Transfer Function Models

of Dynamical Processes

Process Dynamics and Control

Linear SISO Control Systems

General form of a linear SISO control system:

this is a underdetermined higher order differential

equation
the function
must be specified for this ODE to
admit a well defined solution

Transfer Function

Heated stirred-tank model (constant flow,

Taking the Laplace transform yields:

or letting

Transfer functions
3

Transfer Function

Heated stirred tank example

+
+

e.g. The block

is called the transfer function relating Q(s) to T(s)

Process Control
Time Domain

Laplace Domain

Process Modeling,
Experimentation and
Implementation

Transfer function
Modeling, Controller
Design and Analysis

Ability to understand dynamics in Laplace and

time domains is extremely important in the
study of process control

Transfer functions

Transfer functions are generally expressed as a ratio of

polynomials

Where

The polynomial
Roots of
Roots of

is called the characteristic polynomial of

are the zeroes of
are the poles of

Transfer function
Order of underlying ODE is given by degree of
characteristic polynomial
e.g. First order processes

Second order processes

Order of the process is the degree of the characteristic

(denominator) polynomial
The relative order is the difference between the degree of the
denominator polynomial and the degree of the numerator
polynomial
7

Transfer Function

Steady state behavior of the process obtained form the

final value theorem
e.g. First order process

For a unit-step input,

From the final value theorem, the ultimate value of

is

This implies that the limit exists, i.e. that the system is stable.

Transfer function

Transfer function is the unit impulse response

e.g. First order process,

Unit impulse response is given by

In the time domain,

Transfer Function

Unit impulse response of a 1st order process

10

Deviation Variables
To remove dependence on initial condition
e.g.

Compute equilibrium condition for a given

and

Define deviation variables

Rewrite linear ODE

or

11

Deviation Variables
Assume that we start at equilibrium

Transfer functions express extent of deviation from a given

steady-state

Procedure
Find steady-state
Write steady-state equation
Subtract from linear ODE
Define deviation variables and their derivatives if required
Substitute to re-express ODE in terms of deviation variables
12

Process Modeling

Gravity tank

Fo

h
Objectives: height of liquid in tank
Fundamental quantity: Mass, momentum
Assumptions:

F
L

Outlet flow is driven by head of liquid in the tank

Incompressible flow
Plug flow in outlet pipe
Turbulent flow

13

Transfer Functions
From mass balance and Newtons law,

A system of simultaneous ordinary differential equations results

Linear or nonlinear?

14

Nonlinear ODEs
Q: If the model of the process is nonlinear, how do we
express it in terms of a transfer function?
A: We have to approximate it by a linear one (i.e.Linearize)
in order to take the Laplace.

f(x)

f(x0)

!f
( x0 )
!x

x0

15

Nonlinear systems

First order Taylor series expansion

1. Function of one variable

2. Function of two variables

3. ODEs

16

Transfer Function

Procedure to obtain transfer function from nonlinear

process models
Find an equilibrium point of the system
Linearize about the steady-state
Express in terms of deviations variables about the steady-state
Take Laplace transform
Isolate outputs in Laplace domain

Express effect of inputs in terms of transfer functions

17

Block Diagrams

Transfer functions of complex systems can be represented

in block diagram form.

3 basic arrangements of transfer functions:

1. Transfer functions in series
2. Transfer functions in parallel
3. Transfer functions in feedback form

18

Block Diagrams

Transfer functions in series

Overall operation is the multiplication of transfer functions

Resulting overall transfer function

19

Block Diagrams

Transfer functions in series (two first order systems)

