PARTIAL DIFFERENTIAL EQUATIONS

1. FORMATION OF PARTIAL DIFFERENTIAL EQUATIONS 1.1 Elimination of Arbitrary Constants (1)

Let the given equation be f(x,y,z,a,b) = 0 1.1.1 METHOD

f =0 a f =0 B

1. Differentiate Equation (1) partially with respect to a, 2. Differentiate Equation (1) partially with respect to B,

(2) (3)

3. Eliminate a & b from Eqns. (1), (2) & (3), we get F(x,y,z,p,q) = 0 1.1.2 PROBLEMS

Eliminate arbitrary constants from the following equations and form partial differential equations 1) z = (x2 + a2) (y2 + b2) 2) 3)
log z = a log x + z = xy + y 1 - a 2 log y +b

x 2 a 2 +b

4) z = ax 5)

a y +b a +1

1 + a 2 log z + z 2 1 = x + ay + b

1.2 1.2.1

Elimination of Arbitrary Functions Type I f(u,v) = 0 or u = f(x,y,z), where u and v are functions of x&y.

1.2.1.2 METHOD Let f(u, v) = 0 be given

, , and , , 1) Compute x y z x x z u u u v v v

(1)

 u y 2) Obtain P = v y

u u z Q = z v , v z z

u u x , R = x v v x x

u y v y

3) The solution is given by P p + Q q = R. 1.2.1.2 PROBLEMS Eliminate arbitrary functions from the following equations and form partial differential equations 1) z = x + y =f(xy) 2) z = xy + f(x2 +y2 + z2) 3) xy + yz + zx = f x+y 4) f xy + yz + zx, y z =0
xy z

5) f ( x 2 + y 2 + z 2 , xy + yz + zx ) = 0 1.2.2 TYPE II (1) z =F(x, y, z, f, )

1.2.2.1 METHOD 1) Differentiate Eqn.(1) partially with respect to x, p = F1(x,y,z,f,f,,) = 0 2) Differentiate Eqn.(1) partially with respect to y, q = F2(x,y,z,f,f,,) = 0 3) Differentiate Eqn.(2) partially with respect to x, r = F3(x,y,z,f,f,,, f,) = 0 4) Differentiate Eqn.(2) partially with respect to y, s = F4(x,y,z,f,f,,,f,) = 0 5) Differentiate Eqn.(3) partially with respect to y, t = F5(x,y,z,f,f,,,f,) = 0 6) Eliminate f & g from Eqns.(1) to (6), we get G(x,y,z,p,q,r,s,t) = 0 1.2.2.2 PROBLEMS Form p.d.e by eliminating arbitrary functionsfrom the following equations. 1) z = f(y) + (x + y + z) 2) z = xf + y( x)
y x

(2) (3) (4) (5) (6)

3) z = f(x2 + y) + (x2 y)

4) z = f(x + y + z) + (x y) 5) z = f(ax + by) + g(cx + dy)

2.

SOLUTIONS OF FIRST ORDER PARTIAL DIFFERENTIAL EQUATIONS

A general first order homogeneous differential equation is of the form F(x,y,z,p,q) = 0 2.1 STANDARD I 2.1.1 METHOD (1) F(p, q ) = 0

1) Assume the trial solution of F(p, q) = 0 as z = ax + by + c where F(a, b) = 0 2) Express a = (b) or b = (a) 3) Substitute a or b in Eqn(1), the Complete solution is z = (b)x + by + c or z = ax + (a)y + c. (3) 4) Put c = f(b) Eqn (2) becomes z = (b)x + by + f(b)

(2)

(4)

5) To get general solution differentiate Eqn.(4) w.r.to b and eliminate the constant b.

2.1.2

PROBLEMS

Solve the following equations

1)
p + q =1

2) p + q = pq 3) pq = a 4) p2 + q2 = npq 5) p + pq = 1

2.2

STANDARD II

z = px + qy + f(p, q) = 0

(1)

2.2.1

Method (2)
(4)

1) Put p =a & q = b, we get the complete integral as z = ax + by + f(a, b) = 0.

2) Differentiate Eqn.(2) w.r.to a & b, we get x + f =0 a

(3) y +

f =0 b

3) Express Eqn. (3) as x = F(a, b) 4) Express Eqn. (4) as y = G(a, b) 5) Eliminate a& b from Eqns.(3) to (6), we get singular solution.

