Chapter 10 - Stability of Closed-Loop Control Systems

CEB2083 PROCESS INSTRUMENTATION & CONTROL
MAY 2020 SEMESTER
Chapter 10:
Stability of Closed-Loop Control Systems
BY
Assoc. Prof. Dr Marappagounder Ramasamy / Dr Serene Lock /
Dr Mohd Hilmi Noh / Dr Nurul Aini Amran
UNIVERSITI TEKNOLOGI PETRONAS (UTP)

Chapter Objectives
End of this chapter, you should be able to:

1. Explain the concepts of stability of feedback control
systems
2. Determine the stability of control systems
Course learning outcomes
CLO1
Develop dynamic models of chemical processes
CLO2
Design feedback control systems using modern IT tools
CLO3
Select appropriate instruments for control systems
Introduction
An important consequence of feedback control is that it

01
can cause oscillatory responses.
Under certain circumstances, the oscillations may be

02 undamped or even have amplitude that increases with
time until a physical limit is reached.
In these situations, the closed-loop system is said to be

03 unstable.
Control system:
Consider the feedback control system with the following

transfer functions:
1
Gc  K c Gv  G p  Gd 
1
Gm 
1
(10.1)
2s  1 5s  1 s 1
Consider a step change in set-point Ysp ( s)  1 / s . The transfer

function for set-point changes is:
Y K m Gc Gv G p
 (10.2)
Ysp 1  Gc Gv G p Gm
Substituting (10.1) in (10.2), and rearranging gives us

K c ( s  1) 1
Y ( s)  (10.3)
10s 3  17 s 2  8s  1  K c s
 After Kc is specified, y(t) can be obtained.
 Fig. below demonstrates that as Kc increases, the response become

more oscillatory and is unstable for Kc = 15.
We can use MATLAB to plot the step responses the cases of Kc=6 and 2.
What do you observe?
Step Response
1.5
Kc = 6
Kc = 2
1
y
0.5
0
0 5 10 15 20 25 30 35 40
time (min)
General Stability Criterion:
Some industrial processes are stable without feedback controllers.

They are said to be open-loop stable or self-regulating.
Definition of stability: An unconstrained linear system is

said to be stable if the output response is bounded for all
bounded inputs. Otherwise, it is said to be unstable.
By a bounded input, we mean an input variable that stays

within upper and lower limits for all values of time.
The term unconstrained refer to the ideal situation where there is

no physical limits on the input and output variables.
Characteristic equation
For the standard block diagram, we have developed that

Gd K m Gc Gv G p
Y D Ysp (10.4)
1  GOL 1  GOL
where GOL is the open-loop transfer function, GOL = GcGpGvGm.
Consider set-point changes only. Eq. (10.4) reduces to

Y K m G c Gv G p
 (10.5)
Ysp 1  GOL
If GOL is a ratio of polynomials in s, then the closed-loop transfer

function in (10.5) is also a rational function. Then, it can be factored
into poles (pi) and zeroes (zi) as
Y ( s  z1 )( s  z 2 )....(s  z m )
 K (10.6)
Ysp ( s  p1 )( s  p 2 )....(s  p n )
where K’ is a multiplicative constant selected to give the correct

steady-state gain. To have a physically realizable system, the number
of poles, pN must be greater than or equal to the number of zeroes, zM.
Comparing (10.5) and (10.6) indicates that the poles are also the roots
of the following equation, referred to as the characteristic equation of
the closed-loop system:
1  GOL  0 (10.7)
The characteristic equation plays a decisive role in determining system

stability.
For a unit step change in set-point, (10.6) becomes
K  ( s  z1 )( s  z2 )....(s  zm )
Y (10.8)
s ( s  p1 )( s  p2 )....(s  pn )
If there are no repeat poles (all distinct roots of the denominator),

then the partial fraction expansion of (10.8) has the form
A0 A1 A2 An
Y    ...  (10.9)
s ( s  p1 ) ( s  p2 ) ( s  pn )
Taking the inverse Laplace transform of (10.9) gives
y (t )  Ao  A1e p1t  A2 e p2t  ....  An e pnt (10.10)

Suppose that one of the poles is a positive real number;

i.e., pk > 0. Then it is clear from (10.10) that y(t) is unbounded
and thus the closed-loop system is unstable.
If pk is a complex number, pk = ak + jbk, with a positive real

part, then the system is also unstable.
If all the poles are negative (or have negative real parts)
then the system is stable.
Pole-Zero Map
General stability criterion:
A feedback control system is stable if and

only if all roots of the characteristic equation
(1 + GOL) are negative or have negative real
parts. Otherwise, the system is unstable.
Graphical interpretation of stability criterion:
Example 1
 Consider a standard block diagram with Gv=Kv , Gm=1

and Gp=Kp/(τps+1). Determine the stability
characteristics if a proportional controller is used, i.e,
Gc=Kc.
 The characteristic equation is:
Kc Kv K p
1 0
 ps 1
 Which can be rearranged as:
 p s  1  Kc Kv K p  0
Example 1
 So, the root is:

Kc Kv K p  1
s
p
 Since tp is always positive, KcKvKp>0.
 Correct controller action (reverse/direct) is required so

that the root, s<0 (negative).
Routh Stability Criterion
Uses an analytical technique for determining whether any

