  )

)]
)]

= 0.

leads to an equation of the form 2.1.2.7:

sinh(  )

2.1.5. Equations Containing Logarithmic Functions


2.1.5-1. Equations of the form  ( )  + ( ) = 0.
1.

ln2 + ln + ) = 0.

Particular solution:
2.

ln2 +

Particular solution:
3.

7.

8.

10.

ln +

=   (

ln2 (

) + ln
0

+1)

)]

= 0.

+ ( ln2 + ln + ) = 0.

The transformation = ln  ,
2  + (  2 + ' + ( 1 ) = 0.
4

+ V ( ln + ) + 41 UW = 0.

The transformation =  ln  + ( ,
2  +   2  = 0.

ln(

ln

) = 0.

 +1
, where  =  
.
(B + 1)2

+ ) = 0.
The transformation =  ln  +  14 ,
2  +  2 = 0.

= exp lX ln  ( ' )  S .

+ ( ln

Solution:
9.

( 2 ln2 + ) = 0.
 
Particular solution: 0 =   .

Particular solution:

6.

4.

5.

2
2
=  J 4  J 2.

= 

1 2

1 2

= 

leads to an equation of the form 2.1.2.2:

leads to an equation of the form 2.1.2.6:

1 2

leads to an equation of the form 2.1.2.7:

+ = 0.
=L

ln(  ) + L

+ )
Particular solution:
ln + + )
Particular solution:

2(

ln(  ) X [ln(  )]2  .

= 0.
=  ln  + ' + ( .
+ = 0.
=  ln  + '  + ( .

2003 by Chapman & Hall/CRC

Page 257

258

SECOND-ORDER DIFFERENTIAL EQUATIONS

2.1.5-2. Equations of the form  ( )  + ( )  + % ( ) = 0.


11.
12.
13.
14.
15.

+ ln ( ) + [ ln (

Particular solution: 0 =  v .

17.
18.
19.

+ [ ln ( ) + ] + ln (

Particular solution: 0 =  v .
+ ( + ) ln (
Particular solution:

+ ln ( )
Particular solution: 0 =  .

ln

22.
23.

25.
26.

= 0.

= 0.

ln (

= 0.


= 

 .

+ ( ln + ) + ( ln + ) = 0.


Particular solution: 0 =   .

+ (2 ln + 1) + (


=   (L 1 + L
Solution:

ln2 + ln + ) = 0.
2 ln  ).
2

+ ln ( + ) + ( ln2 + 1) = 0.


Particular solution: 0 =   .

ln (

ln

+ ln

+ ( ln + 1) ln
Particular solution: 0 = ln  .

ln

=

.

 .
M P

= 0.

= 0.

+ [( 2 + ) ln ( ) + 2] + ln (
Particular solution: 0 =  + 
 .

+ ( + ln
Particular solution:

+ 1 ( ln
= exp(  
B ).

= 0.

= 0.

+ ) + ( 1) ln
0

= 0.

=  exp O lX ln 

+ ( ln + 1)
Particular solution: 0 = ln  .

+ (

+ ln

= exp lX ln  ( ' )  S .

+ ( ln
0

Particular solution:

24.

+ (ln + 1) = 0.

Particular solution:
21.

ln (
=  +  .

Particular solution:
20.

) ] = 0.

Particular solution:

16.

+ 1) = 0.

+ ( + ln ) + [ ( 1) ln
Particular solution: 0 =  exp(  
B ).

= 0.

] = 0.

2003 by Chapman & Hall/CRC

Page 258

259

2.1. LINEAR EQUATIONS

27.

Solution:
where
28.

29.

30.

31.

32.

+ ln (

ln( ' ) L

ln ( ' ) + L

+ (2 ln + 1) + (

ln ( ' ) ,

+ (2 ln + + 1) + (ln2 +

+ (2 ln + ) + [ln2 + ( 1) ln +

1
(B + 2),
2

) = 0.

ln2 + ) = 0.

 +
  +  2
The substitution = exp( 21  ln2  ) leads to the Bessel equation 2.1.2.126:  2 2

( 2 +   ) = 0.

ln + ) = 0.

The transformation = ln  , = exp( 21 ln2  ) leads to a constant coefficient linear equation:


2  +  2  + (  1) = 0.

+ ] = 0.

The substitution = exp( 12 ln2  ) leads to a linear equation of the form 2.1.2.132:
 2 2   +  2 + ( '  + ( 1) = 0.
+ ln ( ) + [ ln (
Particular solution: 0 =  v .

2
2

+ (

+ ln ) + (

+ ) + ( ln + ) +

+ 2 ln (
Particular solution:

) 1] = 0.
ln ln + 1) = 0.

= exp  12  ln2   .

Particular solution:

= 0.
 ln  + ( 1
 .
 +
M P

= exp O X

+ [ ( ) ln (
=  exp( (
 ).

=   .

) 2 ] = 0.

( ln + ) + ( ln + ) + [(  ) ln +  ] = 0.
Particular solution:

ln

37.

Solution:

ln(

38.

Solution:

ln2

(ln )2

+1

= 0.

= L 1  + L 2  , where  = X ln 
[ ln(

)+

2 2

+1

ln2

+1

M
(

 .

= 0.

+
= L 1   + L 2   , where  = X   ln (  )  .
M
+ (

+ 1) ln

The substitution = X
40.

34.

39.

2+
functions.
( H ) and n ( H ) are the Bessel

Particular solution:

36.

= 0.

+ + ( ln2 + ln 1 ) = 0.
The substitution = ln  leads to an equation of the form 2.1.2.10:  + (  2  + ' 

33.

35.

ln ( ) + ( 2
Particular solution:

ln 

= 0.

leads to a constant coefficient linear equation:  +   + 

= 0.

) + ( + ) = 0.
0 =   .
2

2003 by Chapman & Hall/CRC

Page 259

260

SECOND-ORDER DIFFERENTIAL EQUATIONS

2.1.6. Equations Containing Trigonometric Functions


2.1.6-1. Equations with sine.
1.

+ 2 = sin( ).
Equation of forced oscillations.


Solution:

2.

3.

4.



6.
7.

sin(

+ [
0

sin(

= exp O

 

B +1

+ sin ( ) + [ sin (

Particular solution: 0 =  v .

2
+1

= 0.

+ ( + ) sin ( ) sin (
Particular solution: 0 =  +  .

= 0.

sin (

Particular solution:

+ [

sin (

+

= exp O

sin (

, +1

 

+1

= 0.

+ (

+1

+ sin' ) +

Particular solution:

= exp O


 
B +1

+ sin (
Particular solution:

)
0

[ (

=  v .

 1 ]

= 0.

= 0.

sin' + ) = 0.

+1

+ ( + sin ) + [ ( 1) sin
Particular solution: 0 =  exp(  
B ).

2

+ [( 2 + ) sin ( ) + 2] + sin (
Particular solution: 0 =  + 
 .

] = 0.

) ] = 0.

+ [ sin ( ) + ] + sin (

Particular solution: 0 =  v .

13.

, we obtain the Mathieu equation 2.1.6.29:

+ ( sin2 + ) = 0.
Applying the formula 2 sin2  = 1 cos 2 , we obtain the Mathieu equation 2.1.6.29:
  + ( 1  +  1  cos 2 ) = 0.
2
2

9.

12.

1
2

+ sin ( )
Particular solution: 0 =  .

11.

cos(  ) +

+ [ sin( ) + ] = 0.
Applying the substitution  = 2 +
 + (4  2 cos 2 + 4  2 ) = 0.

8.

10.

Particular solution:
5.

sin(  ) + L

sin(  ) if  ,
 2

L 1 sin(  ) + L 2 cos(  )  cos(  )
if  = .
2

+ 1)w

sin (

)+

] = 0.

+ 2 ] = 0.

2003 by Chapman & Hall/CRC

Page 260

261

2.1. LINEAR EQUATIONS

14.

+ [

sin (

+ ( sin

+ ( sin

19.

+ [ sin( #

sin (

)
0

sin2

.

+ [

) + ] = 0.

sin(2 ) + 2
=L

sin (

= 0.

) = 0.

.

The transformation = 1
 ,
2  + [  sin( ) +  ] = 0.

Solution:
21.

v.
0 = 

Particular solution:
20.

=

sin (

Particular solution:
18.

=

+ ) + ( sin

Particular solution:
17.

v.
0 = 

+ 1) + ( sin

Particular solution:
16.

Particular solution:
15.

) + ] + ( 1)

1) = 0.

sin (

= 


) 1] = 0.

leads to a linear equation of the form 2.1.6.2:

+ [ ( ) sin (

=  exp( (
 ).

) 2 ] = 0.

sin2

sin(  ) + L

= 0.

cos(  ), where = X

tan 
M

 .

+ = 0.

This is a special case of equation 2.1.6.23.


22.

sin2

[ sin2 + ( 1)] = 0,
1
= sin 
 O
sin 

Solution:
23.

sin2

M
M

3P

= 1, 2, 3,

 2
 2
 L 1  + L 2   .

uuu

+ ( sin2 + ) = 0.

Set  = 2 . Applying the trigonometric formulas sin 2 = 2 sin cos and  =  (sin 2 +
cos2 )2 and dividing both sides of the equation by sin 2  , we arrive at an equation of the form
2.1.6.131:   + (  tan2 +  cot2 + 4  + 2  ) = 0.
24.

sin2

{[(

2 2

Particular solution:
25.

[ sin(

)+

sin (

+ (

Particular solution:
27.

( sin


=   sin  (cos  +  sin  ).

+ ] + 

Particular solution:
26.

( + 1)2 ] sin2 + ( + 1) sin 2 + ( 1)} = 0.

0
2

sin(

= 0.

=  sin(  ) + ' + ( .

2 ) ( + ) = 0.
0

=  .

+ ) + ( sin

Particular solution:

+ ) + [(  ) sin

 ]

= 0.

=  .

2003 by Chapman & Hall/CRC

Page 261

262

SECOND-ORDER DIFFERENTIAL EQUATIONS


TABLE 19
The general solution of the Mathieu equation 2.1.6.29 expressed
in terms of auxiliary periodical functions Y 1 ( ) and Y 2 ( )

Constraint
1(
1(

)>1


L 1  2  Y 1 ( ) + L 2 

Y

2(

) < 1


L 1  2 ~ Y 1 ( ) + L 2 

~Y

2(

General solution = ( )

| 1 ( )| < 1

Period of
1 and Y 2

is a real number

 )

 )

L 1 Y 1 ( ) + L 2 "Y 2 ( )

=  + 12 , 2 = 1,
 is the real part of

= @ is a pure
imaginary number,
cos(2 @ ) = 1 ( )

=0

( L 1 cos @! + L 2 sin @! ) Y 1 ( ) +


+ ( L 1 cos @! L 2 sin @! ) Y 2 ( )

1( ) = A 1

Index

2.1.6-2. Equations with cosine.


28.

+ 2 = cos( ).
Equation of forced oscillations.


Solution:
29.




cos(  ) if  ,
 2 2

L 1 sin(  ) + L 2 cos(  ) +  sin(  )
if  = .
2
L

sin(  ) + L

cos(  ) +

+ ( 2 cos 2 ) = 0.
The Mathieu equation.

1 b . Given numbers  and  , there exists a general solution ( ) and a characteristic index
such that
( + ) =  2 zf ( ).

For small values of  , an approximate value of can be found from the equation:
cosh( ) = 1 + 2 sin2 

1
2

  +



(1  )
2

sin    + [ (  4 ).

1 (  ) is the solution of the Mathieu equation satisfying the initial conditions


 (0) = 0, the characteristic index can be determined from the relation:
1

If

cosh(2 ) =

1(

= 1 and

1 (0)

).

The solution 1 ( ), and hence , can be determined with any degree of accuracy by means of
numerical or approximate methods.
The general solution differs depending on the value of 1 ( ) and can be expressed in

terms of two auxiliary periodical functions Y 1 ( ) and Y 2 ( ) (see Table 19).
2 b . In applications, of major interest are periodical solutions of the Mathieu equation that exist
for certain values of the parameters  and  (those values of  are referred to as eigenvalues).
The most important solutions are listed in Table 20.
The Mathieu functions possess the following properties:
ce2  ( ,  ) = (1)  ce2  O  ,  ,
P
2
se2  ( ,  ) = (1) 

se2  O  ,  ,
2
P

 ,  ) = (1)  se2 

+1

O  ,

 ,  ) = (1)  ce2 

+1

O  ,

ce2 

+1 (

se2 

+1 (

2
2

2003 by Chapman & Hall/CRC

Page 262

263

2.1. LINEAR EQUATIONS

TABLE 20
Periodical solutions of the Mathieu equation ce  = ce  ( ,  ) and se  = se  ( ,  ) (for odd B , the
functions ce  and se  are 2 -periodical, and for even B , they are -periodical); certain eigenvalues

 =   (  ) and  =   (  ) correspond to each value of the parameter
 ; B = 0, 1, 2, 
Recurrence relations
for coefficients

Mathieu functions

# 22 =  2 # 20 ;



# =( 2 4)# 22 2# 20 ;
# 22+ +2 =( 2 4 2)# 22+
# 22 + 2,
2

2  +1
# 3 =( 2 +11  )# 12 +1;
# 22+ +1+3 =[ 2 +1(2 +1)2]
2  +1
1
# 22 + +1
+1 # 2 + 1,

$ 42 =(  2 4)$ 22 ;


$ 22+  +2 =(  2 4 2)$ 22+ 
$ 22+  2,
2

$ 32 +1 =(  2 +11  )$ 12 +1;


$ 22+  +1+3 =[  2 +1(2 +1)2]
2  +1
1
$ 22+  +1
+1 $ 2 + 1,

+1(

se2  ( ,  )=

se2 

2
2

7
 ,  )= # 22 +
+ =0

+1(

+1
+1 cos

V (2 +1) W

$ 22+  sin(2  ),

=0

se0 =0

7
 ,  )= $ 22+ 
+ =0

+1
+1 sin

(#

2
4

ce2  ( ,  )= # + cos(2  )

+ =0

ce2 

Normalization
conditions

V (2 +1) W

 )2 + 7  #
+ =0
2 if B
=
1 if B
2
0

=0

# 22 +

2
2

 
+

=0
1

+1 2
+1 =1

2
2
+ $ 2 +  =1
=0

$ 22+ 
+
=0

+1 2
+1 =1

Selecting a sufficiently large


and omitting the term with the maximum number in the

recurrence relations (indicated in Table 20), we can obtain approximate relations for the
eigenvalues   (or   ) with respect to parameter  . Then, equating the determinant of the
corresponding homogeneous linear system of equations for coefficients # + (or $ +  ) to zero,
we obtain an algebraic equation for finding   (  ) (or   (  )).
For fixed real  0, the eigenvalues   and   are all real and different, while:

if
if

> 0 then
< 0 then

<
<

<
1 < 

1
1

<
<

2
2

<
<

2
2

< ))) ;
< 3<

<

< )))

The eigenvalues possess the following properties:

 2  (  ) =  2  (  ),

 2  (  ) =  2  (  ),

 2

+1 (

 ) =  2

+1 (

 ).

The solution of the Mathieu equation corresponding to eigenvalue   (or   ) has B zeros
on the interval 0  < (  is a real number).