Overall operation is the multiplication of transfer functions

Resulting overall transfer function

20

Transfer Functions

DC Motor example:
In terms of angular velocity

In terms of the angle

21

Transfer Functions

Transfer function in parallel

+
+

Overall transfer function is the addition of TFs in parallel

22

Transfer Functions

Transfer function in parallel

+
+

Overall transfer function is the addition of TFs in parallel

23

Transfer Functions

Transfer functions in (negative) feedback form

Overall transfer function

24

Transfer Functions

Transfer functions in (positive) feedback form

Overall transfer function

25

Transfer Function

Example 3.20

26

Transfer Function

Example

A positive feedback loop

27

Transfer Function

Example 3.20

Two systems in parallel

Replace
by
28

Transfer Function

Example 3.20

Two systems in parallel

29

Transfer Function

Example 3.20

A negative feedback loop

30

Transfer Function

Example 3.20

Two process in series

31

First Order Systems

First order systems are systems whose dynamics are

described by the transfer function

where

is the systems (steady-state) gain

is the time constant

First order systems are the most common behaviour

encountered in practice
32

First Order Systems

Examples,
Examples Liquid storage

Assume:
Incompressible flow
Outlet flow due to gravity
Balance equation:
Total
Flow In
Flow Out
33

First Order Systems

Balance equation:

Deviation variables about the equilibrium

Laplace transform

First order system with

34

First Order Systems

Examples:
Examples Cruise control

DC Motor

35

First Order Systems

Liquid Storage Tank

Speed of a car
DC Motor
First order processes are characterized by:
1. Their capacity to store material, momentum
and energy
2. The resistance associated with the flow of
mass, momentum or energy in reaching their
capacity

36

First Order Systems

Step response of first order process

Step input signal of magnitude M

The ultimate change in

is given by

37

First Order Systems

Step response

38

First Order Systems

What do we look for?

Systems Gain: Steady-State Response

Process Time Constant:

Time Required to Reach

63.2% of final value

What do we need?
System initially at equilibrium
Step input of magnitude M
Measure process gain from new steady-state
Measure time constant
39

First Order Systems

First order systems are also called systems with finite

settling time
The settling time
is the time required for the system comes
within 5% of the total change and stays 5% for all times
Consider the step response

The overall change is

40

First Order Systems

Settling time

41

First Order Systems

Process initially at equilibrium subject to a step of

magnitude 1

42

First order process

Ramp response:

Ramp input of slope a

5
4.5
4
3.5
3
2.5
2
1.5
1
0.5
0

0.5

1.5

2.5

3.5

4.5

43

First Order Systems

Sinusoidal response

Sinusoidal
input Asin(t)
2
1.5
1

AR

0.5
0
-0.5

-1
-1.5

10

12

14

16

18

20

44

First Order Systems

10

10

10

Bode Plots

High Frequency
Asymptote

Corner Frequency

-1

-2

10

-2

10

-1

10

10

10

0
-20
-40
-60
-80
-100
10

-2

10

-1

Amplitude Ratio

10

10

10

Phase Shift

45

Integrating Systems
Example: Liquid storage tank

Fi
h
F
Laplace domain dynamics

If there is no outlet flow,

46

Integrating Systems

Example
Capacitor

Dynamics of both systems is equivalent

47

Integrating Systems

Output

Step input of magnitude M

Input

Time

Slope =

Time

48

Integrating Systems

Output

Unit impulse response

Input

Time

Time

49

Integrating Systems

Output

Rectangular pulse response

Input

Time

Time

50

Second order Systems

Second order process:

Assume the general form

where

= Process steady-state gain

= Process time constant
= Damping Coefficient

Three families of processes

Underdamped
Critically Damped
Overdamped

51

Second Order Systems

Three types of second order process:
1. Two First Order Systems in series or in parallel
e.g. Two holding tanks in series
2. Inherently second order processes: Mechanical systems
possessing inertia and subjected to some external force
e.g. A pneumatic valve
3. Processing system with a controller: Presence of a
controller induces oscillatory behavior
e.g. Feedback control system

52

Second order Systems

Multicapacity Second Order Processes

Naturally arise from two first order processes in series

By multiplicative property of transfer functions

53

Transfer Functions

First order systems in parallel

+
+

Overall transfer function a second order process (with one zero)

54

Second Order Systems

Inherently second order process:
e.g. Pneumatic Valve

By Newtons law

p
x

55

Second Order Systems

Feedback Control Systems

56

Second order Systems

Second order process:

