MECE 3350U

Control Systems

Lecture 7
Steady-State Error

1/41
MECE 3350 - C. Rossa 1 / 41 Lecture 7
Videos in this lecture

Lecture: https://youtu.be/Co2K3BaGOdc

Exercise 30: https://youtu.be/rXFcTRi9QXI

Exercise 31: https://youtu.be/3zEXv6krNNU

Exercise 32: https://youtu.be/DgRFSLgK5h4

Exercise 33: https://youtu.be/9aZaIz61kIU

Exercise 34: https://youtu.be/iTTnMuIbsx8

2/41
MECE 3350 - C. Rossa 2 / 41 Lecture 7
Outline of Lecture 7

By the end of today’s lecture you should be able to

• Understand the concept of error and disturbance signals

• Calculate the steady state error of a system

• Analyse the influence of control loop gains

3/41
MECE 3350 - C. Rossa 3 / 41 Lecture 7
Applications

When the steel bars engage in the rolls of the rolling mill, the load on the rolls
increases immediately. How can the speed of the rollers be controlled to
minimize this disturbance?

4/41
MECE 3350 - C. Rossa 4 / 41 Lecture 7
Applications
Automated control of the laser position during eye surgery enables the
ophthalmologist to indicate to the controller where lesions should be inserted.

How can we design a controller that minimizes the transient response of the
positioning system if the retina moves during the surgery?

5/41
MECE 3350 - C. Rossa 5 / 41 Lecture 7
Closed-loop control

6/41
MECE 3350 - C. Rossa 6 / 41 Lecture 7
Open-loop control

In the absence of a controller and without feedback, the disturbance T (s)

directly influences the output Y (s).

An open-loop system operates without feedback and directly generates the

output in response to an input signal.

7/41
MECE 3350 - C. Rossa 7 / 41 Lecture 7
Closed-loop control

A closed-loop system compares the output Y (s) with a desired value R(s).

The error signal E (s) is used by the controller to adjust the actuator.

8/41
MECE 3350 - C. Rossa 8 / 41 Lecture 7
Advantages of closed-loop control

• Decrease sensitivity to variations in the parameters of the process (G(s))

• Improve rejection of the disturbances (T (s))
• Improve noise attenuation (N(s))
• Reduce the steady-state error (E (s))
• Allows for control of the transient response

9/41
MECE 3350 - C. Rossa 9 / 41 Lecture 7
Disturbance

Disturbance is a change in the values of the nominal parameters of a control

system due to external sources.

For H(s) = 1, the response due to disturbance is

G(s)
YD (s) = T (s)
1 + C (s)G(s)
For R(s) = 0 and N(s) = 0.

10/41
MECE 3350 - C. Rossa 10 / 41 Lecture 7
Disturbance

For H(s) = 1, T (s) = 0; and N(s) = 0, the output due to R(s) is

C (s)G(s)
YR (s) = R(s)
1 + C (s)G(s)

For H(s) = 1, T (s) = 0, and R(s) = 0, the response to noise is

C (s)G(s)
YN (s) = − N(s)
1 + C (s)G(s)

11/41
MECE 3350 - C. Rossa 11 / 41 Lecture 7
Disturbance and superposition

The principle of superposition gives the total response as

Y (s) = YR (s) + YN (s) + YT (s)

C (s)G(s) G(s) C (s)G(s)

Y (s) = R(s) + T (s) − N(s)
1 + C (s)G(s) 1 + C (s)G(s) 1 + C (s)G(s)

12/41
MECE 3350 - C. Rossa 12 / 41 Lecture 7
Disturbance and superposition

Defining the error as E (s) = R(s) − Y (s) gives

C (s)G(s) G(s) C (s)G(s)

E (s) = R(s) − R(s) − T (s) + N(s)
1 + C (s)G(s) 1 + C (s)G(s) 1 + C (s)G(s)

 
C (s)G(s) G(s) C (s)G(s)
E (s) = 1− R(s) − T (s) + N(s)
1 + C (s)G(s) 1 + C (s)G(s) 1 + C (s)G(s)

1 G(s) C (s)G(s)
E (s) = R(s) − T (s) + N(s)
1 + C (s)G(s) 1 + C (s)G(s) 1 + C (s)G(s)

13/41
MECE 3350 - C. Rossa 13 / 41 Lecture 7
Poles and zeros
Consider the following function:

s(s + 3)
F (s) =
s 2 + 2s + 5
→ Poles: −1 + 2j, −1 − 2j

→ Zeros: 0, −3

1
Imaginary Axis

-1

-2
-3 -2.5 -2 -1.5 -1 -0.5 0
Real Axis

14/41
MECE 3350 - C. Rossa 14 / 41 Lecture 7
Disturbance rejection

1 G(s) C (s)G(s)
E (s) = R(s) − T (s) + N(s)
1 + C (s)G(s) 1 + C (s)G(s) 1 + C (s)G(s)

To reduce the influence of the disturbance:

→ C (s) must be large to minimize the influence of T (s)
→ C (s) must be small to minimize the influence of N(s)
How to solve this conflict?