Listed below are two leading terms of asymptotic expansions of the Mathieu functions
ce  ( ,  ) and se  ( ,  ), as well as of the corresponding eigenvalues   (  ) and   (  ), as D 0:
1

O 1 cos 2 ,
ce0 ( ,  ) =
2
P
2
ce1 ( ,  ) = cos 

 2 + 7 4 ;
2

128

 1 ( ) = 1 +  ;

cos 4
5 2
;
,  2 ( ) = 4 +
4
3 P
12
 cos(B + 2) cos(B 2) S ,  ( ) = B
ce  ( ,  ) = cos B/ +

4Q
B +1
B 1
ce2 ( ,  ) = cos 2 +

cos 3 ,

 0( ) =

O 1

2(B

2
2

1)

(B 3);

2003 by Chapman & Hall/CRC

Page 263

264

SECOND-ORDER DIFFERENTIAL EQUATIONS

se1 ( ,  ) = sin 

se2 ( ,  ) = sin 2

 1( ) = 1  ;

sin 3 ,

12

12
sin(B 2)
sin(B + 2)

S ,
4Q
B +1
B 1

se  ( ,  ) = sin B/

 2;

 2( ) = 4

sin 4 ,

  ( ) = B

2(B

2
2

1)

(B 3).

The Mathieu functions are discussed in the books by McLachlan (1947), Whittaker &
Watson (1952), Bateman & Erdelyi (1955, vol. 3), and Abramowitz & Stegun (1964) in more
detail.
30.

31.
32.
33.

+ ( cos2 + ) = 0.
Applying the formula 2 cos2  = 1 + cos 2 , we obtain the Mathieu equation 2.1.6.29:
   + ( 12  +  + 12  cos 2 ) = 0.

+ cos ( ) + [ cos (

Particular solution: 0 =  v .

+ [ cos ( ) + ] + cos (

Particular solution: 0 =  v .

= 0.

+ ( + ) cos ( ) cos (
Particular solution: 0 =  +  .

= 0.

+ cos ( )
Particular solution: 0 =  .

35.

38.

cos (

)
0

41.

cos (

+

cos (

, +1

 

+1

= 0.

+ (

+1

+ 1) cos (

+ cos' ) +

+ [

cos (

(


cos

 

+1

= exp O

) + ] + ( 1)

B +1

v.
0 = 
1

)+

+ ( cos

+ ( cos

Particular solution:

+ 1) + ( cos

=

=

.

+ ) + ( cos
0

.

 1 ]

= 0.

+ 2 ] = 0.
= 0.

+ ) = 0.

.
1

cos (

+ ( + cos ) + [ ( 1) cos
Particular solution: 0 =  exp(  
B ).

+ [( 2 + ) cos ( ) + 2] + cos (
Particular solution: 0 =  + 
 .

2

Particular solution:
42.

= exp O

+ cos ( ) [ (

Particular solution: 0 =   v .

Particular solution:

40.

+ [

Particular solution:
39.

Particular solution:

37.

) ] = 0.

34.

36.

)
+

= 0.

] = 0.

) = 0.
1) = 0.

2003 by Chapman & Hall/CRC

Page 264

265

2.1. LINEAR EQUATIONS

43.

cos (

v.
0 = 

Particular solution:
44.

cos (

cos2

Solution:
46.

cos2

[ cos(

( cos

) 2 ] = 0.

= 1, 2, 3,

2


2

uuu

+ ( cos2 + ) = 0.

)+

1
2

leads to a linear equation of the form 2.1.6.23: sin 2

+ ] + 

+ (

0
2

Particular solution:
49.

) 1] = 0.

[ cos2 + ( 1)] = 0,

2


1
 L 1  + L
M
= cos   O
cos   P

cos (

cos (

0 =  exp( (  ).

Particular solution:
48.

+ [ ( ) cos (

The substitution  = +
(  sin2 +  ) = 0.
47.

Particular solution:
45.

+ [

cos(

= 0.

=  cos(  ) + ' + ( .

( + ) = 0.

2 )

=  .

+ ) + ( cos

Particular solution:

  +

+ ) + [(  ) cos

 ]

= 0.

=  .

2.1.6-3. Equations with tangent.


50.

Particular solution:
51.

52.

= 0.

= [cos(  )] J  .

+ ( tan2 + ) = 0.

+ ( ) tan(

The transformation H = sin2  , = cos   , where , is a root of the quadratic equation


, 2 + , +  = 0, leads to the hypergeometric equation 2.1.2.171: H ( H 1) '  + [(1 , )H 12 ] 
1
4 ( , +  ) = 0.

Particular solution:
53.

)]

+ [ + ( ) tan2 (

tan

+  = 0.

= [cos(  )] J  .
= 0.

1 b . The substitution = sin  leads to a linear equation of the form 2.1.2.168: (


(1  )   = 0.

1)  +

2 b . Solution for  = 2:
cos  =

sin( ,! ) + L 2 cos( ,! )


if  + 1 = , 2 > 0,
L 1 sinh( ,! ) + L 2 cosh( ,! ) if  + 1 = , 2 < 0.

3 b . Solution for  = 2 and  = 3:


54.

tan

=L

+ ( tan2 + ) = 0.

cos3  + L

sin  (1 + 2 cos2  ).

This is a special case of equation 2.1.6.131.

2003 by Chapman & Hall/CRC

Page 265

266
55.

56.

57.

SECOND-ORDER DIFFERENTIAL EQUATIONS

2 tan(

+ (

2 tan(

+ (

tan (

+ [ tan (

tan (

tan

+ ( tan

+ [ tan (

+ tan ( tan

+ (

tan (

Particular solution:
64.

tan (

Particular solution:
65.

66.

) = 0.

+
0

tan

1) + ( tan

= .

)
0

tan (

=  +  .

)
0

= sin  cos   .

+ [

+
(

= exp O

tan (

[ (

, +1

+ 4) = 0.

+2

+1

) + ] = 0.

tan

+ 2 + 2) = 0.

= 0.

tan (

 

= 0.

tan (

)
0

) + ( ) tan2 (

+ tan (

+ (

= 0.

2

 1 ]

= 0.

+ ) = 0.

The substitution = cos(  ) leads to a second-order linear equation of the form 2.1.2.64:
 2   + [(  + 2 ) +  ] = 0.

+ [(

+ (

+1

+ (

v .
0 =  

) tan (
0

Particular solution:

tan

+ 1) tan (

)+

) + 2] + tan (

= +
 .

+ tan

Particular solution:
69.

Particular solution:
68.

+1

+1

tan(

Particular solution:
67.

) ] = 0.

v 
.
0 = 

+ ) tan (

= sin  cos  .

) + ]

Particular solution:
63.

J
0 = [cos(  )]   .

Particular solution:
62.

+ [ tan

Particular solution:
61.

v 
.
0 = 

Particular solution:
60.

+ ) = 0.

The substitution = cos(  ) leads to a second-order linear equation of the form 2.1.2.10:
2   + (  2  + '  1 ) = 0.

Particular solution:
59.

The substitution = cos(  ) leads to a second-order linear equation of the form 2.1.2.6:
   + (  2 + ' + ( + 2 ) = 0.

Particular solution:
58.

= exp O

(


 

+ [ (

B +1

tan
+1

  
B ).
0 =  exp( 

+ 2 ] = 0.
= 0.

+ ) = 0.

1) tan' +

] = 0.

2003 by Chapman & Hall/CRC

Page 266

267

2.1. LINEAR EQUATIONS

70.

+ [

tan (

Particular solution:
71.

72.
73.

74.

77.
78.
79.

v.
0 = 
1

tan (

= 0.

+ (1 2 tan ) ( tan + b 2 ) = 0.
Solution: cos  = L 1 ( ) + L 2 n ( ), where ( ) and n ( ) are the Bessel functions.

(2 tan + ) + ( 2 + + + tan ) = 0.
The substitution = cos  leads to a second-order linear equation of the form 2.1.2.131:
 2 2   ,! 2 + [(  + 1) 2 + ' + ( ] = 0.

2
2

+ ( tan

+ ( tan

+ 1)
0

=

+ )

Particular solution:

76.

) + ] + ( 1)

2 2 tan( ) + ( 2 + + ) = 0.
The substitution = cos(  ) leads to a second-order linear equation of the form 2.1.2.115:
 2 2   + [(  + 2 ) 2 + ' + ( ] = 0.

Particular solution:
75.

=

+ tan ( )
Particular solution: 0 = 

+ 2 tan (
Particular solution:

.

+ ( tan

.

+ [
v
.

) = 0.
1) = 0.

tan (

+ [ ( ) tan (
=  exp( (
 ).

)
0

+ ( tan

( tan + ) + ( + ) ^ = 0.
Particular solution: 0 = ( + .
( tan

+ )

+ ( tan

Particular solution:

=  .

+ )

) 1] = 0.
) 2 ] = 0.

+ [(  ) tan

 ]

= 0.

2.1.6-4. Equations with cotangent.


80.

Particular solution:
81.
82.

83.

= 0.

= [sin(  )] J  .

+ ( cot2 + ) = 0.
The substitution  = + z 2 leads to an equation of the form 2.1.6.51:  + (  tan2 +  ) = 0.

+ cot + b ( b + 1) = 0.
The substitution = cos  leads to the Legendre equation 2.1.2.154: (
@ ( @ + 1) = 0.

+ 2 cot(

Particular solution:
84.

)]

+ [ + ( ) cot2 (

+ ( ) cot(

Particular solution:

+ (
0

1)  + 2 

2 ) = 0.
cos( ' )
.
=
sin(  )

)
0

= 0.

= [sin(  )] J  .

2003 by Chapman & Hall/CRC

Page 267

268
85.

86.

SECOND-ORDER DIFFERENTIAL EQUATIONS


+ cot( ) + = 0.
The substitution H =  + z 2 leads to a second-order linear equation of the form 2.1.6.53:
'  1 tan H  +  2 = 0.

cot(

J
0 = [sin(  )]   .

Particular solution:
87.

cot

)]

+ [ + ( ) cot2 (

= 0.

+ ( cot2 + ) = 0.

This is a special case of equation 2.1.6.131.


88.

89.

90.

+ 2 cot(

cot (

+ [ cot (

v 
.
0 = 
v 
.
0 = 

= 0.

= 0.

+ ( + ) cot ( ) cot (
Particular solution: 0 =  +  .

cot (

Particular solution:

+ [

cot (

+

= exp O

cot (

, +1

 

+1

= 0.

2

 1 ]

= 0.

+ 2 cot( ) + ( + ) = 0.
The substitution = sin(  ) leads to a second-order linear equation of the form 2.1.2.64:
 2   + [(  + 2 ) +  ] = 0.

[ (

+ cot

cot (

v .
0 =  

Particular solution:

+ (

+1

Particular solution:

+ (

+ [

cot (

Particular solution:

+ 1) cot (

= exp O


 
B +1

cot' ) + [ (

Particular solution:
99.

98.

) ] = 0.

) + ] + cot (

94.

97.

+ [ cot (

+ cot ( )
Particular solution: 0 =  .

96.

+ ) = 0.

+ 2 cot( ) + ( 2 + 1 2 ) = 0.
The substitution = sin(  ) leads to a second-order linear equation of the form 2.1.2.10:
2   + (  2  + '  1 ) = 0.

93.

95.

Particular solution:
92.

The substitution = sin(  ) leads to a second-order linear equation of the form 2.1.2.6:
2   + (  2 + ' + ( + 2 ) = 0.

Particular solution:
91.

+ (

+1

v.
0 = 

+ 2 ] = 0.

1) cot

) + ] + ( 1)
1

)+

cot' + ) = 0.

  
B ).
0 =  exp( 

cot (

] = 0.

= 0.

2003 by Chapman & Hall/CRC

Page 268

269

2.1. LINEAR EQUATIONS

100.

101.

102.

+2

+ (2 cot + ) + (

+ ( cot

cot(

+ ) = 0.

+ 1) + ( cot

+ ( cot

+ +

cot (
cot (

=

=

.
.

+ [

v.
0 = 

Particular solution:
106. ( cot

+ ) + ( cot

Particular solution:
105.

cot ) = 0.

The substitution = sin  leads to a second-order linear equation of the form 2.1.2.131:
 2 2   + ,! 2 + [(  + 1) 2 + ' + ( ] = 0.

Particular solution:
104.

+ (

The substitution = sin(  ) leads to a second-order linear equation of the form 2.1.2.115:
 2 2   + [(  + 2 ) 2 + ' + ( ] = 0.

Particular solution:
103.

1) = 0.

cot (

+ [ ( ) cot (

=  exp( (
 ).

+ ) + ( cot

Particular solution:

) = 0.

) 1] = 0.
) 2 ] = 0.

+ ) + [(  ) cot

 ]

= 0.

=  .

2.1.6-5. Equations containing combinations of trigonometric functions.


107.

[ sin2 (

) + cos(

Particular solution:
108.

[ cos2 (

109.

+ ( sin + )

+ ( sin

+ ( cos + )

Particular solution:
112.

+ ( cos

= exp

Q 

= 0.
sin( ' )S .

= exp(  cos  ).

+ cos

+1

+ cos ) = 0.

= exp(  cos  ).

+ ( cos sin ) = 0.

= exp(  sin  ).

+ cos ) + ( cos

Particular solution:
113. sin

)]

cos( ' )S .

+ sin ) + ( sin

Particular solution:
111.

Q 

= 0.

+ ( sin + cos ) = 0.

Particular solution:
110.

= exp

) + sin(

Particular solution:

)]

= exp(  sin  ).

+1

sin ) = 0.

+ b ( b + 1) sin

= 0.
The substitution = cos  leads to the Legendre equation 2.1.2.154: (1 2 )  2  +
@ ( @ + 1) = 0.

2003 by Chapman & Hall/CRC

Page 269

270

SECOND-ORDER DIFFERENTIAL EQUATIONS

114. sin

+ (2 + 1) cos

+ ( b )( b + + 1) sin

= 0.

Here, @ is an arbitrary number and B is a positive integer. The substitution = cos  leads to
an equation of the form 2.1.2.156: ( 2 1)  + 2(B + 1)  + (B @ )( @ + B + 1) = 0.
115. sin2

+ sin cos

+ sin ( cos + ) + (  cos2 +  cos + ) = 0.

+ [ b ( b + 1) sin2

] = 0.

Here, @ is an arbitrary number and B is a nonnegative integer.


The transformation = cos  , = sin   leads to an equation of the form 2.1.2.156:
2
( 1)   + 2(B + 1)  + (B @ )(B + @ + 1) = 0.
116. sin2

Set  = 2 . Applying the trigonometric formulas

sin(2 ) = 2 sin cos , cos(2 ) = cos2 sin2 ,  =  (sin2 + cos2 ),


= (sin2 + cos2 ), = = = (sin2 + cos2 )2 ,
and dividing all the terms by sin2  , we arrive at an equation of the form 2.1.6.131:

  + [(   ) tan + (  +  ) cot ]  + [( + = ) tan2 + ( + + = ) cot2 + 2= 2 ] = 0.


117. cos2

+ sin(2 ) + [ cos(2 ) + ] = 0.

