ME153P

System Stability
Stability at first glance

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

How to know if the system is stable?
A system is stable if its output is
bounded for any bounded input.

◦ A system is asymptotically stable is in

the absence if input, the output tends
towards zero irrespective of initial
conditions.
◦ A system is stable if its output is Where
Blue line signal is unbounded
unbounded for any bounded input. Red line signal is bounded this due to
having upper and lower limits

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

How to know if the system is stable?
Stability of the system depends upon poles.

◦ If all the poles are located in the left half of the S-

Plane, then the system is stable
◦ As pole approaches origin then stability
decreases.
◦ When poles (non repeated) are located on
imaginary axis then the system is marginally stable
◦ If poles located in the imaginary axis are
repeated, then the system is unstable
◦ If poles are located in the right hand of S-Plane
then the system is unstable.
◦ The poles which are close to origin are called
dominant poles.

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

CLASSIFICATION OF
SYSTEMS BASED ON
STABILITY
ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]
Classification of systems based on
stability
Absolutely stable system
◦ If the system is stable for all the range of system component values, then it is known as the absolutely
stable system. The open loop control system is absolutely stable if all the poles of the open loop transfer
function present in left half of ‘s’ plane. Similarly, the closed loop control system is absolutely stable if all
the poles of the closed loop transfer function present in the left half of the ‘s’ plane.

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Classification of systems based on
stability
Marginally Stable System
◦ If the system is stable by producing an output signal with constant amplitude and constant frequency of
oscillations for bounded input, then it is known as marginally stable system. The open loop control system
is marginally stable if any two poles of the open loop transfer function is present on the imaginary axis.
Similarly, the closed loop control system is marginally stable if any two poles of the closed loop transfer
function is present on the imaginary axis.

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Classification of systems based on
stability
Conditionally Stable System.
◦ If the system is stable for a certain range of system component values, then it is known as conditionally
stable system.

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Techniques used to Calculate Stability
◦ Routh–Hurwitz stability Criterion
◦ Routh Locus
◦ Bode Plot
◦ Nyquist Plot
◦ Nicholas Chart

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

 ROUTH-HURWITZ
STABILITY CRITERION
Routh-Hurwitz Stability Criterion
Introduced in 1874
◦ Generalizes the previous method
◦ Introduces additional combinations of
coefficients
◦ Based on Sturm's theorem.
Central is the idea of the \Routh Table"

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Routh’s Method
Step 1: Write the polynomial as

Initial layout for Routh table

S4 a4 a2 a0
Step 2: Write the coefficients in 2 rows s3 a3 a1 0
◦ First row starts with an
◦ Second row starts with an-1
s2
◦ Other coefficients alternate between rows s1
◦ Both rows should be same length
s0
◦ Continue until no coefficients are left
◦ Add zero as last coefficient if necessary

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Routh’s Method
Step 3: Complete the third row.
◦ Call the new entries b1, . . . , bk
◦ The third row will be the same length as the first two Initial layout for Routh table
s4 a4 a2 a0
s3 a3 a1 0
◦ The denominator is the first entry from the
previous row. s2 b1 b2 b3
◦ The numerator is the determinant of the entries s1
from the previous two rows:
◦ The first column s0
◦ The next column following the coefficient

◦ If a coefficient doesn't exist, substitute 0.

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Routh’s Method Initial layout for Routh table

s4 a4 a2 a0
Step 4: Treat each following row in the
s3 a3 a1 0
same way as the third row
𝑎4 𝑎2 𝑎4 𝑎0 𝑎4 0
− 𝑎 𝑎 − 𝑎 −
◦ There should be n + 1 rows total, s2 3 1
= 𝑏1 3 0
= 𝑏2
𝑎3 0
= 𝑏1
𝑎3 𝑎3
including the first row. 𝑎3
𝑎3 𝑎2 𝑎3 0 𝑎3 0
− 𝑏 𝑏 − −
s1 1 2
= 𝑐1
𝑏1 0
=0
𝑏1 0
=0
𝑏1 𝑏1 𝑏1
𝑏1 𝑏2 𝑏1 0 𝑏1 0
− − −
s0 𝑐1 𝑏2
= 𝑑1
𝑐1 0
=0
𝑐1 0
=0
𝑐1 𝑐1 𝑐1

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Example 1
Can feedback destabilize the
suspension system?

Is it stable for any k > 0???

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Example 1
Can feedback destabilize the Solution:
suspension system?

Is it stable for any k > 0??? We need to find the bi.

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Example 1
Can feedback destabilize the Start by calculating the coefficients in
suspension system? the first row:

Is it stable for any k > 0???

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Example 1
Can feedback destabilize the The coefficients for the next row are
suspension system?

Is it stable for any k > 0???

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Example 1
Can feedback destabilize the There is only one non-zero coefficient in
suspension system? the last row.

Is it stable for any k > 0???

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Example 1
Can feedback destabilize the Conclusion:
suspension system?
No matter what k > 0 is, the first column
is always positive.
◦ No sign changes for any k.
Is it stable for any k > 0??? ◦ Stable for any k.
Feedback CANNOT destabilize the
suspension system.

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Stability of Quadratics
What about a simple second-order system?

We know the poles are at

Calculate

The Routh table is

Thus a quadratic is stable if and only if both coeficients are positive.

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Stability of 3rd order systems
Now consider a third order system:

The Routh table is

So for 3rd order, stability is equivalent to:

◦ a>0
◦ b>0
◦ b > c/a

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Example 2:
The Closed-Loop Transfer Function is

We close the loop with a gain of size k

Controller: K(s) = k
But this is a third order system!

