Stability: Lecture Notes of ELG3155: Introduction To Control Systems
Stability: Lecture Notes of ELG3155: Introduction To Control Systems
Stability: Lecture Notes of ELG3155: Introduction To Control Systems
Wail Gueaieb
Chapter 6
Stability
Lecture notes of ELG3155: Introduction to Control Systems
Objectives
Introduction
Routh-Hurwitz Criteria
Routh-Hurwitz Table
Routh-Hurwitz Criteria:
Special Cases
Routh-Hurwitz Criteria:
Supplementary
Examples
References
Wail Gueaieb
EECS
University of Ottawa
6.1
Stability
Outline
Wail Gueaieb
1 Introduction
2 Routh-Hurwitz Criteria
Objectives
Introduction
Routh-Hurwitz Table
Routh-Hurwitz Criteria:
Special Cases
4 Routh-Hurwitz Criteria: Special Cases
Case 1: a zero component at the beginning of a nonzero row Routh-Hurwitz Criteria:
Supplementary
Case 2: an entire row of zeros (ROZ) Examples
References
5 Routh-Hurwitz Criteria: Supplementary Examples
6.2
Stability
Objectives
Wail Gueaieb
Objectives
Introduction
This chapter aims at studying how:
Routh-Hurwitz Criteria
• to determine a system’s stability from its transfer function without solving
Routh-Hurwitz Table
for the poles.
Routh-Hurwitz Criteria:
• to determine a system’s parameter to guarantee its stability. Special Cases
Routh-Hurwitz Criteria:
Supplementary
Examples
References
6.3
Stability
Introduction
Wail Gueaieb
A system is said to be unstable if and only if there exists at least one bounded Introduction
input that leads to an unbounded output. Otherwise, it is stable. Routh-Hurwitz Criteria
Routh-Hurwitz Table
Poles of an unstable LTI system Routh-Hurwitz Criteria:
Special Cases
An LTI system is unstable if at least one of its poles is at the origin, or on the
Routh-Hurwitz Criteria:
RHS of the s-plane, or a pair of conjugate poles on the jω-axis is repeated Supplementary
twice or more. Examples
References
jω jω jω
× × ×(2)
× σ × σ × σ
× × ×(2)
6.4
Definition (Stability) Stability
Wail Gueaieb
A system is said to be stable if and only if every bounded input leads to a
bounded output. A stable system is either asymptotic stable or marginally
(critically) stable.
Routh-Hurwitz Criteria:
Special Cases
Routh-Hurwitz Criteria:
Poles of an asymptotically stable LTI system Supplementary
Examples
An LTI system is asymptotically stable if and only if all of its poles are on the
Stability in State Space
LHS of the s-plane.
References
jω jω
×
× ×
× σ σ
× ×
× 6.5
Stability
A system is marginally stable (or critically stable) when its natural response
cn (t) is oscillatory with a constant magnitude.
cn (t)
t Objectives
Introduction
Routh-Hurwitz Criteria
Routh-Hurwitz Table
Poles of a critically stable LTI system
Routh-Hurwitz Criteria:
An LTI system is critically stable if and only if it is stable and has at least one Special Cases
conjugate pair of poles are the imaginary axis of the s-plane. Routh-Hurwitz Criteria:
Supplementary
Examples
×
×
×
× σ σ
×
×
×
6.6
Stability
Wail Gueaieb
Objectives
Unstable
Introduction
Routh-Hurwitz Table
Stable
Routh-Hurwitz Criteria:
Special Cases
Asymptotically stable
Routh-Hurwitz Criteria:
Supplementary
Figure 1: The different types of stability for an LTI system. Examples
References
6.7
Stability
Wail Gueaieb
Objectives
Introduction
Routh-Hurwitz Criteria
Routh-Hurwitz Table
Routh-Hurwitz Criteria:
Special Cases
Routh-Hurwitz Criteria:
Supplementary
Examples
References
Figure 2: (a) Stable system; (b) unstable system (Fig. 6.1 of [Nise, 2015]).
6.8
Stability
Wail Gueaieb
Objectives
Examples of characteristic equations satisfying this condition (the system may or Introduction
may not be stable)
Routh-Hurwitz Criteria
Routh-Hurwitz Criteria:
• −30s8 − 5s7 − 12s4 − 3s2 − 1 = 0 Special Cases
Routh-Hurwitz Criteria:
Supplementary
Examples
6.9
Stability
Routh-Hurwitz Criteria
Wail Gueaieb
Routh-Hurwitz Table
which have positive (> 0) real parts without solving the equation.