Dividing the equation by cos2  and applying the formulas


sin(2 ) = 2 sin  cos  ,

cos(2 ) = cos2  sin2  ,

( = ( (sin2  + cos2  ),

we obtain an equation of the form 2.1.6.131: "  + 2  tan   + [( (  ) tan2  +  + ( ] = 0.


118. cos2 (

+ ( 1) sin(2

Particular solution:
119. cos2

+ , 2 [( 1) sin2 (

= cos  (  ).

) + cos2 (

)]

= 0.

+ cos ( sin + ) + (  sin2 +  sin + ) = 0.

The substitution  = + z 2 leads to a second-order linear equation of the form 2.1.6.116:


sin2  sin (  cos +  )  + ( cos2 + cos + = ) = 0.

120. sin cos2

+ cos ( sin2 + ) + sin

= 0.

1 b . Dividing the equation by sin  cos  and assuming  =  (sin2  +cos2  ), ( = ( (sin2  +cos2  ),
we obtain equation 2.1.6.131:   + [(  +  ) tan  +  cot  ]   + ( (tan2  + 1) = 0.
2

2 b . Particular solutions:
0
0
0

121. sin cos2

= cos 
= tan1  
= sin1   cos

 

+ 1

if ( =  (  + 1),
if ( = (  + 2)(  1),
if ( = 2(  +  1).

+ cos ( sin2 1) + sin3

= 0.

= L 1 (cos  )  1 + L 2 (cos  )  2 , where , 1 and , 2 are roots of the quadratic equation

Solution:
, 2  , +  = 0.
122. sin2 cos2

+ ( sin2 + cos2 + sin2 cos2 ) = 0.

Dividing the equation by sin2  cos2  and assuming  =  (sin2  + cos2  ),  =  (sin2  + cos2  ),
we obtain equation 2.1.6.131:   + (  tan2  +  cot2  +  +  + ( ) = 0.

2003 by Chapman & Hall/CRC

Page 270

271

2.1. LINEAR EQUATIONS

123. sin (

sin (

+[

)+

The transformation = tan(  ),

2  + 

cos

 = 0.

)]

= 0.
leads to an equation of the form 2.1.2.7:

cos(  )

124. cos ( ) + [ 2 cos ( ) + sin 4 ( )] = 0.


The substitution  = z 2 leads to an equation of the form 2.1.6.123.
125.

+ tan

Solution:
126.

+ tan

=L

sin(  ) + L

L
Q

+ tan

128.

129.

=L

cot

130.

131.

)
1

=L

Solution:

= 0.

sin 

+L

1
2

O
B

 .
M

sin  

S , where ( H ) and n ( H )

= 0.

if  12 ,
L 1 | sin  | + L 2 | sin  |1
| sin  | ( L 1 + L 2 ln | sin  |) if  = 1 .
2

2 cot(2

Solution:

1) cot2

a (
=

Solution:

= 0.

cos(  ), where = X cos  

1
2

are the Bessel functions.


127.

cos2 (sin )2

= sin 

Solution:

(cos )2

exp
+

Q 

sin2 (2

sin2 (  )S + L

(sin )2

sin(  ) + L

= 0.
2

exp

Q 

sin2 (  )S .

= 0.

cos(  ), where = X sin  

 .
M

2 cot(2 ) + tan2 = 0.
The substitution = cos  leads to the Euler equation 2.1.2.123:

   + 

= 0.

+ ( tan + cot ) + (  tan2 +  cot2 + ) = 0.


The transformation = sin2  , = sin   cos +  , where B and are roots of the quadratic

equations
2
B 2 + (  1)B + = 0,
(  + 1) + = 0,

leads to the hypergeometric equation 2.1.2.171:


4 ( 1)   + 2[(2B + 2
132. sin(2 )
Solution:

2
=L

+ 2

+ 2 +   ) 2B  1]  + (2B
2

sin2 (tan )2

sin(  ) + L

+B +

+

B = ) = 0.

= 0.

cos(  ), where = X tan  

 .

2.1.7. Equations Containing Inverse Trigonometric Functions


2.1.7-1. Equations with arcsine.
1.

+ ( + + arcsin ) + [ ( + ) arcsin + ] = 0.
Particular solution: 0 = exp( 21  2 ' ).

2003 by Chapman & Hall/CRC

Page 271

272

SECOND-ORDER DIFFERENTIAL EQUATIONS

2.

+ (arcsin )
Particular solution:

3.

+ (arcsin )

Particular solution:
4.
5.
6.

9.

+ (arcsin )
Particular solution: 0 =  .

arcsin

+ [ (

12.
13.

16.
17.

 +

+1

(arcsin )

= ( ' + 1) 

1] +

+ [

+ (

+1

=

.

+ arcsin )

= exp O

+ 2)] = 0.

arcsin

+ arcsin
Particular solution:

2
2

= 0.


 
B +1

= 0.

arcsin + ) = 0.

+1

+ )

+ (

2
0

+ 2 arcsin
Particular solution:

( 2 + )2 + (
Particular solution:
2

+ )2 + (

+ )2 + (

Particular solution:

+ [ (

1 
.

0 = 

+ ( arcsin 1) = 0.
=  .

+ ( arcsin + 2)

+ 1) arcsin

+ )(arcsin )
= 

= 0.

+ ( + arcsin ) + [ ( 1) arcsin +
Particular solution: 0 =  exp(  
B ).

] = 0.

= 0.

(arcsin ) + ] + ( 1)(arcsin )
1

= 0.

+ 1) arcsin + (

+ [( 2 + ) arcsin + 2] + arcsin
Particular solution: 0 =  + 
 .

Particular solution:
18.

= 

Particular solution:
15.

+1

+ 1) arcsin +

Particular solution:
14.

[ (

Particular solution:
11.

+ ( + )(arcsin ) (arcsin )
Particular solution: 0 =  +  .

Particular solution:
10.

= exp O

(arcsin )

Particular solution:
8.

+ [
0

Particular solution:
7.

+ [ (arcsin ) ] = 0.
v 
.
0 = 

+ .

2 [

2 ]

] = 0.

= 0.

(arcsin ) + 1] = 0.

+ [ ( ) arcsin 2 ] = 0.
=  exp( 
 ).

+ )(arcsin )
0 = ( + .

+ )(arcsin )

= 

2
0

(arcsin )

+ .

+ )(arcsin )

= exp O X



M2

+  P

= 0.

(arcsin ) + ] = 0.

+ [ (arcsin ) 2

1] = 0.

2003 by Chapman & Hall/CRC

Page 272

273

2.1. LINEAR EQUATIONS

2.1.7-2. Equations with arccosine.


19.

+ ( + + arccos ) + [ ( + ) arccos + ] = 0.
Particular solution: 0 = exp( 21  2 ' ).

20.

+ (arccos )
Particular solution:

21.

+ (arccos )

Particular solution:
22.
23.
24.
25.

27.

30.
31.
32.

+ (arccos )
Particular solution: 0 =  .

arccos

Particular solution:

+ [ (

34.

[ (

= 

 +

+1

(arccos )

= ( ' + 1) 

1]

+ [

+ (

+1

=

.

+ arccos )
0

= exp O

+ 2)] = 0.

arccos


 
B +1

+ arccos
Particular solution:

Particular solution:

1 
.

0 = 

2
2

= 0.

= 0.

arccos + ) = 0.

+1

+ ( arccos 1) = 0.
=  .

] = 0.

+ ( arccos + 2) + [ (

+ ) + (

= 0.

+ ( + arccos ) + [ ( 1) arccos +
Particular solution: 0 =  exp(  
B ).

] = 0.

= 0.

(arccos ) + ] + ( 1)(arccos )
1

= 0.

+ [( 2 + ) arccos + 2] + arccos
Particular solution: 0 =  + 
 .

+ 1) arccos + (

+ 1) arccos +

Particular solution:
33.

+1

+ ( + )(arccos ) (arccos )
Particular solution: 0 =  +  .

Particular solution:
29.

= exp O

(arccos )

Particular solution:
28.

+ [
0

Particular solution:
26.

+ [ (arccos ) ] = 0.
v 
.
0 = 

2
0

+ )(arccos )
= 

+ .

+ 1) arccos

2 [

2 2

= 0.

(arccos ) + 1] = 0.

+ 2 arccos + [ ( ) arccos 2 ] = 0.
Particular solution: 0 =  exp( 
 ).

( 2 + )2 + (
Particular solution:

+ )(arccos )
0 = ( + .

(arccos )

= 0.

2003 by Chapman & Hall/CRC

Page 273

274
35.

SECOND-ORDER DIFFERENTIAL EQUATIONS


(

+ )2 + (

+ )(arccos )

Particular solution:
36.

+ )2

= 

+ (

Particular solution:

+ .

+ )(arccos )

= exp O X

M2



(arccos ) + ] = 0.

+ [ (arccos ) 2

1] = 0.

+  P

2.1.7-3. Equations with arctangent.


37.
38.
39.

+ ( + + arctan ) + [ ( + ) arctan + ] = 0.
Particular solution: 0 = exp( 21  2 ' ).

+ (arctan )
Particular solution:

+ (arctan )

Particular solution:
40.

+ (

(arctan )

Particular solution:
42.

arctan

Particular solution:
43.

+ [ (

44.

+ [(

+ [

+ (

+ (

+1

= .

[ (
= 

(arctan )

+ arctan

Particular solution:

.

=

= exp O

1] +

= 0.

+ 2)] = 0.

arctan


 
B +1

+ ( arctan


=  .

= 0.

= 0.

+1

= 0.

arctan + ) = 0.

.
1) arctan +

+ ) (arctan
= .

  
B ).
0 =  exp( 
0

] = 0.

= ( ' + 1) 

= 0.

+ 1) arctan + (

arctan ) + [ (

(arctan

Particular solution:
49.

+ arctan ) +

Particular solution:
48.

(arctan ) + ] + ( 1)(arctan )

Particular solution:
47.

(arctan )
1

0 =  +  .

Particular solution:
46.

+1

) arctan + 2] + arctan

Particular solution:
45.

+1

 +

+ 1) arctan +

Particular solution:

=  +  .

(arctan )

= exp O

+ )(arctan )

+ [
0

Particular solution:
41.

+ [ (arctan ) ] = 0.

=  v .

] = 0.

+ ) = 0.

= 0.

2003 by Chapman & Hall/CRC

Page 274

275

2.1. LINEAR EQUATIONS

50.

+ (arctan

Particular solution:
51.

+ arctan

Particular solution:
52.
53.

+ arctan
Particular solution:

2
2

55.

58.

+ (arctan
Particular solution:

61.

arctan

12

= exp 

  .

+ [ (

+ (arctan

64.

+ 1) = 0.

2 )

=

2 )

= 0.

+ arctan

= 0.

+ 1) [ 2 (

+ )

2 ]

= 0.

= 0.

+ 1)(arctan )2 + ] = 0.

+ (

+ ) = 0.

+ [( + ) arctan + 2] + arctan
Particular solution: 0 =  + 
 .

2
0

= (

+ 1) J 2 exp(  arctan  ).

+ )(arctan )
= 

+ .

2 [

(arctan ) + 1] = 0.

+ 2 arctan + [ ( ) arctan 2 ] = 0.
Particular solution: 0 =  exp( 
 ).

+ 1)2

+ [ (arctan )2 + arctan + ] = 0.

The transformation = arctan  ,

=


The transformation = arctan  ,

( 2 + )2 + (
Particular solution:

+1

+1

leads to an equation of the form 2.1.2.6:

+ 1)2 + [ (arctan ) 1] = 0.

+ )2 + (

Particular solution:
65.

= 0.

+ 1) arctan

+ ( arctan

=  .

+
0

+ ) (arctan
= .

2  + '  = 0.

63.

2  + (  2 + ' + ( + 1) = 0.

62.

+ ( arctan 1) = 0.
=  .

+ arctan

Particular solution:
60.

Particular solution:
59.

arctan

1 
.

0 = 

Particular solution:

57.

+ ( arctan + 2)

Particular solution:
56.

=

Particular solution:
54.

+ )2 + (

Particular solution:

=


+ )(arctan )
0 = ( + .

+ )(arctan )

= 

2
0

(arctan )

+ .

+ )(arctan )

= exp O X



leads to an equation of the form 2.1.2.7:

M2

+  P

= 0.

(arctan ) + ] = 0.

+ [ (arctan ) 2

1] = 0.

2003 by Chapman & Hall/CRC

Page 275

276

SECOND-ORDER DIFFERENTIAL EQUATIONS

2.1.7-4. Equations with arccotangent.


66.

+ (

+ arccot ) + [ (

Particular solution:
67.

+ (arccot )

Particular solution:
70.

71.

72.

(arccot )

arccot

Particular solution:

+ [ (

+ [(

+ (

+1

+ (

2
(

=

1] +

= 0.

.


 
B +1
+ [ (

+ 2)] = 0.

arccot

= 0.

= 0.

+1

= 0.

arccot + ) = 0.

.
1) arccot +

] = 0.

+ ( arccot 1) = 0.

=  .

+ )(arccot )

= 

+ ( 1)(arccot )

1 
.

0 = 

arccot

Particular solution:

+ ( arccot + 2) + [ (

] = 0.

= exp O

= 0.

(arccot )

  
B ).
0 =  exp( 

arccot

+ ) + (

+ 1) arccot + (

= ( ' + 1) 

arccot )

Particular solution:
80.

= 

Particular solution:
79.

+ arccot ) +

Particular solution:
78.

+1

(arccot )

[ (

Particular solution:
77.

+1

 +

0 =  +  .

Particular solution:
76.

) arccot + 2] + arccot

Particular solution:
75.

(arccot )

(arccot ) + ]

+ [

= .

Particular solution:
74.

+ 1) arccot +

Particular solution:
73.

' ).

=  +  .

Particular solution:

v

= exp O

+ )(arccot )

+ (

+ [

Particular solution:
69.

=

+ (arccot )

+ ) arccot + ] = 0.

+ [ (arccot ) ] = 0.

Particular solution:
68.

= exp( 21 

+ 1) arccot

2 [

2 ]

= 0.

(arccot ) + 1] = 0.

+ .

+ [ ( ) arccot 2 ] = 0.

=  exp( 
 ).

2003 by Chapman & Hall/CRC

Page 276

277

2.1. LINEAR EQUATIONS

81.

+ 1)2 + [ (arccot )2 + arccot + ] = 0.

The transformation = arccot  ,

=


2  + (  2 + ' + ( + 1) = 0.

82.

The transformation = arccot  ,

84.

( 2 + )2 + (
Particular solution:

=


+ ) + (
2

+ )(arccot )
0 = ( + .

+ )(arccot )

Particular solution:
85.

+1

+1

leads to an equation of the form 2.1.2.6:

+ 1)2 + [ (arccot ) 1] = 0.

  + '  = 0.

83.

+ )2 + (

= 

Particular solution:

(arccot )

+ .

+ )(arccot )

= exp O X

M2



leads to an equation of the form 2.1.2.7:

= 0.

(arccot ) + ] = 0.

+ [ (arccot ) 2

1] = 0.

+  P

2.1.8. Equations Containing Combinations of Exponential,


Logarithmic, Trigonometric, and Other Functions
1.