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Example 2
For the third-order system,

we require
◦ a > 0, which means 10 > 0 The last requirement implies
◦ c > 0, which means 30 + k > 0
◦ b > c/a
So our gain is limited to k < 280

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Limited Special Cases
Case 1: When All the elements in any
one row of the Routh table is Zero
Solution:
The Polynomial whose coefficients are
the elements if the row just above the
row of zeros in the Routh table is called
as “Auxiliary Polynomial.” The order of
auxiliary polynomial is always even.

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Example 3
Identify if the system is stable and if not 𝑠6 1 5 2 -8
how many poles that make the system 𝑠5 1 3 -4 0
unstable.
𝑠4 2 6 -8 0
𝑠 6 + 𝑠 5 + 5𝑠 4 + 3𝑠 3 + 2𝑠 2 − 4𝑠 − 8 = 0
𝑠3 0 0 0 0
𝑠2
𝑠1
𝑠0

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Example 3
Auxiliary Polynomial 𝑠6 1 5 2 -8
𝐴 𝑠 = 2𝑠 4 + 6𝑠 2 − 8 = 0 𝑠5 1 3 -4 0
𝑠4 2 6 -8 0
𝑠3 0 0 0 0
𝑠2
𝑠1
𝑠0

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Example 3
Differentiating 𝑠6 1 5 2 -8
𝑑𝐴(𝑠) 𝑠5 1 3 -4 0
= 8𝑠 3 + 12𝑠 = 0
𝑑𝑠 𝑠4 2 6 -8 0
𝑠3 0 0 0 0
> 𝑠3 8 12 0 0
𝑠2
𝑠1
𝑠0

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Example 3
Complete Routh Table 𝑠6 1 5 2 -8
𝑠5 1 3 -4 0
Since there is 1 sign change in the first 𝑠4 2 6 -8 0
column, there is only one pole that is in 𝑠3 0 0 0 0
the right side of the S-Plane > 𝑠3 8 12 0 0
𝑠2 3 -8 0 0
𝑠1 33.33 0 0 0
𝑠0 -8 0 0 0

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Limited Special Cases
Case 2: When the first term in any row of Solution:
Routh table is Zero, while the rest of the 1
row has at least one nonzero term. 1. Substitute in characteristic equation
∈
and form Routh table again.
2. Replace 0 with ∈

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Example 4
Identify if the system is stable and if not
how many poles that make the system 𝑠4 1 2 5
unstable.
𝑠3 1 -2 0
𝑠 4 + 𝑠 3 + 2𝑠 2 − 2𝑠 − 5 = 0
𝑠2 0 5 0
𝑠1
𝑠0 5

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Example 4
1
Substitute in characteristic equation
∈
and form Routh table again. 𝑧4 5 2 1
𝑧3 2 1 0
4 3 2 𝑧2
1 1 1 1
+ +2 −2 −5=0 𝑧1
∈ ∈ ∈ ∈
𝑧0

Modified Characteristic Equation

5𝑧 4 + 2𝑧 3 + 2𝑧 2 + 𝑧 + 1 = 0

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Example 4
Complete the New Routh Table
𝑧4 5 2 1
Since there is 2 sign changes in the first 𝑧3 2 1 0
column, there are 2 poles that are in the 𝑧2 -0.5 1 0
right-hand side of the S-Plane thus 𝑧1 5 0 0
making the system unstable.
𝑧0 1 0 0

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Example 4 (Alternate Solution)
Identify if the system is stable and if not
how many poles that make the system 𝑠4 1 2 5
unstable.
𝑠3 1 -2 0
𝑠 4 + 𝑠 3 + 2𝑠 2 − 2𝑠 − 5 = 0
𝑠2 0 5 0
𝑠1
𝑠0 5

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Example 4 (Alternate Solution)
Replace 0 with ∈
𝑠4 1 2 5
2 ∈ −5 𝑠3 1 -2 0
lim
∈ ∈ 𝑠2 ∈ 5 0
5
lim 2 − 𝑠1 −∞
∈ ∈
−∞ 𝑠0 5

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

Example 4 (Alternate Solution)
Since there is 2 sign changes in the first
column, there are 2 poles that are in the 𝑠4 1 2 5
right-hand side of the S-Plane thus
𝑠3 1 -2 0
making the system unstable.
𝑠2 ∈ 5 0
𝑠1 −∞
𝑠0 5

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

MATLAB Implementation

By using a fuction named rhStabilityCriterion.m located at

https://www.mathworks.com/matlabcentral/fileexchange/17483-
routh-hurwitz-stability-criterion

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

End of Presentation
Thank you!

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]

REFERENCES:
◦ Schaum’s Outline Series “Feedback and Control Systems”, 2nd edition, by DiStefano III, J.J., Stubberud
A.R., and Williams I.J.
◦ Control Systems Engineering, 3rd edition, by Nise N.S.
◦ Control Systems Engineering; Principles and Design, 2nd edition, by Gopal M.
◦ Feedback Control Systems, 4th edition, by C.L.Phillips and R.D.Harbor.
◦ https://www.tutorialspoint.com/control_systems/control_systems_stability.htm#:~:text=If%20the%20system
%20is%20stable,half%20of%20's'%20plane
◦ http://control.asu.edu/Classes/MAE318/318Lecture10.pdf

ME153P- CONTROL SYSTEMS ENGINEERING [JADCNohay]