Routh-Hurwitz Criteria:
Special Cases
Routh-Hurwitz Criteria:
Supplementary
Examples
Routh-Hurwitz Criteria (necessary and sufficient condition)
Stability in State Space
The number of roots of a polynomial which are on the RHS of the s-plane is References
equal to the number of sign changes in the first column of the Routh-Hurwitz
table.
6.10
Stability
Routh-Hurwitz Table
Wail Gueaieb
Introduction
Routh-Hurwitz Table
sn an an−2 an−4 ··· ··· ··· ···
Routh-Hurwitz Criteria:
sn−1 an−1 an−3 an−5 ··· ··· ··· ··· Special Cases
sn−2 zi−2,1 zi−2,j+1 Routh-Hurwitz Criteria:
sn−3 zi−1,1 zi−1,j+1 Supplementary
Examples
..
. zi,j Stability in State Space
s1 References
s0
z zi−2,j+1
− i−2,1
zi−1,1 zi−1,j+1
where zi,j =
zi−1,1
6.11
Stability
Wail Gueaieb
Objectives
Example 6.1 Introduction
R(s) E(s) C(s)
Determine the stability of the following system, +
−
G(s) Routh-Hurwitz Criteria
1000
where G(s) = (s+2)(s+3)(s+5) . Routh-Hurwitz Table
Routh-Hurwitz Criteria:
Special Cases
Routh-Hurwitz Criteria:
Supplementary
Examples
References
6.12
Stability
Wail Gueaieb
Verification
Fig. 3 shows the step response of this system. It is clear that the response
goes out of bound, hence the system is unstable.
Step Response
#10 25
5
4
Objectives
3
Introduction
2
Routh-Hurwitz Criteria
Amplitude
1
Routh-Hurwitz Table
0
Routh-Hurwitz Criteria:
-1 Special Cases
-2 Routh-Hurwitz Criteria:
Supplementary
-3 Examples
6.13
Stability
Wail Gueaieb
Routh-Hurwitz Criteria:
Special Cases
Routh-Hurwitz Criteria:
Facilitating the manual computation of the RH table Supplementary
Examples
6.14
Stability
Wail Gueaieb
Objectives
Example Introduction
Consider the following system, where G(s) = R(s) E(s) C(s) Routh-Hurwitz Criteria
K + G(s)
and K ∈ R. For what values of −
(s+1)(s2 +2s+1) Routh-Hurwitz Table
K is the system stable? Routh-Hurwitz Criteria:
Special Cases
Routh-Hurwitz Criteria:
Supplementary
Examples
References
6.15
Stability
Wail Gueaieb
Verification (Fig. 4)
0.8 1.6
Objectives
-0.5
0.7 1.4
Introduction
0.6 -1 1.2
Routh-Hurwitz Criteria
Amplitude
Amplitude
Amplitude
0.5 1
-1.5
0.4 0.8
0.3 -2 0.6
Routh-Hurwitz Table
0.2 0.4
-2.5
Routh-Hurwitz Criteria:
0.1 0.2
Special Cases
0 -3 0
0 2 4 6 8 10 12 14 16 18 0 10 20 30 40 50 60 70 80 90 100 0 5 10 15 20 25 30 35 40
Time (seconds) Time (seconds) Time (seconds) Routh-Hurwitz Criteria:
Supplementary
K =2 K = −4 K =8 Examples
6.16
Stability
Routh-Hurwitz Criteria: Special Cases
Case 1: a zero component at the beginning of a nonzero row Wail Gueaieb
Objectives
Introduction
Routh-Hurwitz Criteria:
Supplementary
Examples
References
6.17
Stability
Wail Gueaieb
Verification
Fig. 5 shows the step response of this system. It is clear that the response
goes out of bound, hence the system is unstable.