+  + [ +  tan(


Particular solution: 0 = cos( ' ).

2.

3.

+   + [   cot(
Particular solution: 0 = sin( ' ).

)]

= 0.

)]

= 0.

+ cosh ( ) + [ + cosh (
Particular solution: 0 = cos( ' ).

4.

+ cosh ( ) + [ cosh (
Particular solution: 0 = sin( ' ).

5.

7.
8.

cosh (

+ sinh ( )
Particular solution:

+ sinh

(
Particular solution:

sinh (

+ tanh ( )
Particular solution:

 [

cosh (

= exp O  

+ [ + sinh (
= cos( ' ).

Particular solution:
9.

Particular solution:
6.

 [

sinh (

= exp O  

)]

= 0.

) cot(

)]

= 0.



+ ] = 0.

+ [ sinh (
= sin( ' ).
+

) tan(

) tan(

)]

= 0.

) cot(

)]

= 0.

) 

+ ] = 0.

+ [ + tanh (
= cos( ' ).

) tan(

)]

= 0.

2003 by Chapman & Hall/CRC

Page 277

278
10.
11.

SECOND-ORDER DIFFERENTIAL EQUATIONS


+ tanh ( )
Particular solution:

tanh (

Particular solution:
12.
13.
14.

 [

tanh (

+ coth ( )
Particular solution:

coth (

 [

+ ln ( ) + [ + ln (
Particular solution: 0 = cos( ' ).

17.

ln (

Particular solution:

cos (

sin (

Particular solution:

tan (

Particular solution:
21.

cot (

Particular solution:
22.

 [

+ ln

+ ( 

Particular solution:

 [

 [

cot (

) cot(

)]

= 0.

= exp O  

+ ] = 0.

) 

ln

+ ] = 0.

+ ] = 0.



+ ] = 0.



+ ] = 0.



+ ] = 0.

 (

= exp O  

) 

= 0.

= 0.

tan (

)]

)]

) cot(

sin (

= exp O  

= 0.

) tan(

cos (

 [

)]

= exp O  

 [

+ ] = 0.

) tan(

) 

= exp O  

= exp O  

ln (



= 0.

Particular solution:

20.

+ ln ( ) + [ ln (
Particular solution: 0 = sin( ' ).

19.

coth (

= exp O  

)]

+ [ coth (
= sin( ' ).

) cot(

+ coth ( ) + [ + coth (
Particular solution: 0 = cos( ' ).

16.

18.

= exp O  

Particular solution:
15.

+ [ tanh (
= sin( ' ).

+ ) = 0.

23.

+ (   + cos ) + (   cos sin ) = 0.


Particular solution: 0 = exp(  sin  ).

24.

+ ( 

+ cos

Particular solution:

 (


= exp O  

cos

+ ) = 0.

2003 by Chapman & Hall/CRC

Page 278

279

2.1. LINEAR EQUATIONS

25.

+ ( 

+ cos

Particular solution:
26.

+ ( 

+ ( cosh

+ ( cosh

+ ( cosh

+ ( cosh

+ ( cosh

+ ( cosh

( 

= cos  +1  .

)
0

 (

= exp O  

+ ( sinh

= sin1   .

)
0

= exp O  

tan + 1) = 0.

+ ) = 0.

+ 

 .
M P

tan

cot 1) = 0.

cot

+ ) = 0.

= exp(  sin  ).

= exp O  X cos + 

+ sin

= exp O  X sin + 

cos sin ) = 0.
cosh

+ cos ) + ( sinh

Particular solution:

= exp(  sin  ).

cos

sin ) = 0.

sin + cos ) = 0.

cosh

 .
M P

= cos  +1  .
= sin1   .

 .
M P

= exp(  cos  ).

sin + cos ) = 0.

+ cot ) + ( 1)( cosh

Particular solution:
39.

+ sin

+ ) = 0.

+ tan ) + ( + 1)( cosh

Particular solution:
38.

= exp O  X sin  

+ sin

Particular solution:
37.

= exp O  

+ sin ) + ( cosh

Particular solution:
36.

sin

+ cos' ) + cos

Particular solution:
35.

 (

+ cos ) + ( cosh

Particular solution:
34.

+ cot

Particular solution:
33.

+ cot ) + ( 1)(-

+ ( 

Particular solution:
32.

sin + cos ) = 0.

= exp(  cos  ).

+ tan

+ ( 

Particular solution:
31.

cos sin ) = 0.

 .
M P

+ tan ) + ( + 1)(-

+ ( 

Particular solution:
30.

(  

= exp O /X cos  

+ sin

+ ( 

Particular solution:
29.

+ sin

+ ( 

Particular solution:
28.

+ cos

+ sin ) + ( 

+ ( 

Particular solution:
27.

sin +

cos ) = 0.

tan + 1) = 0.
cot 1) = 0.

cos sin ) = 0.

2003 by Chapman & Hall/CRC

Page 279

280
40.

SECOND-ORDER DIFFERENTIAL EQUATIONS

+ ( sinh

+ cos

Particular solution:

+ cos

= exp O  X cos + 

41.

+ ( sinh + sin ) + ( sinh


Particular solution: 0 = exp(  cos  ).

42.

+ sin

Particular solution:
43.

+ ( sinh

+ ( sinh

+ sin

 .
M P

= cos  +1  .

= sin1   .

+ cot ) + ( 1)( sinh

Particular solution:
45.

+ ( tanh + cos ) + ( tanh


Particular solution: 0 = exp(  sin  ).

46.

+ ( tanh

+ cos
0

+ cos

= exp O  X cos + 

47.

+ ( tanh + sin ) + ( tanh


Particular solution: 0 = exp(  cos  ).

48.

+ ( tanh

+ sin

Particular solution:

+ ( tanh

+ ( tanh

+ sin

= cos  +1  .

= sin1   .

+ ( coth + cos ) + ( coth


Particular solution: 0 = exp(  sin  ).

52.

+ cos' ) + cos

Particular solution:

= exp O /X cos 

53.

+ ( coth + sin ) + ( coth


Particular solution: 0 = exp(  cos  ).

54.

+ sin' ) + sin

Particular solution:
55.

+ ( coth

tanh

= exp O /X sin + 

+ tan )

Particular solution:

cos

sin ) = 0.

tanh

sin +

cos ) = 0.

tan + 1) = 0.
cot 1) = 0.

cos sin ) = 0.
coth

cos

sin ) = 0.

 .
M P

sin + cos ) = 0.

 .
M P

+ ( + 1)( coth

= cos  +1  .

sin + cos ) = 0.

+ ( coth

cos sin ) = 0.

 .
M P

51.

+ ( coth

cot 1) = 0.

+ cot ) + ( 1)( tanh

Particular solution:

cos ) = 0.

 .
M P

= exp O /X sin 

sin +

tan + 1) = 0.

+ tan ) + ( + 1)( tanh

Particular solution:
50.

Particular solution:

49.

sinh

= exp O  X sin + 

sin ) = 0.

sin + cos ) = 0.

+ tan ) + ( + 1)( sinh

Particular solution:
44.

cos

 .
M P

+ ( sinh

sinh

coth

sin +

cos ) = 0.

tan + 1) = 0.

2003 by Chapman & Hall/CRC

Page 280

281

2.1. LINEAR EQUATIONS

56.

+ ( coth

+ cot )

Particular solution:
57.

+ ( ln

+ ( ln

+ ( ln

+ ( ln

+ ( ln

+ ( ln



64.

Particular solution:
66.

Particular solution:
67.

cosh (

Particular solution:
69.

cosh (

Particular solution:
70.

Particular solution:
71.

cosh(

= cos( ' ).

+ [ 

= sin( ' ).

tan + 1) = 0.

cot 1) = 0.

)]

= 0.

cos(

)]

= 0.
) ln (

[ +

+ [ + cosh (

[ + sinh(

+ [ cosh (

[ + sinh(

[ +

sinh(

= cos( ' ).

= sin( ' ).

= cosh( ' ).

sinh(

)]

= 0.

)]

= 0.

)]

= 0.

) sin (

)]

= 0.

) cos(

)]

= 0.

) tan (

)]

= 0.

) cot (

)]

= 0.

) cos (

= cosh( ' ).

cos ) = 0.

 .
M P

[ + sinh(

= cosh( ' ).

sin ) = 0.

sin +

sin(

= cosh( ' ).

) cot (

Particular solution:

ln

= cosh( ' ).

) sin(
0

+ [ + 

) tan (

cosh(

= sin1   .

cos

 .
M P

= cos  +1  .

) sin (

cosh(

= exp O /X sin + 

) cos(

Particular solution:
68.

sin + cos ) = 0.

= exp(  cos  ).

) cos (

cosh(

) ln (

cosh(

sin(

Particular solution:
65.

cos(

Particular solution:

+ 

= exp O /X cos 

ln

+ cot ) + ( 1)( ln

Particular solution:
63.

= exp(  sin  ).

+ tan ) + ( + 1)( ln

Particular solution:
62.

cot 1) = 0.

cos sin ) = 0.

+ sin' ) + sin

Particular solution:
61.

+ ( ln

+ sin ) + ( ln

Particular solution:
60.

+ cos' ) + cos

Particular solution:
59.

+ ( 1)( coth

= sin1   .

+ cos )

Particular solution:
58.

) sin(

2003 by Chapman & Hall/CRC

Page 281

282
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.

SECOND-ORDER DIFFERENTIAL EQUATIONS


+ sinh( ) ln ( ) [ +
Particular solution: 0 = sinh( ' ).

+ sinh( ) cos ( ) [ + cosh(


Particular solution: 0 = sinh( ' ).

+ sinh ( ) cos( ) + [ + sinh (


Particular solution: 0 = cos( ' ).
+ sinh( ) sin ( ) [ + cosh(
Particular solution: 0 = sinh( ' ).

+ sinh ( ) sin( ) + [ sinh (


Particular solution: 0 = sin( ' ).

+ sinh( ) tan ( ) [ +
Particular solution: 0 = sinh( ' ).

+ sinh( ) cot ( ) [ +
Particular solution: 0 = sinh( ' ).

cosh(

cosh(

+ tanh ( ) cos( ) + [ +
Particular solution: 0 = cos( ' ).

= 0.

) sin(

)]

= 0.

) sin (

)]

= 0.

) cos(

)]

= 0.

) tan (

)]

= 0.

) cot (

)]

= 0.

) sin(

)]

= 0.

) cos(

)]

= 0.

) sin(

)]

= 0.

) cos(

)]

= 0.

+ tanh ( ) sin( ) + [ tanh (


Particular solution: 0 = sin( ' ).

+ coth ( ) cos( ) + [ +
Particular solution: 0 = cos( ' ).

coth (

+ coth ( ) sin( ) + [ coth (


Particular solution: 0 = sin( ' ).

+ ln ( ) cos( ) + [ +
Particular solution: 0 = cos( ' ).

ln (

)]

= 0.

) cos(

)]

= 0.

 2

) ln (

+ ( +

 2

) ln (

Particular solution:

+ ( +

 2

+ ( +

 2

Particular solution:

+   .

+ [( 

+ [( 

+ [(

=  

+   .

=  

) tan (

+ [(

=  

) sin (

Particular solution:

=  

) cos (

Particular solution:

+ ( +  2

+   .

+   .

) sin(

= 0.

)]

tanh (

85.

88.

)]

) cos (

+ ln ( ) sin( ) + [ ln (
Particular solution: 0 = sin( ' ).

87.

84.

86.

) ln (

cosh(

 2

) ] = 0.

) cos (
) sin (
) tan (

) ] = 0.

) ] = 0.
) ] = 0.

2003 by Chapman & Hall/CRC

Page 282

283

2.1. LINEAR EQUATIONS

89.

+ ( +

 2

) cot (

Particular solution:
90.

91.

92.
93.
94.
95.
96.
97.
98.
99.
100.
101.
102.
103.
104.

=  

 2

+ [(

+   .

) cot (

) ] = 0.

+ ( sn2 + ) = 0.
The Lame equation in the form of Jacobi, sn  is the Jacobi elliptic function. See the books
by Whittaker & Watson (1952), Bateman & Erdelyi (1955, Vol. 3), and Kamke (1977) for
information on this equation.

+ [ 4 ( ) + ( ] = 0.
The Lame equation in the form of Weierstrass, < ( ) is the Weierstrass function. See the
books by Whittaker & Watson (1952), Bateman & Erdelyi (1955, Vol. 3), and Kamke (1977)
for information on this equation.

+ ( ln +
Particular solution:

 )

+ (1 

+ ( 
 

= 
.

ln ) + 
Particular solution: 0 = ln  .

+ (1 cosh ln ) +
Particular solution: 0 = ln  .

ln + 1) = 0.

= 0.

+ ( ln + cosh ) + ( cosh


Particular solution: 0 =   .

cosh

+ ( ln + sinh ) + ( sinh


Particular solution: 0 =   .

+ (1 sinh
Particular solution:

ln )
= ln  .

+ sinh

+ (1 tanh ln )
Particular solution: 0 = ln  .

ln + 1) = 0.

= 0.

+ tanh

+ ( ln + coth ) + ( coth


Particular solution: 0 =   .

+ (1 coth ln ) +
Particular solution: 0 = ln  .

coth

+ ( ln + cos ) + ( cos


Particular solution: 0 =   .
+ (1 cos
Particular solution:

ln ) + cos
= ln  .

+ ( ln + sin ) + ( sin


Particular solution: 0 =   .

= 0.

+ ( ln + tanh ) + ( tanh


Particular solution: 0 =   .

ln + 1) = 0.

ln + 1) = 0.
= 0.
ln + 1) = 0.
= 0.

ln + 1) = 0.
= 0.
ln + 1) = 0.

2003 by Chapman & Hall/CRC

Page 283

284
105.

SECOND-ORDER DIFFERENTIAL EQUATIONS

sin

+ (1

Particular solution:

ln )

106.

ln + tan

+ (

tan

+ (1

Particular solution:
108.

+ (

+ (1

cot

Particular solution:
110.

+ ( ln +

Particular solution:
111.

2
2
2
2
2
2
2
2

+ ( 

ln + 1) = 0.

= 0.

ln ln + 1) = 0.

+ ( cosh

+ ( sinh

+ ( tanh

+ ( coth

= exp( 21  ln2  ).
= exp( 21  ln2  ).

+ ( cos

+ ( sin

+ ( tan

+ ( cot

+ ( 

= exp( 21  ln2  ).

ln ln + 1) = 0.

= exp( 21  ln2  ).
= cot 
= cot 

1
2

1
2

 .

)
 .

ln ln + 1) = 0.
ln ln + 1) = 0.

ln ln + 1) = 0.

ln ln + 1) = 0.

= exp( 21  ln2  ).
= exp( 21  ln2  ).

ln ln + 1) = 0.

= exp( 21  ln2  ).

+ sin ( + cosh

Particular solution:

= 0.

= exp( 21  ln2  ).

+ sin ( + -

Particular solution:
120. sin2

cot

ln + 1) = 0.

= exp( 21  ln2  ).

+ ( ln + cot

Particular solution:
119. sin2

+ ( ln + tan

Particular solution:
118.

  )

+ ( ln + sin

Particular solution:
117.

= ln  .