Assume the general form

where

= Process steady-state gain

= Process time constant
= Damping Coefficient

Three families of processes

Underdamped
Critically Damped
Overdamped

57

Second Order Systems

Roots of the characteristic polynomial

Case 1)

Two distinct real roots

System has an exponential behavior

Case 2)

One multiple real root

Exponential behavior

Case 3)

Two complex roots

System has an oscillatory behavior
58

Second Order Systems

Step response of magnitude M

=0

1.8

=0.2

1.6
1.4
1.2
1
0.8
0.6

=2

0.4
0.2
0

9 59

10

Second Order Systems

Observations
Responses exhibit overshoot
Large

when

yield a slow sluggish response

Systems with
As
(with
oscillatory

yield the fastest response without overshoot

) becomes smaller, system becomes more

60

Second Order Systems

Characteristics of underdamped second order process
1. Rise time,
2. Time to first peak,
3. Settling time,
4. Overshoot:

5. Decay ratio:

61

Second Order Systems

Step response

62

Second Order Systems

Sinusoidal Response

where

63

Second Order Systems

64

More Complicated Systems

Transfer function typically written as rational function of polynomials

where
and
can be factored following the discussion on partial
fraction expansion

s.t.

where

and

appear as real numbers or complex conjuguate pairs

65

Poles and zeroes

Definitions:
the roots of

are called the zeros of G(s)

the roots of

are called the poles of G(s)

Poles: Directly related to the underlying differential equation

If

, then there are terms of the form

If any

in

vanishes to a unique point

then there is at least one term of the form

does not vanish
66

Poles
e.g. A transfer function of the form

with

can factored to a sum of

A constant term from
A
from the term
A function that includes terms of the form

Poles can help us to describe the qualitative behavior of a complex system

(degree>2)
The sign of the poles gives an idea of the stability of the system
67

Poles

Calculation performed easily in MATLAB

Function POLE, PZMAP

e.g.
s=tf(s);
sys=1/s/(s+1)/(4*s^2+2*s+1);
Transfer function:
1
------------------------4 s^4 + 6 s^3 + 3 s^2 + s
pole(sys)
ans =
0
-1.0000
-0.2500 + 0.4330i
-0.2500 - 0.4330i

MATLAB
68

Poles
Function PZMAP
pzmap(sys)

MATLAB

69

Poles

One constant pole

integrating feature

One real negative pole

decaying exponential

A pair of complex roots with negative real part

decaying sinusoidal

What is the dominant feature?

70

Poles

Step Response

Integrating factor dominates the dynamic behaviour

71

Poles

Example

Poles

-1.0000
-0.0000 + 1.0000i
-0.0000 - 1.0000i

One negative real pole

Two purely complex poles

What is the dominant feature?

72

Poles

Step Response

Purely complex poles dominate the dynamics

73

Poles

Example

Poles

-8.1569
-0.6564
-0.1868

Three negative real poles

What is the dominant dynamic feature?

74

Poles

Step Response

Slowest exponential (
) dominates the dynamic
feature (yields a time constant of 1/0.1868= 5.35 s
75

Poles

Example

Poles
-4.6858
0.3429 + 0.3096i
0.3429 - 0.3096i
One negative real root
Complex conjuguate roots with positive real part
The system is unstable (grows without bound)
76

Poles

Step Response

77

Poles

Two types of poles

Slow (or dominant) poles
Poles that are closer to the imaginary axis
Smaller real part means smaller exponential term and
slower decay
Fast poles
Poles that are further from the imaginary axis
Larger real part means larger exponential term and faster
decay

78

Poles

Poles dictate the stability of the process

Stable poles
Poles with negative real parts
Decaying exponential
Unstable poles
Poles with positive real parts
Increasing exponential
Marginally stable poles
Purely complex poles
Pure sinusoidal

79

Poles

Oscillatory
Behaviour

Integrating
Behaviour

Pure Exponential

Stable

Unstable
Marginally
Stable

80

Poles

Example

Stable Poles
Unstable Poles
Fast Poles

Slow Poles

81

Poles and zeroes

Definitions:
the roots of

are called the zeros of G(s)

the roots of

are called the poles of G(s)