Step Response
#10 27
7
6
Objectives
5 Introduction
4 Routh-Hurwitz Criteria
Amplitude
3 Routh-Hurwitz Table
Routh-Hurwitz Criteria:
2
Special Cases
1 Case 1: a zero component
at the beginning of a
nonzero row
0
Case 2: an entire row of
zeros (ROZ)
-1
0 20 40 60 80 100 120 140
Routh-Hurwitz Criteria:
Time (seconds)
Supplementary
Examples
Figure 5: Step response of the system Stability in State Space
References
6.18
Stability
Wail Gueaieb
Objectives
Example 6.2
Introduction
Find the distribution of the poles on the s-plane for a system whose transfer
Routh-Hurwitz Criteria
function is
10 Routh-Hurwitz Table
T (s) = 5
s + 2s4 + 3s3 + 6s2 + 5s + 3 Routh-Hurwitz Criteria:
Special Cases
Case 1: a zero component
at the beginning of a
nonzero row
Case 2: an entire row of
zeros (ROZ)
Routh-Hurwitz Criteria:
Supplementary
Examples
References
6.19
Stability
Wail Gueaieb
Verification
Objectives
Introduction
Routh-Hurwitz Criteria
Routh-Hurwitz Table
Routh-Hurwitz Criteria:
Special Cases
Case 1: a zero component
at the beginning of a
nonzero row
Case 2: an entire row of
zeros (ROZ)
System poles: −1.6681, 0.3429 ± j 1.5083, and −0.5088 ± j 0.7020 Routh-Hurwitz Criteria:
Supplementary
Examples
References
6.20
Stability
Routh-Hurwitz Criteria: Special Cases
Case 2: an entire row of zeros (ROZ) Wail Gueaieb
Objectives
Introduction
Routh-Hurwitz Criteria:
Supplementary
Examples
References
6.21
Stability
Wail Gueaieb
Verification
Fig. 6 shows the step response of this system. At steady state, the response is
oscillatory with a fixed amplitude and frequency, hence the system is critically
stable.
Step Response
4
3.5
Objectives
Introduction
3
Routh-Hurwitz Criteria
2.5
Amplitude
Routh-Hurwitz Table
2
Routh-Hurwitz Criteria:
1.5 Special Cases
Case 1: a zero component
1 at the beginning of a
nonzero row
0
0 5 10 15 20 25 30 35 40 45 50 Routh-Hurwitz Criteria:
Time (seconds) Supplementary
Examples
6.22
Stability
Wail Gueaieb
f (−s) = f (s), ∀s ∈ D
Objectives
Introduction
Routh-Hurwitz Criteria
So, if f is an even function, then for every root s of f in the LHS of the s-plane Routh-Hurwitz Table
there exists another root −s on the RHS of the plane, and vice versa. Routh-Hurwitz Criteria:
Special Cases
Case 1: a zero component
jω at the beginning of a
−s nonzero row
Routh-Hurwitz Criteria:
σ Supplementary
Examples
6.23
Stability
Wail Gueaieb
Auxiliary equation
Objectives
• A system’s auxiliary equation exits only when there is a ROZ in the RH Introduction
• The auxiliary equation is the equation corresponding to the row in the RH Routh-Hurwitz Table
Routh-Hurwitz Criteria:
Supplementary
Examples
References
6.24
Stability
Wail Gueaieb
Remarks
• A ROZ exists in the RH table if (not iff) the characteristic equation has one
or more pairs of complex conjugate roots in the jω-axis. In this case,
• The characterictic polynomial, say Q(s), can be factorized into a product of an Objectives
even auxiliary polynomial, say P(s), and a non-even polynomial, say P 0 (s). Introduction
That is,
0 Routh-Hurwitz Criteria
Q(s) = P(s) P (s).
Routh-Hurwitz Table
Thus, the roots of P(s) and P 0 (s), are also roots of Q(s).