+ ( ln + cos

Particular solution:
116.

+ ( ln + coth

Particular solution:
115.

ln ) +

+ ( cot

   .
0 = 

+ ( ln + tanh

Particular solution:
114.

+ ( ln + sinh

Particular solution:
113.

= ln  .

+ ( ln + cosh

Particular solution:
112.

= 0.

+ ( tan

ln ) + tan

ln + cot

Particular solution:
109.

   .
0 = 

Particular solution:
107.

= ln  .

+ sin

ln ln + 1) = 0.
ln ln + 1) = 0.

cos ) = 0.

+ ( cosh

cos ) = 0.

2003 by Chapman & Hall/CRC

Page 284

285

2.1. LINEAR EQUATIONS

121. sin2

+ sin ( + sinh

Particular solution:
122. sin2

Particular solution:

Particular solution:

Particular solution:

= cot 

Particular solution:

+ ( coth

 )

= cot 
= cot 
= cot 
= cot 
= cot 

+ cos ( + ln

+ ( tanh

1
2

1
2

+ cos ( + coth

Particular solution:
130. cos2

 .
 .
 .

+ ( ln

+ cos ( + tanh

Particular solution:
129. cos2

1
2

+ ( sinh

1
2

+ cos ( + sinh

Particular solution:
128. cos2

= cot 

+ cos ( + cosh

Particular solution:
127. cos2

+ cos ( +

Particular solution:
126. cos2

= cot 

+ sin ( + ln

125. cos2

+ sin ( + coth

124. sin2

= cot 

+ sin ( + tanh

123. sin2

= cot 

 .

+ ( 

1
2

 +

1
2

 +

1
2

 +

1
2

 +

1
2

 +

1
2

 +




+ ( sinh

+ sin ) = 0.

.

1
4

+ ( tanh

.

1
4

+ ( coth

.

.

1
4

+ ( ln
1
4

+ sin ) = 0.

.

1
4

cos ) = 0.

+ ( cosh

+ sin ) = 0.

.

1
4

cos ) = 0.

cos ) = 0.

cos ) = 0.

+ sin ) = 0.
+ sin ) = 0.

+ sin ) = 0.

2.1.9. Equations with Arbitrary Functions


]

Notation:  =  ( ) and = ( ) are arbitrary functions;  ,  , ( , , B , , , , , , , and = are

M
arbitrary parameters.
2.1.9-1. Equations containing arbitrary functions (but not containing their derivatives).
1.

=! .

Equation of forced oscillations without friction.


Solution:

where 





X   ( ) sin[ , ( )]
M
0 
1
L 1 cosh( ,! ) + L 2 sinh( ,! ) + , X   ( ) sinh[ , ( )]
M
0

L 1  + L 2 + X  ( )  ( )
M
0





cos( ,! ) + L

sin( ,! ) + ,

if  = ,

if  = ,

> 0,
2

< 0,

if  = 0,

is an arbitrary number.

2003 by Chapman & Hall/CRC

Page 285

286

SECOND-ORDER DIFFERENTIAL EQUATIONS

2.

+ + = ! .
Equation of forced oscillations with friction. The substitution = exp( 21  ) leads to an
equation of the form 2.1.9.1: u   + (  14  2 ) =  exp( 12  ).

3.

=" .
=L

Solution:

+ X  O L

4.

+ ! + ( ! ) = 0.

Particular solution: 0 =  .

5.

+ (

+ ! ! = 0.
Particular solution: 0 =  .

7.

+ (! +

+ ! [(
Particular solution:

9.

+ (

16.

+ [(

 

+1

= exp O

+ )! +

B +1

+ 1) ! + (
 .
0 =  

 

+1

'

] = 0.

+ 2)] = 0.

=

+ [( + 1) ! + 1] + 2

Particular solution: 0 = (  + 1)  .

= 0.

+ [( 2 + ) ! + 2] + ! = 0.
Particular solution: 0 =  + 
 .

+ (! +

+1

Particular solution:

15.

B +1

) = 0.

+ ) + ( 1) ! = 0.

Particular solution:

14.

+ )

8.

13.

Particular solution:

12.

= exp O

6.

11.


 , where  = X  .
M u
P M
M

Particular solution:

10.

+X 

(!

= exp O

+ ) = 0.


 
B +1

+1

+ ( ! + ) + [( 1) ! +
Particular solution: 0 =  exp(  
B ).

+ [( + 1) ! + + 1 2 ] +
Particular solution: 0 = (   + 1) exp(   ).

] = 0.

2 1 !

= 0.

+  +  = ! .
g
The nonhomogeneous Euler equation. The substitution  =  leads to an equation of the
g



form 2.1.9.2: g g + ( 1) g + =  (  ).

2
2

+ + ( 2 b 2 ) = ! .
The nonhomogeneous Bessel equation. The general solution is expressed in terms of Bessel
functions: = L 1 ( ) + L 2 n ( ) + 12 n ( ) X  ( )  ( )  12 ( ) X "n ( )  ( )  .

2003 by Chapman & Hall/CRC

Page 286

287

2.1. LINEAR EQUATIONS

17.
18.

+ ! + ( ! 1) = 0.
Particular solution: 0 =  .

+ ( ! + 2 )

Particular solution:
19.

21.
22.
23.

26.
27.

+ )!

2(

+2

 

] = 0.

2 +1

2B + 1

= 0.

+ ) + ( 2 + ) ! [(
Particular solution: 0 =  + 
 .

+ )2 + (

+ ) ! (

+ )2 + (


0 =

+ .

( + ) + (
Particular solution:

= (

+ ) !
 + .
0 = 

+ )2 + (

30.

+ 

(!

Particular solution:

+ (! +

 )

Particular solution:

+ ( +



^ ]

 ( !

= exp O  

+ ] = 0.

+ 1) = 0.

+ ) ! + ( 1)

] = 0.

+ ) = 0.

1) = 0.

+ ) = 0.

+ [(

+ ) = 0.

=  

= exp O  

+ ) = 0.

 2 ) !

Particular solution:


M
.
  +  P

= exp O X

+ [ 2 + ( + ) +
Particular solution: 0 =  + ( .

+ ) ! + ( !

! ( )
+

 +  )1 J .
0 = (

+ ) ! [(
=   + ' .

( + )2 + ( + ) !

1)

+ (

+  )+ J 2.

) ! + 2 ] = 0.

+ ) = 0.

+ ) !

( + ) + (
Particular solution:

29.

32.

+ ( 1)] = 0.

+ 2 ! + [( ) ! 2 ] = 0.
Particular solution: 0 =  exp(
 ).

Particular solution:

31.

Particular solution:
28.

+1

( 2 + + ) + ( + ) !
Particular solution: 0 =  + , .

Particular solution:

25.

  .
0 = 

 exp O
0 = 

Particular solution:

24.

Particular solution:
20.

+ )!

+ [(

+ [(

 2 ) !

+   .

] = 0.

2003 by Chapman & Hall/CRC

Page 287

288
33.

SECOND-ORDER DIFFERENTIAL EQUATIONS


( 

+ )2

+ ( 

Particular solution:
34.

[ +

Particular solution:
36.

+ (1

37.

+ (! +

ln ) +

2
2

sin(

cos(

cos

+ (! +

sin2

48.

= 0.

= 0.

1
4

(ln +

+ 2) !hWU = 0.

= exp( 21  ln2  ).

cos(

)]

= 0.

sin(

)]

= 0.

= cos(  ).

= exp(  cos  ).

= exp(  sin  ).

sin

)
0

cos

(!

= exp O lX cos  


+

sin

(!

cos sin ) = 0.

 .
M P

sin + cos ) = 0.

= exp O  X sin  

 .
M P

+ sin ( ! + ) + ( ! cos ) = 0.

Particular solution:
47.

= 0.

= sin(  ).

Particular solution:
46.

cos ) + ( ! cos sin ) = 0.

+ (! +

+ (! +

sin ) + ( ! sin + cos ) = 0.

+ (! +

Particular solution:
45.

+ ( ! ln ln + 1) = 0.

+ [ +

Particular solution:
44.

)]



=   .

Particular solution:
43.

sinh(

+ [

Particular solution:
42.

= 0.

= ln  .

Particular solution:
41.

)]

+ ( ! + ln )

cosh(

 .

 (ln  +  ).
0 =

Particular solution:
40.

+ 2 (ln + ) ! + V

Particular solution:
39.



ln ) + ( ! ln + 1) = 0.

Particular solution:
38.

+ 

(!


= cosh(  ).

Particular solution:

+ 
v

= sinh( ).

cosh(

+ ) !

[ +

Particular solution:
35.

=   

sinh(

= cot 

1
2

 .

cos2 + cos ( + ! ) + ( ! + sin ) = 0.


Particular solution: 0 = cot  12  + 14  .

+ [ +

Particular solution:

tan(

) + ( ) tan2 (

J
0 = [cos(  )]  .

)]

= 0.

2003 by Chapman & Hall/CRC

Page 288

289

2.1. LINEAR EQUATIONS

49.

tan )

+ (! +

Particular solution:
50.

+ [

Particular solution:
52.

= cos

+1

 .

+ tan ( ! + 1) + [( tan2 1) ! + 2 + 2] = 0.

Particular solution:
51.

+ ( + 1)( ! tan + 1) = 0.

+ (! +

= sin  cos  .

cot(

) + ( ) cot2 (

J
0 = [sin(  )]  .

)]

= 0.

cot ) + ( 1)( ! cot 1) = 0.

Particular solution:

= sin1  .

2.1.9-2. Equations containing arbitrary functions and their derivatives.


53.

(!

! )

= 0.

Particular solution:
54.

[ ( + 1) !

+ 2 ! + ( !

= (L

Solution:
56.

57.

+ (1 ) !

= exp O X

! )

(!

= 0.

! )

= 0.

Particular solution:

= exp O X

+"

= exp O X

+ ( !" "

+ 2 !

+ (!

+ 2 ! + ( !

2   +  ( 2  + ' 

 .
M P

= 0.

 .
M P

+ ) = 0.

exp O X


M P

exp OX

+ ! +

! )

) = 0.

+ ) = 0.

exp OX

+ ( ! + " ) + ( !" +

leads to a constant coefficient linear equation:

 leads to a linear equation of the form 2.1.2.10:


M P

) = 0.

+ (2 ! + ) + ( !

Particular solution:

The substitution =
   +   +  = 0.
61.

= 0.

 .
M P

The substitution

60.

! ]

 + L 1 ) exp O X 

The substitution
2   +  = 0.
59.

 .
M P
 .
M P

Particular solution:
58.

Particular solution:
55.

= exp O X


M P

leads to a constant coefficient linear equation:

= 0.

= exp O X

 .
M P

2003 by Chapman & Hall/CRC

Page 289

290
62.

SECOND-ORDER DIFFERENTIAL EQUATIONS

+ (! +

Particular solution:
63.

+ (

+ ( ! +

Particular solution:
64.

( + )

66.

= exp O Xv

 .
M P

Particular solution:

+ 2

+ (

67.

68.

1 
 .
 + M P

( + 1)] = 0.
+1

+ (!

+ (2 ! + ) + [ !
=

+ [

The transformation

 2   + ( 
2

70.

The substitution

 .
M P

) = 0.

   +
 leads to the Bessel equation 2.1.2.126:  2 2
M P

( ! +

+ ) + ] = 0.

exp O X

+ ) + [ !
exp O X

= 0.

 leads to an equation of the form 2.1.2.115:


M P

exp O X 

+  ) = 0.

+ (2 ! +

+ ( 1) ! +

    +   + ( '  + ( ) = 0.


+ 2

 ) = 0.

The substitution

69.

exp OX

exp O X

= exp O X

The substitution
2

= 0.

2
+ ' + ( ) = 0.

+ (2 ! + 1)

 + (
 2

!

=

The transformation

 2 2   + ( 

+ [

= 0.

= exp O X

 .
M P

! )

+ ( ! + )

Particular solution:
65.

= 0.

=  exp O Xv

+ )

! )

+ ] = 0.

 leads to a linear equation of the form 2.1.2.132:


M P

 leads to a linear equation of the form 2.1.2.115:


M P

+ (

1) ! +

+

+

+ ] = 0.

  leads to a linear equation of the form 2.1.2.146:


M P

2

2




2

  + (  +  )  + (  +   + = ) = 0.
=

71.

2 ! +

!

= 0.

The substitution = Xv


72.

Solution:

!
=L

1 2

= 0.
+L

 leads to a constant coefficient linear equation: 2 + 

, where =  X


M

= 0.

 .

2003 by Chapman & Hall/CRC

Page 290

291

2.1. LINEAR EQUATIONS

73.

!

=L

Solution:
74.

(!

! 2D +1 = 0.


 + L  , where

+ !
=L

Solution:

exp O

quadratic equation
75.

76.

 +L
M P

! 2 (

= 0.

= exp O  X 

 .
M P

( ! + 2 ! ) + ( ! +

! 2
! 2

2!

2 ! 3 )

 exp O/Xv
0 = 

+ ! ( ! + ) +

+ ! (!

exp O

! )

The substitution = Xv


78.

X

!"I

=L

Solution:
80.

"

+ 2 !

+ (!


2

The substitution

+L

and

are roots of the

 .
M P

 leads to a constant coefficient linear equation:  +   +  = 0.

+ ( !"

!"I )
+

 , where
M P

= 0.

+ 2" + )

( !

X

= 0.

+"

+ " + ) = 0.

The transformation = Xv 1  , = exp O Xv


M
equation:   +   + ' = 0.
79.

 .
M

 +  = 0.

Particular solution:
77.

= 0.

( ! + ! )

Particular solution:

! 3

= /Xv

#! 2D +1 " 2 +1




 leads to a constant coefficient linear


M P

= 0.

, where = 3Xv

+ 

 

 
M

+ ) = 0.

= exp O X 
leads to a linear equation of the form 2.1.3.5:
M P
2   + (   2   +    + ( ) = 0.

81.

82.

2 2

Solution:

=L

2 2

=L

!

Solution:
83.

4 ! 2 [2 !
Solution:




= 0.

exp O Xv

+L

M P

M P

+L

cos O lX 

exp O  XZ

 .
M P

 .
M P

= 0.
1

 + L 23
 X
(!

)2 + ]

 exp O

1
2

 X 

 .
= 0.

 + L 2  exp O 21  X 
M P

1
1
 + L 2  sin O 12 |  | v
L 1  cos O 2 |  | Xv
X 

M P

L 1  + L 2  X  1 

M
L



= 0.

sin O lX 

=L

Solution:
84.


M P
1

M P

if  > 0,
if  < 0,
if  = 0.

2003 by Chapman & Hall/CRC

Page 291

292
85.

SECOND-ORDER DIFFERENTIAL EQUATIONS

L

+ 2 ( ! )

Solution:

! 2 2


Bessel functions.
86.

! !
!

=L

Solution:
87.

!
!

Solution:
89.

!
!

"
"

= % [ L

)+ L

2 ( ! )2

! 2+

2
2

+L

1
2

  , where + ( H ) and n + ( H ) are the


P

= 0.

+L

 +

+ (

+ 

!
!

.

+ (!

)2

= 0.

n  (  )], where  (  ) and n  (  ) are the Bessel functions.