(5) (6)

2.2.2

PROBLEMS

Solve the following equations 1) z = px + qy + pq 2) z = px + qy + p2q2 3) z = px + qy + p2 q2 4) z = px + qy + p2q + pq2 5)

z = px + qy + 1 +p 2 +q 2

2.3 2.3.1

STANDARD III METHOD

f(z, p, q) = 0

(1)

1) Assume the trial solution as z = f(u). 2) Put u = x + ay, so that p =

dz dz , q =a in Eqn. (1) du du dz dz ,a =0 du du

3) Then Eqn. (1) becomes f z ,

4) Separate du on one side and g(z)dz on other side 5) Integrate on both sides we get ux + b = F(z) 6) The complete solution is x + ay + b = F(z)

2.3.2

PROBLEMS

Solve the following equations 1) p(1 + q) = qz2 2) q(1 + p2) = pz 3) z2 = 1 + p2 + q2 4) 16(p2z + q2) = 25 5) z = p2 +q2 6) p2z2 + q2 = p2q 7) q2 = z2p2(1 p2) 8) p2 + q = z

2.4 STANDARD IV 2.4.1 METHOD

f(x,y,p,q) = 0

(1)

1) Separate x & p on one side and y & q on other side, so that F(x, p) = G(y, q). (2) 2) Let f1(x, p) = f2(y, q) = a 3) Rewrite the above equation as p = f1(x, a) and q = f2(y, a)
4) Then the complete integral is z = pdx + qdy

(3)

2.4.2 PROBLEMS Solve the following equations 1) 2) 3)

4) 5)

p2y(1 + x2) = qx2 p x2 = q + y2 p + q = sin x+ sin y

yp = 2yx + log q p2 + q 2 = x + y

2.5

EQUATIONS REDUCIBLE TO STANDARS FORMS

2.5.1 TYPE I 2.5.2 METHOD CASE I If m 1 & n 1

f(xmp, ynq) = 0(1)

f(z, xmp, ynq) = 0

(2)

1) Put X = x1-m, Y = y1-n .

2) Then xmp = P(1 m) and ynp = Q(1- n), where P = z z , Q= X X

3) Then Eqn. (1) becomes STANDARD I problem and Eqn. II becomes STANDARD III problem CASE II If m = n = 1. 1) Put X = log x & Y = log y 2) Then P = xp & Q = yq 3) Now Eqn.(1) reduces to STANDARD I problem and Eqn. (2) reduces to STANDARD II problem.

2.5.3

PROBLEMS

Solve the following equations 1) x2p + y2q = 0 4) xp +yq = z 7) x2p2 + y2q2 = z2 2.5.3 TYPE II m1 2) xp +yq = 0 5) x4 p2 +y2 zq = 2z2 8) z2(p2x2 +q2) = 1 f(zmp, zmq) = 0 (1) f(x, y, zmp, zmq) = 0 (2) 3) x2p + y2q = z 6) p2 + x2y2q2 = x2z2

CASE I

1) Put Z = zm+1
m 2) Then z p =

P Q & z mq = m +1 m +1

3) Now Eqn.(1) reduces to STANDARD I problem and Eqn.(2) reduces to STANDARD IV problem. CASE II If m = -1

1) Put X = log z
2) Then P = p q & Q= z z

3) Now Eqn.(1) reduces to STANDARD I problem and Eqn.(2) reduces to STANDARD IV problem. 2.5.4 PROBLEMS

Solve the following equations 1) z2(p2 + q2) = x2 + y2 3) z(p2 - q2) = x2 - y2 5) 4zq2 = y + 2zp x 2.6 LAGRANGES LINEAR EQUATION An equation of the form pP + qQ = R, where P, Q, R are functions of x,y&z is called Lagranges equation. 2.6.1 Type I Grouping Method 2.6.1.1 METHOD 1) Form the auxiliary equation P = Q = R
dx dy dz

2) (p2 + q2) = z2( x2 + y2) 4) (zp + x)2 + (zq +y)2 = 1

2) Take any two equations which do not contain the third variable. 3) Let the solutions be a = u(x, y, z) & b = v(x, y, z) 4) The general solution is (u, v) = 0 PROBLEMS Solve the following 1) px + qy = z 2) p q = log(x + y) 3) p yz + q xz = xy

4) p x2 + q y2 = z2

5) p y2z + q x2z = xy2

6)

y2z p + xz q = y 2 x

2.6.2 Type II Multiplier Method 2.6.2.1 METHOD 1) Form the auxiliary equation P = Q = R
dx dy dz

2) Choose multipliers l, m, n & l, m, n and write the above equation as

dx dy dz ldx + mdy + ndz l 'dx + m ' dy + n ' dz = = = = ' . P Q R lP + mQ + NR l P + m 'Q + N ' R

3) Choose mulitipliers so that lP + mQ + NR = 0 , so that ldx + mdy + ndz = 0 or derivative of the denominator is in the numerator, so that on integrating we get v = log( Dr). 4) Obtain u = a & v= b, the general solution is (u, v ) = 0. Note: Usually the multipliers are in the following pairs: 1, 1, 1; x, y, z; 1/x, 1/y, 1/z or the pairs with sign changed. 2.6.2.2 PROBLEMS

