NDJ30103

CONTROL SYSTEM

CHAPTER 6
SYSTEM STABILITY
 SUB CHAPTERS
1. Introduction
2. System Stability
3. Determining System Stability
4. Routh-Hurwitz Criterion
5. Construction of Routh Table
6. Special Case 1: Zero in First Column
7. Special Case 2: Zero for Entire Row
8. Stability Via Routh-Hurwitz
9. Computational Approach
1. INTRODUCTION

 Total response of a system is the sum of forced and natural response.

c(t ) = c forced (t ) + cnatural(t )

 Definition of stability, for linear, time-invariant system by using natural response:
 A system is stable if the natural response approaches zero or steady state as time
approaches infinity.
 A system is unstable if the natural response approaches infinity as time approaches
infinity.
 A system is marginally stable if the natural response neither decays nor grows but
remains constant or oscillates.
1. INTRODUCTION

Types of stability analysis

 Absolute Stability (Qualitative) :
i. The absolute stability indicates whether the system is stable or not.
ii. This is indicated by the presence of one or more poles in right-hand-plane (RHP).

 Relative Stability (Comparative) :

i. Relative stability refers to the degree of stability of a stable system described by above.
ii. This depends on the transfer function of the system, which is represented by both the
numerator (that yields the zeros) and denominator (that yields the poles).
iii. This can then be referred to in the study of system response either in time or
frequency domain.
2. SYSTEM STABILITY

 Stable systems have closed-loop transfer functions with poles only in the left
half-plane.
2. SYSTEM STABILITY

 Unstable systems have closed-loop transfer functions with at least one pole in the right half-plane and/or
poles of multiplicity greater than 1 on the imaginary axis.
2. SYSTEM STABILITY

 Marginally stable systems have closed-loop transfer functions with only

imaginary axis poles of multiplicity 1 and poles in the left half-plane.
3. DETERMINING SYSTEM STABILITY
 To determine stability of a given system, we have to consider the manner in which the
system is operating, whether open-loop or closed-loop.
 Step:
i. If the system is operating in closed-loop, first find output, C(s) of the closed loop transfer
function.
ii. Find the closed-loop poles.
iii. If the order of the system is 2 or less, factorise the denominator of the transfer function.
This will provide the roots of the polynomial, or the closed-loop poles of the system.
iv. If the system order is higher than 2nd-order, construct Routh table and apply Routh-
Hurwitz Criterion.
v. Any poles that exist on the RHP will indicate that the system is unstable.
4. ROUTH-HURWITZ CRITERION

 Routh-Hurwitz Stability Criterion: The number of roots of the polynomial that

are in the right half-plane is equal to the number of changes in the first column.
 Systems with the transfer function having all poles in the LHP is stable.
 Hence, we can conclude that a system is stable if there is no change of sign in
the first column of its Routh table.
 However, special cases exists when:
i. There exists zero only in the first column.
ii. The entire row is zero.
 This method will show number of poles in RHP, LHP & jω-axis.
4. ROUTH-HURWITZ CRITERION

 If a polynomial is given by: T (s) = an s n + an −1s n −1 + ..... + a1s + a0 = 0

Where,
an, an-1, …, a1, a0 are constants
n = 1, 2, 3,…, ∞

 The necessary conditions for stability are:

i. All the coefficients of the polynomial are of the same sign. If not, there are
poles on the right hand side of the s-plane.
4. ROUTH-HURWITZ CRITERION
 For the sufficient condition, we must form a Routh-array.
5. CONSTRUCTION OF ROUTH TABLE

Equivalent closed-loop transfer function

Initial layout for Routh table

5. CONSTRUCTION OF ROUTH TABLE

Completed Routh table

5. CONSTRUCTION OF ROUTH TABLE
 Example 1: How many roots exist on the right-half plane?
5. CONSTRUCTION OF ROUTH TABLE

Solution: 2 changes in 1st column. 2 poles in RHP. System Unstable

 6. SPECIAL CASE I : ZERO IN FIRST COLUMN

 If the first element of a row is zero, division by zero would be required to form the
next row.
 To avoid this, an epsilon, , is assigned to replace the zero in the first column.
 The value of epsilon,  is allow to approach zero from either positive or negative
side.
6. SPECIAL CASE I : ZERO IN FIRST COLUMN
 Example 2: Consider the following closed-loop transfer function T(s).