1

!
!

+ (!

)2

= 0.

(  )+ L 2 n (  )], where  (  ) and n  (  ) are the Bessel functions.

1
2

!
!

"
"

1
2

"
"

"
"

"
"

"
"

+("I )2 = 0.

( )+ L 2 n ( )], where  ( ) and n  ( ) are the Bessel functions.

"

"

!
!
2
!
!

"
"

!
!

+ ("I )2 = 0.

( ) + L 2 n ( )], where  ( ) and n  ( ) are the Bessel functions.

+(2 1)

Solution:

 

+ 

!
!


  [ L

"
"

!

!

!
!

 (


   [L

=  [L

Solution:
91.

+
=

Solution:
90.

 +

=  [L

!
!

1
2

+ (2 1)

Solution:
88.

1
2

= 0.

"
"
1

+2

% "

%
"

+(2 1)

"
"

+2

%
%

%
%

+("I )2 = 0.

( ) + L 2 n ( )], where  ( ) and n  ( ) are the Bessel functions.

2.1.10. Some Transformations


]

Notation:  , , and % are arbitrary composite functions of their arguments, which are written in
parentheses following the name of a function (the argument is a function of  ).
1.

2.

3.

(1 

= 0.

  +  ( ) = 0.
The transformation = 1
 , = 
 leads to the equation p

+
= 0.
+ P
 + 
, =
The transformation =
( +
(  +
M
M
where =  ( .

+ (

+ V

+ ) 4 !IO

! (

+ )+

1
4

1
4

The transformation =   +  , = 

W


1
2

  +
leads to a simpler equation: u

 ( ) = 0,

= 0.

  +( B )2  ( ) = 0.
leads to a simpler equation: u

2003 by Chapman & Hall/CRC

Page 292

293

2.1. LINEAR EQUATIONS

4.

5.

+ ( ! + ) + ( " + ) = 0,
! = ! ( ), " = " ( ).
The substitution =   , where , is a root of the quadratic equation ,
leads to the equation  3   + (  +  + 2 , )3 + ( + ,  ) = 0.

g ( )

i ( )

g (


6.

1 (

1) g

)=

) = 0,
7
, i (
1 (

)=

=0


 ( )   + [  ( ) + 2 ,
1

 ) = ,!

1 (

7
=0

+ (  1) , +  = 0,

1 (

)=

7 1
=0

=   , where , = 1  0
 0 , leads to an equation of the same form:

The substitution

where  


 ( )]  + [ 

1 (

 )+ 

1 (

 )] = 0,

[  ( ) + ( , 1)  ( )].

i ( )

) = 0,
7 1
g 1 ( ) = , i (
=0

1 (

)=

=0

1 (

)=

7 1

=0


, = | 1|  , where , is a root of the quadratic equation
 1
  1 , 2 + (     1 ) , + (  1 = 0, leads to an equation of the same form:
.
.
.
( 1)  1 ( )   + [2(1 , )  1 ( )  ( )]  + [ , ( , 1)  1 ( ) +   1 ( )] = 0,

The transformation =

where

.



7.

1 (

)=

1 ( ) =

+ V

 + + ( 1)  +

 ( ) =

=0

( + + ( 1)  +

=0

 ! (  

+ )

The transformation =  

 

 + + ( 1)  + ,
=0

1 ( ) =

WU = 0.

+  , =  J

1
4




1 (

) + ,  ( ) + , ( , 1) 
1

1 (

  + (  )2  ( ) = 0.
leads to the equation 3

+ ! (  ) + " (  ) = 0.



The substitution H =   leads to the equation 2 H 2 ' + H [  ( H ) + ]  + ( H ) = 0.

9.

11.

+ V

+ sinh 4 (

)!

 coth(

)  W

The transformation = coth(  ), =

+ V

1
4

+ cosh 4 (

 2

(  

)!

= 0.

sinh(  )

leads to the equation 3  +

 tanh( )  WU = 0.
  +
leads to the equation 3
The transformation = tanh(  ), =
cosh(  )

+ )

! O
4




+ P

12.

 ( ) = 0.

 ( ) = 0.

= 0.



   +
  J2
,
=


(  +
(  +
M
( e )2  ( ) = 0, where =  M ( .
M
The transformation =

8.

10.

  +
leads to a simpler equation: u

+ (2 ! tanh + " ) + (" tanh + % ) = 0,


! = ! ( ), " = " ( ), % = % ( ).
The substitution = cosh  leads to a simpler equation: "    + !  + ( %  ) = 0.

2003 by Chapman & Hall/CRC

Page 293

294
13.

14.

15.

SECOND-ORDER DIFFERENTIAL EQUATIONS

+ (2 ! coth + " )

+ (" coth + % ) = 0,

+ V ! ( ln + ) + 14 W = 0.

The transformation =  ln  +  , = 
(

1)2 + I
! O ln

+1 P

[  ( ) 1] = 0.

17.

2!

(ln ) +

+[

"

20.

21.

22.

= % ( ).

 ( ) = 0.

|

leads to a simpler equation: 4u  +

1|

(ln ) + % (ln ) = 0.

The substitution = ln  leads to the equation  ( )" + [ ( )  ( )]  + % ( ) = 0.


2

)!

+ sin 4 (

(cot(

))] = 0.

+ V

)!

+ cos 4 (

 tan(

sin(  )

)  W

  +
leads to a simpler equation: u

 ( ) = 0.

leads to a simpler equation: u  +

 ( ) = 0.

= 0.

cos(  )

sin( + )
= 0.
sin4 (
+ )
sin( + ) P
sin(  +  )
, =
leads to a simpler equation: u  +
The transformation =
sin(  +  )
sin(  +  )
[ sin(   )]2  ( ) = 0.

+ n

!IO

+ (" 2 ! tan ) + ( % " tan ) = 0,

+ (" + 2 ! cot )

= ! ( ),

"

= " ( ),

= % ( ).

The substitution = cos  leads to a simpler equation: "    + !  + (  + % ) = 0.

+ ( % + " cot ) = 0,

= ! ( ),

"

= " ( ),

= % ( ).

The substitution = sin  leads to a simpler equation: "    + !  + (  + % ) = 0.


(

+ 1)2

+ ! (arctan + ) = 0.

[  ( ) + 1] = 0.
(

[  ( ) + 1] = 0.

=


+1

=


+1

+ 1)2 + ! (arccot + ) = 0.

The transformation = arccot  +  ,

24.

  + 
J leads to a simpler equation: u

The transformation = arctan  +  ,

23.

= " ( ),

1 2

 
,
=
 +1

The transformation = tan(  ), =


19.

"

= 0.

The transformation = cot(  ), =


18.

= ! ( ),

The substitution = sinh  leads to a simpler equation: "    + !  + ( %  ) = 0.

The transformation = ln

16.

  +
leads to a simpler equation: u

  +
leads to a simpler equation: u

+ ! ( ) = 0.

| Y  | leads to an equation of the same form: u'  +


The transformation  = Y ( H ), =
'
'

'





1Y
3 Y  2

` ( H ) = 0, where ` (H ) =

O
+ ( Y )2  ( Y ).

2 Y 
4 Y  P

2003 by Chapman & Hall/CRC

Page 294

2.2. AUTONOMOUS EQUATIONS

( ,

295

2.2. Autonomous Equations w = c ( , )


Preliminary remarks. Equations of this type often arise in different areas of mechanics, applied
mathematics, physics, and chemical engineering science.
1 b . The substitution ( ) =  leads to a first-order equation:

 =


 ( , ).

(1)

2 b . The solution of the original autonomous equation can be represented in implicit form:

 =X
where = ( , L

1)

( M , L ) + L 2,
1

(2)

is the solution of the first-order equation (1).

3 b . The solution of the original autonomous equation can be written in parametric form:

 (U , L )
1
* (U , L ) M U + L 2 ,
1

 =X

= (U , L

1 ),

(3)

where = (U , L 1 ), = (U , L 1 ) is a parametric form of the solution of the first-order equation (1).


Formula (2) is a special case of formula (3) with = U .
4 b . For the special cases 

=  ( ), see equations 2.9.1.1 and 2.9.4.35.

=  ( ) and 

2.2.1. Equations of the Form

$  = ( )

Preliminary remarks. Equations of this type arise in the theory of combustion and the theory of
chemical reactors.
1 b . The substitution ( ) =  leads to the Abel equation *3 =  ( ), which is considered in

Subsection 1.3.1 for some specific functions  .
2 b . The solution of the original autonomous equation can be written in the parametric form (3),
where = (U , L 1 ), = (U , L 1 ) is a parametric form of the solution to an Abel equation of the
second kind *  =  ( ).

1.

2(

+ 1)

EA

( + 3)2
2
Solution in parametric form:

 =
=L
2.

= A 2

2 1

+1
2

+3
ln O  L
1

2
1

U O  L

1
1

1
1

A 1,

3.

U
M + +1 + L 2 ,
1 AC
P
U
2
U
+
M + +1 + L 2 1 .
1 AI
P
U

+ 1 X
+3

+ 1 X
+3

Solution in parametric form:

 = ln L
Q

X exp(A"U 2 ) U + L 2 S ,
M

=  L

exp(AU 2 ) L

X exp(AU 2 ) U + L 2 S
M

2003 by Chapman & Hall/CRC

Page 295

296
3.

SECOND-ORDER DIFFERENTIAL EQUATIONS


= 29 + 16
3 ) 2 1 ) 2 .
9
Solution in parametric form:

 = 3 ln , L

exp(U ) V exp(3U ) + L

9
= 100
A 1009 8 ) 3
Solution in parametric form:

5 3

 = 54 ln V A (U
=  (U
5.

3
= 16
643 8 ) 3
Solution in parametric form:

 =L

sin  3 U  W. ,

V 2 exp(3U ) L 2 sin  3 U  + 3 L 2 cos  3 U  W 2


.
V exp(3U ) + L 2 sin  3 U  W 2

=  exp(2U )
4.

3U + L

5 3

2 ln[sin U cosh(U + L

6U
1)

+ 4L

J V A (U

3 2

U 3)W + L 2 ,

6U

+ 4L

U 3)W

9 8

2)

+ cos U sinh(U + L

=  [tan U + tanh(U + L

2 )],

2 )]

J .

3 2

In the solutions of equations 69, the following notation is used:

L 1 (U ) + n (U ) for the upper sign,


L 1  (U ) + ; (U ) for the lower sign,
n (U ) are the Bessel functions, and  (U ) and ; (U ) are the modified Bessel
where (U ) and

functions.
6.

=  1 ) 2 .
Solution in parametric form:

 = 2 XwU

where @ = 13 , # = T
7.

: 
*

(U

 J

1
3

) U +L

= U

2,

J :

4 3

[(U

: 
*

1
3

)2 AIU

],

1 3 2
.
3

=  2 .
Solution in parametric form:

 =T

2
3

XU

: 2V : 
(U
*

1
3

)2 AIU

2 1

= 2 U 4 J

U + L 2,

: 2V : 
(U
*

1
3

)2 AIU

2 1

where @ = 13 , # = 36  3 .
8.

= 2  2  1 ) 2 .
Solution in parametric form:

 = A 2 XU
where @ = 0, #
9.

=  1J 2.

:  1 :
* ) (U A

=  1 ) 2 + 2 ( 2 +
Solution in parametric form:

 = 2 XwU
where # = (1 @ 2 ) $

( 1 )
1

(U

: 
*

: 
* )M U

+L

2,

:  2 :
* ) (U A

= (

:  2
*),

) U +L

2,

=$

(U

: 
*

)2 ,

2003 by Chapman & Hall/CRC

Page 296

2.2. AUTONOMOUS EQUATIONS

In the solutions of equations 1014, the function

U =X

A (4 <

<

1)

( ,

297

) is defined in implicit form:

1.

The upper sign in this formula corresponds to the classical elliptic Weierstrass function
< = < (U + L 1 , 0, 1).
10.

9
=  2 625
 1 .
Solution in parametric form:

= 5  (U 2 <IT

 = 5 ln U + L 2 ,
11.

<

where

# =A

6 1
.
125

= 5  (U 2 <pT 1),

6 1
.
125

# =A

where

= 12 +  5 ) 2 .
Solution in parametric form:

A (4 <

X<

1), # = T 147  7 J 2.

 pA 2U'< 2 

2
7

= -<

U + L 2,

J  pA 2U'< 2 

6 7

4 7

= 63
+  5 ) 3 .
4
Solution in parametric form:

 = 34 X ( pA 2U'< 2 )(U + 2 < )1 U + L 2 ,


M
2
2J 3
.
where  = A (4 < 3 1), # = 128
3  (2  )

= 2 spA 2U'<

 J (U + 2 < )9 J 8 ,

2 3 2

In the solutions of equations 1518, the following notation is used:

=X

+ L 1 (incomplete elliptic integral of the second kind),


A (4U 1)

A (4U 3 1), 1 = 2UlT , 2 = U 1 (2U $T 2 1).

=  1 ) 2 12
.
49
Solution in parametric form:

1 1

 = 7 XwU
16.

15.

= 5 U

6
=  2 + 25
.
Solution in parametric form:

where  =

 =T

14.

6 1
.
125

6
=  2 25
.
Solution in parametric form:

 = 5 ln U + L 2 ,
13.

# =A

where

 = 5 ln U + L 2 ,
12.

1
2 ),

= 6 +  4 .
Solution in parametric form:

 = 15 X U

U + L 2,
M

1 1
1

U + L 2,

= 7 U

= U

2 4

where

3 5 2 5
,
1

# =A

where

12
49 (7

 )1 J 2 .

# = T 150 5.

2003 by Chapman & Hall/CRC

Page 297

298
17.

SECOND-ORDER DIFFERENTIAL EQUATIONS


= 20 +  1 ) 2 .
Solution in parametric form:

1
3

 =
18.

20.

21.

22.

23.

24.

U + L 2,
M

= 15
+  7 .
4
Solution in parametric form:

1  4U 12 T 22  1 M U
=  + ( 1 ( 2
 = Xv

19.

1 1
1 2

XZ

+L
3

2,

= 

4 3 2
2,
1

= 

1 2
1

# = A 108 3 J 2 .

where

J  4U 2 T
1

22  3 J 8 ,

# =A

where

3 8
.
4

The substitution ( ) =   leads to an Abel equation of the form 1.3.1.5:


*  = # + $ 1 $ 2 3 .


3
= 16
+  1 ) 3 + ( 5 ) 3 .
The substitution ( ) =   leads to an Abel equation of the form 1.3.1.61:

*  = 3 + #
16


J +$

5 3

J .

J +$

7 5

1 3

5
= 36
+  3 ) 5 + ( 7 ) 5 .
The substitution ( ) =   leads to an Abel equation of the form 1.3.1.62:

*  = 5 + #
36


J .

3 5

= 94 + 2  2 + 2  2 3 .
The substitution ( ) =   leads to an Abel equation of the form 1.3.1.14:

=  

*  =


/( 

+ /( 

2 2 1

4
9

+ 2#

+ 2#

2 3

The substitution ( ) =   leads to an Abel equation of the form 1.3.1.6:


*  = #  1 , $  + , $ 2 2  1 .


= A

2
2

A 8

Solution in parametric form:

 =T X
where
25.