Routh-Hurwitz Criteria:
• The distribution of the roots of the auxiliary polynomial P(s) on the s-plane is Special Cases
Case 1: a zero component
determined from the sign changes of the coefficients in the first column of the at the beginning of a
RH table corresponding to the rows of sp down to s0 , where p is the order of nonzero row
Case 2: an entire row of
P(s). See Fig. 7. zeros (ROZ)
• The distribution of the roots of the remaining (non-even) polynomial P 0 (s) is Routh-Hurwitz Criteria:
Supplementary
determined from the sign changes of the coefficients in the first column of the Examples
RH table corresponding to the rows of sn down to sp , where n is the order of
the control system (and of the characteristic polynomial). Stability in State Space
References
6.25
Stability
Wail Gueaieb
Col. 1
Objectives
Routh-Hurwitz Criteria
Routh-Hurwitz Table
Aux. poly. P(s) Routh-Hurwitz Criteria:
ROZ Special Cases
Case 1: a zero component
at the beginning of a
Routh-Hurwitz Criteria:
Supplementary
Examples
Figure 7: Pole distribution on the s-plane in the existence of an auxiliary equation. Stability in State Space
References
6.26
Stability
Wail Gueaieb
Routh-Hurwitz Table
Routh-Hurwitz Criteria:
Special Cases
Definition (Critical gain) Case 1: a zero component
at the beginning of a
In the case of a free gain, the critical gain is the value of this gain which makes nonzero row
Routh-Hurwitz Criteria:
Supplementary
Examples
References
6.27
Stability
Wail Gueaieb
Example 6.4
Find the distribution of the poles on the s-plane for a system whose transfer
function is
10
T (s) = 5 Objectives
s + 7s4 + 6s3 + 42s2 + 8s + 56
Introduction
Routh-Hurwitz Criteria
Routh-Hurwitz Table
Routh-Hurwitz Criteria:
Special Cases
Case 1: a zero component
at the beginning of a
nonzero row
Case 2: an entire row of
zeros (ROZ)
Routh-Hurwitz Criteria:
Supplementary
Examples
References
6.28
Stability
Wail Gueaieb
Verification
Fig. 8 shows the step response of this system. At steady state, the response is
oscillatory with a fixed amplitude and frequency, hence the system is critically
stable.
Step Response
1.8
1.6 Objectives
1.4 Introduction
1 Routh-Hurwitz Table
0
0 5 10 15 20 25 30 35 40 Routh-Hurwitz Criteria:
Time (seconds) Supplementary
Examples
6.29
Stability
Wail Gueaieb
Remarks
Objectives
• If a system is marginally stable, then it must have a ROZ in its RH table. Introduction
• However, a ROZ doesn’t necessary mean that the system is marginally Routh-Hurwitz Criteria
stable (see Example 6.5 and Problem 6.19, for instance). Routh-Hurwitz Table
• A ROZ simply means that the characteristic polynomial can be factorized Routh-Hurwitz Criteria:
Special Cases
as a product of two polynomials, one of which is even. Case 1: a zero component
at the beginning of a
nonzero row
Case 2: an entire row of
zeros (ROZ)
Routh-Hurwitz Criteria:
Supplementary
Examples
References
6.30
Stability
Wail Gueaieb
Example 6.5
Find the distribution of the poles on the s-plane for a system whose transfer
function is
20
T (s) =
s8 + s7 + 12s6 + 22s5 + 39s4 + 59s3 + 48s2 + 38s + 20
Objectives
Introduction
Routh-Hurwitz Criteria
Routh-Hurwitz Table
Routh-Hurwitz Criteria:
Special Cases
Case 1: a zero component
at the beginning of a
nonzero row
Case 2: an entire row of
zeros (ROZ)
Routh-Hurwitz Criteria:
Supplementary
Examples
References
6.31
Stability
Wail Gueaieb
Verification
Fig. 9 shows the step response of this system. It is clear that the response
goes out of bound, hence the system is unstable.
Step Response
#10 10
2.5
2
Objectives
1.5 Introduction
1 Routh-Hurwitz Criteria
Amplitude
0
Routh-Hurwitz Criteria:
Special Cases
-0.5 Case 1: a zero component
at the beginning of a
nonzero row
-1
Case 2: an entire row of
zeros (ROZ)
-1.5
0 10 20 30 40 50 60
Time (seconds)
Routh-Hurwitz Criteria:
Supplementary
Examples
Figure 9: Step response of the system Stability in State Space
References
System poles: 0.5 ± j 3.1225, ±j 1.4142, ±j, and two poles at −1.
6.32
Stability
Routh-Hurwitz Criteria: Supplementary Examples
Wail Gueaieb
Objectives
Introduction
Example 6.9
Routh-Hurwitz Criteria
Consider the following system, where G(s) =
R(s) E(s) C(s)
K Routh-Hurwitz Table
s(s+7)(s+11)
and K ∈ R. For what values of K +
−
G(s)
Routh-Hurwitz Criteria:
is the system stable, unstable, and marginally Special Cases
stable? Routh-Hurwitz Criteria:
Supplementary
Examples
References
6.33
Stability
Wail Gueaieb
1.6 1
1.4
1.4
Introduction
0.5
1.2
1.2
Routh-Hurwitz Criteria
Amplitude
0
Amplitude
Amplitude
1
1 -0.5
0.8
0.6
0.8 -1 Routh-Hurwitz Table
0.6 -1.5
0.4 0.4 -2 Routh-Hurwitz Criteria:
0.2 0.2 -2.5 Special Cases
0 0 -3
0 1 2 3 4 5 6 7 8 9 10 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 0 10 20 30 40 50 60 70 80 90
Time (seconds) Time (seconds) Time (seconds) Routh-Hurwitz Criteria:
Supplementary
K = 1000 K = 1386 K = 2000 Examples
6.34
Stability
Wail Gueaieb
Objectives
Example 6.10
Introduction
Manually compute the roots of the polynomial Routh-Hurwitz Criteria
4 3 2 Routh-Hurwitz Table
s + 3s + 30s + 30s + 200
Routh-Hurwitz Criteria:
Special Cases
Routh-Hurwitz Criteria:
Supplementary
Examples
References
6.35
Stability
Wail Gueaieb
Remark
p
−b ± b2 − 4ac
The roots of f (x) = ax 2
+ bx + c, with a 6= 0, are x1,2 = .