Solve the following equations 1) (y2 + z2)p xy q + xz = 0 2) (x2- y2 z2) p + 2xy q = 2 xz 3) z(x y) = px2 q y2 4) x(y z) p + y(z x) q = z( x y) 5) (x2 yz)p + (y2 xz) q = (z2 xy) 6) x(y2 z2) p + y(z2 x2) q = z(x2 y2) 7) (y + z) p + (z + x) q = x + y 8) (x + y)(y + 2z) q (x + y)(x + 2z) p = z(y x) 9) (y + xz) p q(yz + x) = y2 x2 10) x( y2 + z ) p y(x2 + z) q = (x2 y2)z 11) x2(y z) p + y2(z x) q = z2( x y)

3. SOLUTION OF HIGHER ORDER DIFFERENTIAL EQUATIONS WITH CONSTANT COEFFICIENTS A general higher order partial differential equation with constant coefficient is of the form. a0 Dn + a1 Dn-1 D + a2 Dn-2 D2+.+an Dn)z = F(x, y) or f(D, D) = F(x, y) 3.1 METHOD 1) Compute Complementary Function (C.F) using 3.1.1. 2) Obtain Particular Integral (P.I) using 3.1.2. 3) The general solution is z = C.F + P.I 3.1.1 Method of finding C.F (1) (1)

1) Replace D by m and D by1 in f(D, D) = 0 2) Form auxiliary equation f(m) = 0, (2) which is a polynomial of degree m.

3) Let the roots of Eqn.(3) be m1, m2, m3,..mn. 4.1) Let all the roots m1, m2, m3,..mn are real and distinct, then the C.F = 1(y + m1x) + 2(y + m2x) + . n(y + mnx) 4.2) Let m1 = m2 = m, m3, m4, are real and distinct, then the C.F = 1(y + mx) + x 2(y + mx) +3(y + m3x) + . n(y + mnx) Method of Finding P.I
P.I = F ( x, y ) ( D, D ' )

Case 1 F(x, y ) = eax + by 3.1.2.1 METHOD

1) Replace D by a and D by b, if (a, b) 0.

2) If (a, b) = 0, then
P.I = e ax +by a D D' b
r

x r e ax +by r!

3.1.2.2

PROBLEM

Solve the following equations 1. (D2 + 5D D + 6 D2)z = e2x + y 2. (D3 3 D D2 + 2D3) = e2x y + ex + y 3. (D2 D D 6 D2)z = e3x + y 4. (D3 -3D2 D + 3 D D2 + 1)z = ex + y 5. (D2 + 7 D D + 14 ) z = e2x - y Case 2. 3.1.2.3 F(x, y ) = xr ys METHOD 1) Make the denominator as (1 (D, D)). 2) Rewrite P.I as P.I = (1 (D, D))-1xr ys 3) Expand (1 (D, D))-1 as a series of powers of D & D. 4) Operate with xr ys. 3.1.2.4 PROBLEMS Solve the following equations 1) ( D2 + D D - 12 D2)z = x2y 2) (D2 4 D D + 4 D2) = x2 + y2 3) (6D2 + D D D3) z = xy 4) (D2 + 5 D D + 6 D2) z = xy2 5) (2D2 + 5 D D + 6D3) z = x + y Case 3. 3.1.2.5 F(x, y) = sin (ax +by) or cos(ax + by) METHOD

1. Rewrite the denominator as (D2, DD,D2). 2. Replace D2 by a2, D2 by b2 and DD by ab, if (-a2, -ab, -b2) 0

2 3. If (-a2, -ab, -b2) = 0, then Denominator has a factor of the form D

a2 2 D' b2

In this case

P.I =

cos (ax + by ) 2 a 2 D b 2 D'

2

x sin(ax + by ) 2a

Or

P.I =

sin(ax + by ) x cos(ax + by ) = 2 2a 2 a 2 D b 2 D'

3.1.2.6

PROBLEMS

Solve the following equations 1. (D3 4 D2 D + 4 D D2) z = sin (3x 4y) 2. (D3 7 D D2 6 D3) z = cos (2x y) 3. (4D2 9 D2 ) z = sin ( 3x 2y)
4.

( D 2 DD ' ) z = cos x cos 2 y.

5. (D2 + 2 D D + 2D2) z = cosx

CASE 4

F(x, y ) = eax + by (x, y) 3.1.2.7 METHOD

1. Replace D by D +a and D by D + b. 2. Proceed as in Case 2 or Case 3. 3.1.2.8 PROBLEMS 1. ( D 3 + D 2 D 'DD'2 D '3 ) z = e xCos 2 y. 2. (D D D D ) z = (y -1) e
2 2 2 2 x

3. (D + DD + 6 D ) z = y cos x 4. ( D + 2 D D + D) z = ( 2x + 3y)e
2 2 y

5. (D + D ) z = (xy + e )

x 2

CASE 5 3.1.2.4.

P.I =

F ( x, y ) D mD'

METHOD 1. Replace y by y mx in f(x, y) and integrate w.r.to x, treating y as constant. 2. In the resulting integral replace y by y + mx.