26.

27.

(

= X exp(T"U 2 ) U + L

2
1,

A 2

) U +L

2,

= A !

(

T 2

),

 = 2U ZA exp(T"U 2 ).

=  + ( exp(2 ).
The substitution ( ) =   leads to an Abel equation of the form 1.3.1.8:

*  = # + $ exp(2
# ).


= 2  2 ] (  + 1) ] + .
The substitution ( ) =   leads to an Abel equation of the form 1.3.1.73:

*  =  2  2    (  + 1)   +  .


= 2  2 ] +  ] +  ] .


The substitution ( ) =   leads to an Abel equation of the form 1.3.1.74:

*  =  2  2   +    +    .


2003 by Chapman & Hall/CRC

Page 298

2.2. AUTONOMOUS EQUATIONS

( ,

299

$ + ( ) + = 0

2.2.2. Equations of the Form


2.2.2-1. Preliminary remarks.

Equation of this form are often encountered in the theory of nonlinear oscillations, where  plays
the role of time.
1 b . The transformation

H = 21

+ ,

= 

leads to an Abel equation:

*  = ( H ) + 1,

where ( H ) =  ( ) 
,

=A

2(  H ),

whose special cases are outlined in Subsection 1.3.2.


2 b . For oscillatory systems with a weak nonlinearity

    +  ( )   + = 0,
two leading terms of the asymptotic solution, as ID

0, are described by the formula

= # cos( + $ ),
where the functions # = # ( ) and $ = $ ( ) depend on the slow variable =  ; they are determined
from the autonomous system of first-order differential equations:

#
#  =

 ( # cos Y ) sin2 Y

1
$  = X

Y ,

 ( # cos Y ) sin Y cos Y

Y .

The right-hand sides of these equations depend only on # . The system is solved consecutively
starting from the first equation.
2.2.2-2. Solvable equations and their solutions.
1.

+ + = 0.
Solution in parametric form:

 = #X
2.

U (L

1 + 2#

+L

2,

=  cos( )

1
32

ln |U | 2 #*U )1 J

= (L

1 +2

ln |U | 2 #*U )1 J 2 ,

where # =

W (1 2 ) + = 0.
Van der Pol oscillator.
1 b . Solution, as ID

0:

where

4
1 + (4 L

2
1

1) 

In applications,  plays the role of time, L


initial phase with = 0.

 3 sin[3( )] + [ ( 2 ),
=

1
7
ln  
8
64

1 2
 +L
16

2.

is the initial oscillation amplitude, and L

is the

2 b . As CD + F , the periodic solution of the Van der Pol equation consists of intervals with
{
fast and slow oscillations and describes damping oscillations with period = (3 2 ln 2) +
[ ( 1 J 3 ).

2003 by Chapman & Hall/CRC

Page 299

300
3.

SECOND-ORDER DIFFERENTIAL EQUATIONS

+ ( 2 + ) + = 0.
1 b . The transformation H = 12
*2 = (2 !H +  ) + 1.

=  leads to an Abel equation of the form 1.3.2.1:

2 b . Solution in parametric form with  < 0:

:
:
1:
2
4
 1 J 2
U +L
, X U 1 J 3 A , 2 U 2 J 3 1 OU  +
S
* 3 P 
3
Q 3
M
:
:
1:
9
4
 1J 2
= A , 2 U 2 J 3 1 OU  +
S
,
 = , 3 ,
* 3 P 
Q 3
4

 =T

where

J U ) + L 2n
1 1 J 3 (U ) + L 2 ;

2,

J U ) for the upper sign,


1 J 3 ( U ) for the lower sign,
1 J 3 ( U ) and n 1 J 3 ( U ) are the Bessel functions, and 1 J 3 ( U ) and ; 1 J 3 ( U ) are the modified Bessel

1 1 3(

1 3(

functions.
4.

5.

+ ( 2 + )2 + = 0.
The transformation H = 21 2 ,
*2 = (2 !H +  )2 + 1.

= 

+ ( 2 + )1 ) 2 + = 0.
1 b . The transformation H = 12 2 ,
*  = (2 !H +  )1 J 2 + 1.

leads to an Abel equation of the form 1.3.2.2:

=  leads to an Abel equation of the form 1.3.2.4:

2 b . Solution in parametric form:

 =  L
where
6.

= exp O X

pO 2 +

2
1

X O L

U
1


 P

+ P

1 2

+ = 0.

1
2

U +

2

= O L

2,

2
1

 1J 2
,
 P

, =  leads to an Abel equation of the form 1.3.2.3:

*  = O# 1 + 1,
#2HrP

+L

U + P

The transformation H =

7.

+ exp( 2 ) + = 0.
The transformation H = 21 2 ,
*2 =  exp(2 H ) + 1.

= 

where

# = 2 .

leads to an Abel equation of the form 1.3.2.7:

8.

+ [ exp( 2 ) + exp( 2 )] + = 0.
The transformation H = 21 2 , =  leads to an Abel equation of the form 1.3.2.8:
*2 = [  exp(2 H ) +  exp(2 H )] + 1.

9.

+ 2 exp[ ( 2 )] + = 0.
Solution in parametric form:

 = T 2 , X  4 , 2 U
where  = T

1
4

ln | ,

1|  1 J 2 U
M

= X exp(T U 2 ) U + L

+L

2,

=   4, 2U

ln | ,

1 |  1 J 2 ,

1.

2003 by Chapman & Hall/CRC

Page 300

2.2. AUTONOMOUS EQUATIONS

10.
11.

12.

+  cosh( 2 ) + = 0.
This is a special case of equation 2.2.2.8 with  =  =

( ,
1
2

+  sinh( 2 ) + = 0.
This is a special case of equation 2.2.2.8 with  =  =

+ 2  sinh2 [  ( ( 2 )] + 2 

Solution in parametric form:

 = 2 lX ( 

+ 2

)4

# .
1
2

# .

+ = 0.

301

U + L 2,
M

= ;

# = 14 

where

13.

= X exp(U 2 ) U + L 1 ,
M
2
= $ 4  arcsinh[!

2  cosh2 [  ( 2

Solution in parametric form:

 = 2 X (

 = 2U + exp(U 2 ),
1

1 (

)] 2 

)4

2 )],

4 = 

+ 8

arcsinh H = ln  H + H

2,

+1 .

+ = 0.

U + L 2,

= ;

# = 14 

where

14.

15.

= X exp(U 2 ) U + L 1 ,  = 2U exp(U 2 ), 4 =  2 + 2 2 8 2  2 ,
M
2
= $ + 4  arccosh[ ! 1  1 (  2 + 2 2 )], arccosh H = A ln H + H 2 1  .

+  cos( 2 ) + = 0.
The transformation H = 12 2 ,
*2 = # cos(2 H ) + 1.

+  sin( 2 ) + = 0.
The transformation H = 12 2 ,
*2 = # sin(2 H ) + 1.

2.2.3. Lienard Equations

= 

leads to an Abel equation of the form 1.3.2.11:

= 

leads to an Abel equation of the form 1.3.2.12:

$ + ( ) + ( ) = 0

2.2.3-1. Preliminary remarks.


Equations of this form are encountered in various fields of applied mathematics, mechanics, and
physics.
1 b . For  ( ) = 0, see equation 2.9.1.1.

2 b . The substitution ( ) =  leads to an Abel equation of the second kind:

*  +  ( ) + ( ) = 0,


whose special cases are outlined in Subsection 1.3.3.


3 b . The transformation ( H ) =  , H = Xv ( )

*  = Y ( H ),

where

leads to an Abel equation of the second kind:

Y (H ) = ( )
 ( ),

whose special cases are outlined in Subsection 1.3.1.

2003 by Chapman & Hall/CRC

Page 301

302

SECOND-ORDER DIFFERENTIAL EQUATIONS

2.2.3-2. Solvable equations and their solutions.


1.

+ + 3 = 0.
Duffing equation. This is a special case of equation 2.9.1.1 with  ( ) = 
1 b . Solution:

 = AuXL

The period of oscillations with amplitude L


integral of the first kind:

4
=
1 +  L

 L

; O
2

2 + 2  L

2 b . The asymptotic solution, as eD

=L

cos[(1 + 38  L

2
1)

4 1 2

12 

+L

 + L 2] +

2.

is expressed in terms of the complete elliptic

; (

where

)=X

0, has the form:

1
32

 L

3
1

cos[3(1 + 38  L

zJ 2

a
1 M sin2 a .

2
1)

 + 3 L 2 ] + [ (  2 ),

where L 1 and L 2 are arbitrary constants. The corresponding asymptotics for the period of
{ = 2 (1 3  L 2 ) + [ ( 2 ).
oscillations with amplitude L is described by the formula:
8
2.

+ + + ^ = 0.
The transformation ( H ) =   , H = 12 
*  = 2  H + ( .
2

3.

4.

7.

8.

leads to an Abel equation of the form 1.3.1.2:

= ( + 3 ) + ^ 3 2 2 2 .
The substitution ( ) =   leads to an Abel equation of the form 1.3.3.1:

*  = ( 


+ 3  ) + (

 

2

= (3 + ) 2 3 2 + ^ .
The substitution ( ) =   leads to an Abel equation of the form 1.3.3.2:
+  ) 

2 3

 

+( .

2 = (7 + 5 ) 3 2 3 2 ^ 2 3 2 .
The substitution ( ) =   leads to an Abel equation of the form 1.3.3.3:
2*  = (7 

6.

*  = (3 


5.

+ 5  ) 3 

2 3

2(

3

= 1 [(1 + 2 ) + ] 2 ( + ).
The substitution ( ) =   leads to an Abel equation of the form 1.3.3.8:

*  = 


[(1 + 2B ) + B ] B

 ( +  ).

= ( ) ) 1 + [ 2 (2 + 1) + ( + 1) 2 ] 2 1 .
The substitution ( ) =   leads to an Abel equation of the form 1.3.3.9:

*  =  ( B ) 


+( [

+ B (B + 1)  2 ] 2 

(2B + 1) 

= [ (2 + )  + ] 1 + ( 2 2   + ) 2 1 .
The substitution ( ) =   leads to an Abel equation of the form 1.3.3.10:

*  = [  (2B + , )  +  ] 


+ (  2 B

    + ( ) 2 

2003 by Chapman & Hall/CRC

Page 302

2.2. AUTONOMOUS EQUATIONS

9.

10.

= [ (2

+ )

+ (2

)]

 1

( ,
2

x 4

303

^ 2

2 ) 2

The substitution ( ) =   leads to an Abel equation of the form 1.3.3.11:


*  = [  (2 + , ) 2  +  (2 , )] +  1 (  2 4  + ( 2  +  2 ) 2 +


]

]

2 1

 .

2 1

Solution in parametric form:

 =

XU

L

+#

ln |U | #*U 

where # = 
 .
11.

= ( ] + ) +

 2]

U + L 2,

ln ( L

Q 

+#

ln |U | #*U )S ,

] 2 .

The substitution ( ) =   leads to an Abel equation of the form 1.3.3.67:

*  = (    +  ) + (  2       2 .

12.

13.

14.

15.

16.

17.

18.

= [ (2
+ )-] + ] 

 ]

+ (

= ( ] + ) + [

 ]

(2  +

Df]

(2

= ( cosh + ) sinh + .

] + ) 

C 2 ]

 ]

The substitution ( ) =   leads to an Abel equation of the form 1.3.3.68:


*  = [  (2 + )   +  ]    + (  2  2      + ( )  2   .


 ]

2 2

+ 1) ] +

].

The substitution ( ) =   leads to an Abel equation of the form 1.3.3.69:


*  = (    +  ) + ( [  2  2  +   ( + 1)   +  2 ].


+ )

] ( 2

+ ).

The substitution ( ) =   leads to an Abel equation of the form 1.3.3.70:


*  =   (2  +  +  )  2  (  2 2 +   + ( ).

2

+ 2 + ) +

 2fD ]

+ ).

The substitution ( ) =   leads to an Abel equation of the form 1.3.3.71:


*  =   (2  2 + 2 +  ) +  2  (  4  2 + ( ).

The substitution ( ) =   leads to an Abel equation of the form 1.3.3.75:

*  = (  cosh +  )   sinh + ( .


= ( sinh + ) cosh + .

The substitution ( ) =   leads to an Abel equation of the form 1.3.3.76:

sin = 0.

*  = (  sinh +  )   cosh + ( .


This is the equation of oscillations of the mathematical pendulum, where the variable  plays
the role of time, and is the angle of deviation from the equilibrium state.

2003 by Chapman & Hall/CRC

Page 303

304

SECOND-ORDER DIFFERENTIAL EQUATIONS


1 b . Solution:

 = AuX (2  cos + L 1 )1 J

+L

2.

2 b . With  > 0 and the initial conditions (0) = L > 0 and " (0) = 0, the oscillations of the
mathematical pendulum are described by

= sn    ,
= sin .

2
2
where sn = sn( H ) is the Jacobi elliptic function defined parametrically by the following
relations:
:

sn( H ) = sin , H = X
M 2 2 .
0
1
sin

3 b . The period of oscillations of the mathematical pendulum is expressed in terms of the


complete elliptic integral of the second kind:
sin

4
=


; (

At small amplitudes, as LD

19.

),

zJ 2

)= X

sin2

0, the following asymptotic formula holds for the period:

2
1
=  O 1+
L
16


+ [ (L

L D

),

0.

+ sin( ) + sin( ) = 0.
Solution in parametric form:
 = #Xa 1   2 2 
where # = 
 , 

20.

; (

where

= #*a #

ln |a | + L

2 1 2

a + L 2,

arccos O


P

1.

+ cos( ) + cos( ) = 0.
The substitution = + z 2 leads to an equation of the form 2.2.3.19:
    sin( )   sin( ) = 0.

2.2.4. Rayleigh Equations

$ + ( ) + ( ) = 0

2.2.4-1. Preliminary remarks. Some transformations.


Equations of this form arise in the theory of nonlinear oscillations.
1 b . Let us discuss the special case ( ) = , which corresponds to the equation

    +  (   ) + = 0.

(1)

Differentiating equation (1) with respect to  and substituting H ( ) =  , we obtain the equation of

nonlinear oscillations:

H    + ` ( H ) H   + H = 0,

where

` (H ) =   ( H ),

(2)

which is considered in Subsection 2.2.2.


The solution of equation (1) can be written in parametric form:

 =  (U , L 1 , L 2 ),
where  =  (U , L

1,

=  (H )

H 
* ,
 

L 2 ), H = H (U , L 1 , L 2 ) is a parametric representation of the solution of equation (2).

2003 by Chapman & Hall/CRC

Page 304

2.2. AUTONOMOUS EQUATIONS

2 b . The transformation

= 21 (   )2 +  ,

( ,

305

=  (   ),

reduces equation (1) to an Abel equation of the second kind:

*  = 5 ( ) + 1, where 5 ( ) = H 1 ` ( H ), H = A 2(  ),
(3)

where function ` = ` ( H ) is defined above in equation (2). Specific equations of the form (3) are
outlined in Subsection 1.3.2.