( 2a
sign(−a) , if x1 < x < x2
Assuming x1 ≤ x2 , then sign(f (x)) =
sign(a) , if x < x1 or x > x2
Objectives
Introduction
Routh-Hurwitz Criteria
Routh-Hurwitz Criteria:
Consider the following RH table corresponding to a certain polynomial, where Special Cases
K ∈ R. For what values of K does the polynomial have roots on the RHS of the
Routh-Hurwitz Criteria:
s-plane? Supplementary
Examples
s2 K +7 ?
s1 K2 + K − 2
s0 5
6.36
Stability
Wail Gueaieb
Objectives
Problem 6.34 of Nise
Introduction
Consider the following system, where G(s) =
K (s−1)(s−2) R(s) E(s) C(s) Routh-Hurwitz Criteria
(s+2)(s2 +2s+2)
and K ∈ R. For what values of +
−
G(s)
Routh-Hurwitz Table
K is the system stable, unstable, and marginally
Routh-Hurwitz Criteria:
stable? Special Cases
Routh-Hurwitz Criteria:
Supplementary
Examples
References
6.37
Stability
Wail Gueaieb
Amplitude
0.3
Amplitude 0.4
-1
0.2
0.1 0.2
-1.5
Routh-Hurwitz Table
0 0
-2 Routh-Hurwitz Criteria:
-0.1 -0.2
Special Cases
-0.2 -0.4 -2.5
0 5 10 15 20 25 30 0 5 10 15 20 25 30 35 40 45 50 0 50 100 150 200 250
Time (seconds) Time (seconds) Time (seconds) Routh-Hurwitz Criteria:
Supplementary
K =1 K = 1.57 K = −4.27 Examples
6.38
Stability
Stability in State Space
Wail Gueaieb
Objectives
For a system represented in state space, Introduction
• The poles of the system are the eigenvalues of the system’s matrix A. Routh-Hurwitz Criteria
• The system’s characteristic equation is det(sI − A) = 0 (the roots of Routh-Hurwitz Table
which are the system’s poles). Routh-Hurwitz Criteria:
Special Cases
The Routh-Hurwitz table is drawn starting using the characteristic equation
Routh-Hurwitz Criteria:
det(sI − A) = 0. Supplementary
Examples
References
6.39
Stability
Wail Gueaieb
Routh-Hurwitz Criteria:
Special Cases
Find the distribution of the system poles on the s-plane.
Routh-Hurwitz Criteria:
Supplementary
Examples
6.40
Stability
Wail Gueaieb
Attention!
• In many examples in the textbook, the author jumps to the conclusion that
the system is stable simply by showing that there are no sign changes in
the first column of the RH table.
• This is not a proper proof that a system is stable.
Objectives
• The number of sign changes in the first column of the RH table only Introduction
indicates the number of roots in the right-hand side of the s plane. Routh-Hurwitz Criteria
• You still need to show that there are no poles at the origin of the plane and Routh-Hurwitz Table
that there are no repeated complex conjugate roots on the imaginary axis Routh-Hurwitz Criteria:
to conclude that a system is stable. Special Cases
Routh-Hurwitz Criteria:
Supplementary
Examples
Examine ch6p1, ch6p2, and ch6p3 (Appendix B), and ch6sp2 (Appendix E).
6.41
Stability
References
Wail Gueaieb
Objectives
Introduction
Routh-Hurwitz Criteria
Nise, N. S. (2015).
Control Systems Engineering. Routh-Hurwitz Table
Routh-Hurwitz Criteria:
Supplementary
Examples
References
6.42