10
T ( s) = 5
s + 2s 4 + 3s 3 + 6s 2 + 5s + 3
6. SPECIAL CASE I : ZERO IN FIRST COLUMN

 To determine the system stability,

sign changes were observed after
substituting  with a very small
positive number or alternatively
a very small negative number.

▪ 2 changes sign
▪ 2 poles at RHP
▪ System Unstable
7. SPECIAL CASE II : ZERO FOR ENTIRE ROW

 An entire row of zeros will appear in the Routh table when a purely even or
purely odd polynomial is a factor of the original polynomial.
 Example: s4 + 5s2 + 7 has an even powers of s.
 Even polynomials have roots that are symmetrical about the origin.
i. Roots are symmetrical & real
ii. Roots are symmetrical & imaginary
iii. Roots are quadrantal
7. SPECIAL CASE II : ZERO FOR ENTIRE ROW
7. SPECIAL CASE II : ZERO FOR ENTIRE ROW
10
 Example 3: T ( s) = 5
s + 7 s 4 + 6s 3 + 42 s 2 + 8s + 56
1 6 8
7 1 42 6 56 8
01 03 0
3 8 0
0

8
7. SPECIAL CASE II : ZERO FOR ENTIRE ROW

Aux Equation is getting

from the row above zero
row


7. SPECIAL CASE II : ZERO FOR ENTIRE ROW


8. STABILITY VIA ROUTH-HURWITZ
 Example 4: Find the range of gain K for the system below that will cause the
system to be stable, unstable and marginally stable. Assume K > 0.

 Closed-loop transfer function:

K
T ( s) = 3
s + 18 s 2 + 77 s + K
8. STABILITY VIA ROUTH-HURWITZ

 Forming the Routh table:

8. STABILITY VIA ROUTH-HURWITZ
 Solution:

 If K < 1386 :
All the terms in 1st column will be positive and since there are no sign changes, the system
will have 3 poles in the left-half plane and is stable.

 If K > 1386 :
The s1 in the first column is negative. There are 2 sign changes, indicating that the system has
two right-half-plane poles and one left-half plane pole, which make the system unstable.

 If K = 1386:
The entire row of zeros, which signify the existence of jω poles. Using the equation from
the row above the RoZ can find the poles on the imaginary axis of the s-plane.
 Further Reading

i. Nise, N.S. (2008). Control System Engineering (5th

Ed), John Wiley & Sons. Chapter 6
ii. Dorf, R.C., Bishop R.H. (2001). Modern Control
Systems (9th Ed), Prentice Hall. Chapter 6.
9. COMPUTATIONAL APPROACH
Calculating the Routh-Hurwitz stability criterion can be done in Scilab using
the instruction
 [R [, num]] = routh_t(P)
R is the output argument for the Routh Table of polynomial, P.
num is the optional output argument to display the number of sign changes in the Routh
Table.

 R = routh_t(G, K)
R is the output argument for the Routh Table for the closed-loop characteristic equation,
G when gain, K = 1.
K is the variable gain which can be applied as the variable name, or a number between
any value 1 < K < ∞.
9. COMPUTATIONAL APPROACH
Example 1 @ slide 15
clear; clc;
den = poly([-2 -3 -5],'s','roots');
G = syslin('c',1,den);
[R, num] = routh_t(G,1000);
//Routh table example for Special Case 1 (0 in first column) clear; clc; H1 = poly([3 5 6 3 2 1],'s','coeff'); [R1,num1]= routh_t(H1);

9. COMPUTATIONAL APPROACH
Example 2 @ slide 17 for
Special Case 1 (0 in first column)
clear; clc;
H1 = poly([3 5 6 3 2 1],'s','coeff’);
[R1,num1]= routh_t(H1);
 9. COMPUTATIONAL APPROACH
Example 3 @ slide 21 for
Special Case II (0 for entire row)
RoZ. Replaced with
clear; clc; auxillary equation

H2 = poly([56 8 42 6 7 1],'s','coeff’);
[R2,num2]= routh_t(H2);
 9. COMPUTATIONAL APPROACH
Example 4 @ slide 24 to
find system stability
clear; clc;
s = poly(0,'s'); k = poly(0,'k’);
H3 = s*(s+7)*(s+11)
R3 = routh_t(1/H3,k)