3 b . The equation of the special form

    +  (   )2 + ( ) = 0
(4)

"
u


2
+ 2 +
is reduced, with the aid of the substitution ( ) = (  ) , to the first-order linear equation

2 ( ) = 0. Therefore, the solution of equation (4) can be written in implicit form:
1 J 2
 = L 2 A X V L 1  2  4 ( )W
,
where 4 ( ) = 2  2  XZ 2  ( ) .
M
M
4 b . The equation of the special form
   +  (   )4 +  (   )2 + ( ) = 0
(5)

"
3


2
2

+ 2
+ 2 +
is reduced, with the aid of the substitution ( ) = ( ) , to the Riccati equation

2 ( ) = 0, which is outlined in Section 1.2.
5 b . For the oscillatory systems with a weak nonlinearity
   +  (   ) + = 0,
two leading terms of the asymptotic solution, as ID 0, are described by the formula
= # cos( + $ ),
where the functions # = # ( ) and $ = $ ( ) depend on the slow variable =  and are determined
from the autonomous system of first-order differential equations:
1 2z
1 2z
X
X
#  =
 ( # sin Y ) sin Y Y , #2$  =
 ( # sin Y ) cos Y Y .
2 0
2 0
M
M


The right-hand sides of these equations depend only on # . The system is solved consecutively
starting from the first equation.
2.2.4-2. Solvable equations and their solutions.
1.

2.

+ ( )2 + = 0.
This equation describes small oscillations in the case where the drag force is proportional to
the speed squared.
1 J 2
.
Solution in implicit form:  = L 2 AClX V L 1  2  2  +  ( 21  )W

+ W V 31 ( )3 W + = 0.

Van der Pol equation.

1 b . Differentiating the equation with respect to  and passing on to the new variable ( ) = / ,
we arrive at an equation of the form 2.2.2.2: u   (1 2 )2 + = 0.
2 b . Solution, as ID

3.

0:
2L
=
1L

1
2

2 1L

 cos  +
1 L 2

2
1

sin  + [ ( 2 ).

+ ( )4 + ( )2 + = 0.
The transformation = 12 (  )2 , =  (  )4  (  )2 leads to an Abel equation of the
form 1.3.2.1: *3  = (8  + 2  ) + 1.

2003 by Chapman & Hall/CRC

Page 305

306
4.

SECOND-ORDER DIFFERENTIAL EQUATIONS

+ (

)2 V (

)2 + W

+ = 0.

 )2 , = (  )2 [  (  )2 +  ]1 leads to an equation of the


The transformation =
*

= 2  (  2  )2 + 1.
form 1.3.2.2:
21 (

5.

+  exp[ ( )2 ] + ( + = 0.
Differentiating the equation with respect to  and passing on to the new variable ( ) = / ,
we arrive at an equation of the form 2.2.2.7: u   + 2 #3 exp( 2 )2 + = 0.

6.

+  cosh[ ( )2 ] + ( + = 0.
Differentiating the equation with respect to  and passing on to the new variable ( ) = / ,
we arrive at an equation of the form 2.2.2.11: p   + 2 #3 sinh( 2 )2 + = 0.

7.

+  sinh[ ( )2 ] + ( + = 0.
Differentiating the equation with respect to  and passing on to the new variable ( ) = / ,
we arrive at an equation of the form 2.2.2.10: p   + 2 #3 cosh( 2 )2 + = 0.

8.

+ ( )2 + sin = 0.
This equation describes the oscillations of the mathematical pendulum in the case where the
drag force is proportional to the speed squared.
1 J 2
2
(cos 2  sin )S
.
Solution in implicit form:  = L 2 AX L 1  2  + 2
4 + 1
M
Q

9.

+  cos[ ( )2 ] + ( + = 0.
Differentiating the equation with respect to  and passing on to the new variable ( ) = / ,
we arrive at an equation of the form 2.2.2.15: p   2 #3 sin( 2 )2 + = 0.

10.

+  sin[ ( )2 ] + ( + = 0.
Differentiating the equation with respect to  and passing on to the new variable ( ) = / ,
we arrive at an equation of the form 2.2.2.14: p   + 2 #3 cos( 2 )2 + = 0.

2.3. EmdenFowler Equation w =


2.3.1. Exact Solutions
2.3.1-1. Preliminary remarks. Classification table.
In this subsection, the value of the insignificant parameter # is in many cases defined in the form of
a function of two (one) auxiliary coefficients  and  ,

# = Y (  ,  ),

(1)

and the corresponding solutions are represented in parametric form,

 =  1 (U , L 1 , L 2 ,  ),

=  2 (U , L

1,

L 2 ,  ),

(2)

where U is the parameter, L 1 and L 2 are arbitrary constants, and  1 and  2 are some functions.
Having fixed the auxiliary coefficient sign  > 0 (or  > 0), one should express the coefficient 
in terms of both # and  with the help of (1). As a result, one obtains:

 = j ( # ,  ).
Substituting this formula into (2), we find a solution of the equation under consideration (where
the specific numerical value of the coefficient  can be chosen arbitrarily). The case  < 0 (or

2003 by Chapman & Hall/CRC

Page 306

2.3. EMDENFOWLER EQUATION

TABLE 21
Solvable cases of the EmdenFowler equation

307

No

Equation

No

1
2
3
4
5

0
3

21 ( + 3)

arbitrary
arbitrary

arbitrary
arbitrary
arbitrary
0
1

2.3.1.2
2.3.1.3
2.3.1.4
2.3.1.1
2.3.1.5

Isolated points
7
7
5
2
2
2
5
3
5
3

6
7
8
9
10
11
12

1
3
1
2
2
1
10
3
7
3

5
3
5
3
5
3
5
3
7
5
7
5
1
2
1
2
1
2
1
2
1
2
1
2
1
2
2
2
2

13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28

One-parameter families

2.3.1.15
2.3.1.16
2.3.1.22
2.3.1.28
2.3.1.27
2.3.1.10
2.3.1.8

   = #*  +
B

Equation

5
6
1
2
1
2
13
5
1
7
2
5
2
2
4
3
7
6
1
2
1
5
20
7
15
7

2.3.1.23
2.3.1.24
2.3.1.7
2.3.1.9
2.3.1.14
2.3.1.13
2.3.1.12
2.3.1.6
2.3.1.26
2.3.1.17
2.3.1.18
2.3.1.25
2.3.1.11
2.3.1.19
2.3.1.21
2.3.1.20

 < 0), which may lead to another branch of the solution or to a different domain of definition of the
variables  and in (2), should be considered in a similar manner.
One can also use a different approach by setting one of the auxiliary coefficients (e.g.,  ) equal
to A 1 in (1) and (2); then the other coefficients will be identically expressed in terms of # by means
of (1).
Table 21 presents all solvable EmdenFowler equations whose solutions are outlined in Subsection 2.3.1. The one-parameter families (in the space of the parameters B and ) and isolated points

are presented in a consecutive fashion. Equations are arranged in accordance with the growth of

and the growth of B (for identical ). The number of the equation sought is indicated in the last

column in this table.

2.3.1-2. Solvable equations and their solutions.


1.

Solution:

2.





#* 

+ L 1 + L
(B + 1)(B + 2)
# ln | | + L 1  + L 2



#X ln | |  + L 1  + L
M

Solution:  =

+2




AuXO

2#

+1

+1

A X (2 # ln | | + L

if B = 2;
if B = 1.

1 2

+L
1)

if B 1, B 2;

1 2

+L

+L

1,

if
if

= 1.

2003 by Chapman & Hall/CRC

Page 307

308

SECOND-ORDER DIFFERENTIAL EQUATIONS


Special cases.

= 1
2, the solution can be written in the parametric form:

1 b . In the case

 =  L 13 (U

2 b . In the case

3U + L

A (4U

1)2 ,

 J

4 2 3 2
.
9

# =A

where

= 4, the solution can be written in the parametric form:

 =  L
where =

4 2
1(

= L

2 ),

3 b . In the case

5 1
1

O 2U*X

1), # = T 6

+L

U3T

2 1
,
1

= L

 .

2 5

= 2, the solution can be written in the parametric form:

1
1

 =  L

U ,

2
1

= L

<

2 1

# = A 6

the function < of the parameter U is defined in implicit form:

U =X

A (4 <

<

1)

2.

The upper sign in this formula corresponds to the classical elliptic Weierstrass function
< = < (U + L 2 , 0, 1).
4 b . In the case

<

= 5
2, the solution can be written in the parametric form:

7 2
1

 =  L

A (4 <

1) A 2U'< 2 W ,

<

4 2
,
1

= L

 J .

2 7 2

# = T 3

where

The function < of the parameter U is defined in the previous case.


3.

=  3 .
1 b . Solution in parametric form with

 =  L 1+

X (1 AIU +

) J

+1 1 2

1:

U + L 2S

+ 1 + +1 1 +


.
2
2 b . Solution in parametric form with

= L 1+

+1

U X (1 AIU +
Q

) J

+1 1 2

U + L 2S

where # = A

 =L

= 1:

X exp(T"U 2 ) U + L 2 S
Q
M

=  exp(T U 2 ) X exp(T"U 2 ) U + L

where # = T 2  2 .
4.

= 2 .
1 b . Solution in parametric form with

+3

8
U +
Q
+1
8
=  L 2 U exp XO
U +
Q
+1

 =  L

where # = O

 2P

1
2

2
2

exp 2 X O

2
2

exp 2 Xp 8 ln |U | + U

where # =  2
 .

+1
+1

+U

+U

2
2

+L

+L

1 2
1
1

1 2

U!S ,
U!S ,

2 b . Solution in parametric form with

 =  L

1:

+L

= 1:

1 2

U!S ,

= L

U exp I
X  8 ln |U | + U
Q

+L

1 2

U!S ,

2003 by Chapman & Hall/CRC

Page 308

2.3. EMDENFOWLER EQUATION

5.
6.

309

=  5 ) 2 1 ) 2 .
Solution in parametric form:

3 3
1 (

3U + L

2)

= L

1(

1)2 (U

3U + L

2)

where

# =A

=  5 ) 3 .
Solution in parametric form:

 J  J

4 1 2 3 2
.
9

6U

+ 4L

U 3),

= L

9 3
1(

3U + L

2)

J ,

3 2

# =A

where

=  7 ) 3 5 ) 3 .
Solution in parametric form:

 J

9 3 8 3
.
64

 =A
9.

= .
For B 2, see equation 2.1.2.7. For B = 2, see equation 2.1.2.123.

 = A  L 18 (U
8.

 =  L
7.

6U

 L

8
1

+ 4L

U 3

=A

L 1 (U 3 3U + L 2 )3 J 2
,
U 4 6U 2 + 4 L 2 U 3

where

# =A

 J  J

9 1 3 8 3
.
64

=  2 5 ) 3 .
1 b . Solution in parametric form with # < 0:

 =  L

2
1

cos U cosh(U + L

3

where # = 16

2 )[tan

U + tanh(U + L 2 )],

= L

3
1 [cos

U cosh(U + L 2 )]3 J 2 ,

 J .

4 8 3

2 b . Solution in parametric form with # > 0:

 =  L 12 [sinh U + cos(U + L 2 )],


where # = 34 
10.

= L

3
1 [cosh

U sin(U + L 2 )]3 J 2 ,

 J .

4 8 3

=  10 ) 3 5 ) 3 .
1 b . Solution in parametric form with # < 0:

 =  L 12 [cos U cosh(U + L 2 )]1 [tan U + tanh(U + L 2 )]1 ,


= L 1 [cos U cosh(U + L 2 )]1 J 2 [tan U + tanh(U + L 2 )]1 ,
3 4J 3 8J 3
  .
where # = 16

2 b . Solution in parametric form with # > 0:

 =  L

2
1 [sinh

U + cos(U + L 2 )]1 ,

= L

1 [cosh

U sin(U + L 2 )]3 J 2 [sinh U + cos(U + L 2 )]1 ,

where # = 34  4 J 3  8 J 3 .
11.

=  1 ) 2 .
Solution in parametric form:

 =  L

= L

2
1

where # = 16

exp(U ) V exp(3U ) + L

exp(2U ) V 2 exp(3U ) L

sin  3 U  W ,
2

sin  3 U  + 3 L

2
cos  3 U  W ,

 J .

3 3 2

2003 by Chapman & Hall/CRC

Page 309

310

SECOND-ORDER DIFFERENTIAL EQUATIONS

In the solutions of equations 1214, the following notation is used:

8
8

12.

= exp(3U ) + L

=2

1( 2)

 (

sin  3 U  ,

 8
1)
*

=  7 ) 2 1 ) 2 .

8 8

= 2 exp(3U ) L

13.

1
1
7 5

1
1 ,

= L

4
1

exp(2U )

5
1

= L

3,

=  13 ) 5 7 ) 5 .
Solution in parametric form:

 =  L
]

= ) .
Solution in parametric form:

 =  L

14.

exp(U )

4
1

exp(2U )

sin  3 U  + 3 L

cos  3 U  ,

1 2.

Solution in parametric form:

 =  L

1
3 ,

= L

exp(U )

1 2
1
2,

where

8 5J 2
,

exp  25 U 

where

8 5J 2 8

exp  12 U 

1
3 ,

# = 16(  )3 J 2.

# =

where

 J

3 12 5
5
.
1024

# =

 J  J

3 5 12 5
5
.
1024

In the solutions of equations 1518, the following notation is used:

 = 2U (U ) + L 2 2
U T ,

(U ) = X

A (4U 3 1),

where (U ) is the incomplete elliptic integral of the second kind in the form of Weierstrass.
15.

16.

17.

=  7 .
Solution in parametric form:

 =  L 18 [4U

TCU

 =  L 18 [4U 2 TU
3 5 8
  .
where # = A 64

=  3 7 .
Solution in parametric form:

= L

(  1)2 ]1 ,

 J

3 1 2
,
1

 J

5 1 2
[4
1

= L

# =A

where

U

TCU

3 3 8
.
64

(  1)2 ]1 ,

=  4 ) 3 1 ) 2 .
Solution in parametric form:
9 3
,
1

=  7 ) 6 1 ) 2 .
Solution in parametric form:

4 2
(
1

= L

 1)2 ,

# =A

where

 =  L
]

(  1)2 ],

 =  L

18.

9 3
,
1

= L

5 2 3
(
1

 1)2 ,

In the solutions of equations 1924, the function

U =X

<

where

J  J

4 2 3 3 2
.
3

# =A

J  J

4 5 6 3 2
.
3

) is defined in implicit form:

L 2.
1)

The upper sign in this formula corresponds to the classical elliptic Weierstrass function
< = < (U + L 2 , 0, 1).
19.

A (4 <

=  5 2 .
Solution in parametric form:

 =  L 1 U

= L

3 1
1

<

where

# = A 6 3 

2003 by Chapman & Hall/CRC

Page 310

2.3. EMDENFOWLER EQUATION

20.

=  15 ) 7 2 .
Solution in parametric form:
7 7
,
1

7 7
,
1

= L

=  1 ) 2 5 ) 2 .
Solution in parametric form:

U (U 2 p
< T 1),

 J 

6 1 7 1
.
49

# =A

where

U <IT 1),

6 6 2
(
1

 J 

6 6 7 1
.
49

# =A

where

< V A

7 2
1

(4 <

1) A 2U'< 2 W

3
1

= L

J  J .