01 roots of a polynomial have positive real parts.
02 Characteristic equation
a n s n  a n 1 s n 1    a1 s  a0  0 (10.11)
where an >0. According to the Routh criterion, if any of the

coefficients a0, a1, aK, an-1 are negative or zero, then at
least one root of the characteristic equation lies in the RHP,
and thus, the system is unstable.
Routh Array
 On the other hand, if all of the coefficients are positive, then one
must construct the Routh Array shown below:
 For stability, all elements in the first column must be

positive.
 The first two rows of the Routh Array comprise the
coefficients in the characteristics equation. The elements in
the remaining rows are calculated from coefficients by the
using the formulas:
Note: Recall the determinant of a 2x2 matrix.
an 1an  2  an an 3 an 1an  4  an an 5
b1  b2 
an 1 an 1
b1an 3  an 1b2 b1an 5  an 1b3

c1  c2 
b1 b1
 (n+1 rows must be constructed, n = order of the

characteristic eqn.)
Routh Stability Criterion
A necessary and sufficient condition for all roots of the

01 characteristic equation to have negative real parts is
that all of the elements in the left column of the Routh
array are positive.
Example 2: Determine the stability of a system that

02
has the characteristic equation
s 4  5s 3  3s 2  1  0
Solution: Because the s term is missing, its coefficient is

zero. Thus the system is unstable.
Example 3
Find the values of the controller gain Kc that make the

feedback control system with the following characteristic
equation stable.
10s 3  17 s 2  8s  1  K C  0
All coefficients are positive provided that 1+Kc > 0 or Kc

> -1.
Example 3
The Routh array is:
n3 n3
10 8 0 10 8
17 1  K C 0 17 1  KC
b1 b2 17(8)  10 1  K C 
0
c1 17
1  KC
To have a stable system, each element in the left column must

be positive, Kc < 12.6 (from b1) and Kc > -1 (from c1).
Thus we conclude -1 < Kc < 12.6.

Example 4
Use Direct Substitution method to determine Kcm for the

system with the characteristic equation below:
10s  17 s  8s  1  K c  0
3 2
Solution
Substitute s=jw and Kc=Kcm into the above equation.
 10 jw 3  17w 2  8 jw  1  K cm  0
(1  K cm  17w 2 )  j (8w  10w 3 )  0

Example 4
The above equation is satisfied if both the real and

imaginary parts are identically zero. In other words,
1  K cm  17w 2  0
8w  10w 3  0
Hence,
w  0.894
K cm  12.6
Example 5
Use Routh Stability Criterion to determine the range of Kc

that results in a stable closed-loop system.
Conclusion
You have been introduced to the:
01 Concept of Stability
02 General Stability Criterion
03 Routh Stability Criterion
04 Direct Substitution Method

Chapter 10 - Stability of Closed-Loop Control Systems

Uploaded by

Copyright:

Available Formats

Chapter 10 - Stability of Closed-Loop Control Systems

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chapter 10 - Stability of Closed-Loop Control Systems

Uploaded by

Copyright:

Available Formats

CEB2083 PROCESS INSTRUMENTATION & CONTROL

MAY 2020 SEMESTER

UNIVERSITI TEKNOLOGI PETRONAS (UTP)

End of this chapter, you should be able to:

An important consequence of feedback control is that it

Under certain circumstances, the oscillations may be

In these situations, the closed-loop system is said to be

Consider the feedback control system with the following

Consider a step change in set-point Ysp ( s)  1 / s . The transfer

Substituting (10.1) in (10.2), and rearranging gives us

 Fig. below demonstrates that as Kc increases, the response become

Some industrial processes are stable without feedback controllers.

Definition of stability: An unconstrained linear system is

By a bounded input, we mean an input variable that stays

The term unconstrained refer to the ideal situation where there is

For the standard block diagram, we have developed that

Consider set-point changes only. Eq. (10.4) reduces to

If GOL is a ratio of polynomials in s, then the closed-loop transfer

where K’ is a multiplicative constant selected to give the correct

The characteristic equation plays a decisive role in determining system

If there are no repeat poles (all distinct roots of the denominator),

Taking the inverse Laplace transform of (10.9) gives

y (t )  Ao  A1e p1t  A2 e p2t  ....  An e pnt (10.10)

Suppose that one of the poles is a positive real number;

If pk is a complex number, pk = ak + jbk, with a positive real

A feedback control system is stable if and

 Consider a standard block diagram with Gv=Kv , Gm=1

 Which can be rearranged as:

 So, the root is:

 Since tp is always positive, KcKvKp>0.

 Correct controller action (reverse/direct) is required so

Uses an analytical technique for determining whether any

where an >0. According to the Routh criterion, if any of the

 For stability, all elements in the first column must be

b1an 3  an 1b2 b1an 5  an 1b3

 (n+1 rows must be constructed, n = order of the

A necessary and sufficient condition for all roots of the

Example 2: Determine the stability of a system that

Solution: Because the s term is missing, its coefficient is

Find the values of the controller gain Kc that make the

All coefficients are positive provided that 1+Kc > 0 or Kc

To have a stable system, each element in the left column must

Thus we conclude -1 < Kc < 12.6.

Use Direct Substitution method to determine Kcm for the

(1  K cm  17w 2 )  j (8w  10w 3 )  0

The above equation is satisfied if both the real and

Use Routh Stability Criterion to determine the range of Kc

You have been introduced to the:

02 General Stability Criterion

03 Routh Stability Criterion

04 Direct Substitution Method

You might also like