Zeros:
Do not affect the stability of the system but can modify the dynamical
behaviour

82

Zeros

Types of zeros:
Slow zeros are closer to the imaginary axis than the dominant poles of the
transfer function
Affect the behaviour
Results in overshoot (or undershoot)
Fast zeros are further away to the imaginary axis
have a negligible impact on dynamics
Zeros with negative real parts,
Slow stable zeros lead to overshoot

, are stable zeros

Zeros with positive real parts,

, are unstable zeros
Slow unstable zeros lead to undershoot (or inverse response)

83

Zeros
Can result from two processes in parallel

+
+

If gains are of opposite signs and time constants are different then a right half
plane zero occurs

84

Zeros

Example

Poles

Zeros

-8.1569
-0.6564
-0.1868

-10.000

What is the effect of the zero on the dynamic behaviour?

85

Zeros

Poles and zeros

Fast, Stable
Zero

Dominant
Pole

86

Zeros

Unit step response:

Effect of zero is negligible

87

Zeros

Example

Poles

Zeros

-8.1569
-0.6564
-0.1868

-0.1000

What is the effect of the zero on the dynamic behaviour?

88

Zeros

Poles and zeros:

Slow (dominant)
Stable zero

89

Zeros

Unit step response:

Slow dominant zero yields an overshoot.

90

Zeros

Example

Poles

Zeros

-8.1569
-0.6564
-0.1868

10.000

What is the effect of the zero on the dynamic behaviour?

91

Zeros

Poles and Zeros

Dominant
Pole

Fast Unstable
Zero

92

Zeros

Unit Step Response:

Effect of unstable zero is negligible

93

Zeros

Example

Poles

Zeros

-8.1569
-0.6564
-0.1868

0.1000

What is the effect of the zero on the dynamic behaviour?

94

Zeros

Poles and zeros:

Slow
Unstable Zero

95

Zeros

Unit step response:

Slow unstable zero causes undershoot (inverse reponse)

96

Zeros
Observations:
Adding a stable zero to an overdamped system yields overshoot
Adding an unstable zero to an overdamped system yields
undershoot (inverse response)
Inverse response is observed when the zeros lie in right half
complex plane,
Overshoot or undershoot are observed when the zero is dominant
(closer to the imaginary axis than dominant poles)

97

Zeros

Example: System with complex zeros

Dominant
Pole
Slow
Stable Zeros

98

Zeros

Poles and zeros

Poles have negative real parts: The system is stable
Dominant poles are real: yields an overdamped behaviour
A pair of slow complex stable poles: yields overshoot

99

Zeros

Unit step response:

Effect of slow zero is significant and yields oscillatory behaviour

100

Zeros

Example: System with zero at the origin

Zero at
Origin

101

Zeros

Unit step response

Zero at the origin: eliminates the effect of the step

102

Zeros

Observations:
Complex (stable/unstable) zero that is dominant yields an
(overshoot/undershoot)
Complex slow zero can introduce oscillatory behaviour
Zero at the origin eliminates (or zeroes out) the system response

103

Delay
Fi

Control loop
h

Time required for the

fluid to reach the valve
usually approximated as
dead time

Manipulation of valve does not lead to immediate

change in level
104

Delay
Delayed transfer functions

e.g. First order plus dead-time

Second order plus dead-time

105

Delay

Dead time (delay)

Most processes will display some type of lag time
Dead time is the moment that lapses between input changes and
process response

Step response of a first order plus dead time process

106

Delay

Delayed step response:

107

Delay

Problem
use of the dead time approximation makes analysis (poles and
zeros) more difficult

Approximate dead-time by a rational (polynomial)

function
Most common is Pade approximation

108

Pade Approximations

In general Pade approximations do not approximate dead-time very

well

Pade approximations are better when one approximates a first order

plus dead time process

Pade approximations introduce inverse response (right half plane

zeros) in the transfer function

109

Pade Approximations

Step response of Pade Approximation

Pade approximation introduces inverse response

110