→ Make C (s) large at low frequencies

→ Make C (s) small at high frequencies
15/41
MECE 3350 - C. Rossa 15 / 41 Lecture 7
Transient and steady-state response

16/41
MECE 3350 - C. Rossa 16 / 41 Lecture 7
Steady-state error

If N(s) = T (s) = 0, the error is:

1
E (s) = R(s)
1 + C (s)G(s)
The steady state error can be obtained using the final value theorem:

lim e(t) = lim sE (s)

t→∞ s→0

For the closed loop system (step input):

    
1 1 1
ec (∞) = lim s =
1 + C (s)G(s) s 1 + C (0)G(0)

C (0)G(0) is called the "DC gain".

17/41
MECE 3350 - C. Rossa 17 / 41 Lecture 7
Open-loop vs closed-loop

Open-loop: The error is E (s) = R(s) − Y (s)

KKa
E (s) = R(s) − R(s)
(τ s + 1)(τ1 s + 1)
What is the steady-state error?

18/41
MECE 3350 - C. Rossa 18 / 41 Lecture 7
Open-loop vs closed-loop

Closed-loop: The error is E (s) = R(s) − Y (s)

KKa
E (s) = R(s) − R(s)
(τ s + 1)(τ1 s + 1) + KKa
What is the steady-state error?

19/41
MECE 3350 - C. Rossa 19 / 41 Lecture 7
Open-loop vs closed-loop

Open-loop: The steady error is

E (s) = 1 − KKa

Closed-loop: The steady error is

1
E (s) =
1 + KKa

20/41
MECE 3350 - C. Rossa 20 / 41 Lecture 7
Exercise 30
A robotic arm and camera are used to pick fruit as shown in the figure. The
camera is used to close the feedback loop to a micro-controller, which controls
the arm. The transfer function of the process is
1
G(s) =
(s + 10)2
and the controller is a proportional gain so that C (s) = K . Calculate the
steady-state error of the gripper for a step command A as a function of K .

21/41
MECE 3350 - C. Rossa 21 / 41 Lecture 7
Exercise 30 - continued
1
G(s) = , C (s) = K , R(s) =
(s + 10)2

22/41
MECE 3350 - C. Rossa 22 / 41 Lecture 7
Exercise 30 - continued
A 1
 
E (s) =
s 1 + K /(s + 10)2

23/41
MECE 3350 - C. Rossa 23 / 41 Lecture 7
Exercise 31

A hard drive requires a motor to position a read/write head over the tracks of
data on a spinning disk. The motor and head have the transfer function
100
G(s) =
s(0.001s + 1)
The controller is C (s) = K . Calculate K that yields a steady-state error of 0.1
mm for a ramp input of 10 cm/s.

24/41
MECE 3350 - C. Rossa 24 / 41 Lecture 7
Exercise 31 - continued
100
G(s) = , C (s) = K , R(s) =
s(0.001s + 1)

25/41
MECE 3350 - C. Rossa 25 / 41 Lecture 7
Exercise 31 - continued

26/41
MECE 3350 - C. Rossa 26 / 41 Lecture 7
Exercise 32

A submarine has a depth control system as illustrated.

(a) Determine the transfer function F (s) = Y (s)/R(s)

(b) Determine the steady state error for T (s) = 1/s and R(s) = 0
(c) Calculate the response y (t) for R(s) = 1/s when K = K2 = 1 and
1 < K < 10. Select K1 for the fastest response.

27/41
MECE 3350 - C. Rossa 27 / 41 Lecture 7
Exercise 32 - continued
(a) The transfer function Y (s)/R(s)

28/41
MECE 3350 - C. Rossa 28 / 41 Lecture 7
Exercise 32 - continued

(b) The state state error for T (s) = 1/s

29/41
MECE 3350 - C. Rossa 29 / 41 Lecture 7
Exercise 32 - continued
(c) y (t) for R(s) = 1/s, K = K2 = 1 and 1 < K1 < 10 for the fastest response.

KK1
Y (s) = R(s) =
s + K1 (K + K2 )

30/41
MECE 3350 - C. Rossa 30 / 41 Lecture 7
Exercise 33 - Design problem

The position control system of a laser system for ophthalmologic surgery is

shown in the figure. Select the gain K to ensure an appropriate transient
response to a step change in R(s) and the effect of disturbance is minimized.
The steady-state error must be zero. The system is stable provided that
K < 10. Plot the step response of the system for 1 < K < 15 using Matlab.

31/41
MECE 3350 - C. Rossa 31 / 41 Lecture 7
Exercise 33 - continued

32/41
MECE 3350 - C. Rossa 32 / 41 Lecture 7
Exercise 33 - continued

33/41
MECE 3350 - C. Rossa 33 / 41 Lecture 7
Exercise 33 - continued

34/41
MECE 3350 - C. Rossa 34 / 41 Lecture 7
Exercise 34 - Design problem

Consider the speed control system shown.

Determine:

(a) The range of K1 so that the steady state tracking error is e ≤ 1%

(b) KK1 so that the steady state error for T (t) = 2t mrad/s for 0 ≤ t ≤ 5 sec
is less than 0.1 mrad/s.

35/41
MECE 3350 - C. Rossa 35 / 41 Lecture 7
Exercise 34 - continued

36/41
MECE 3350 - C. Rossa 36 / 41 Lecture 7
Exercise 34 - continued

37/41
MECE 3350 - C. Rossa 37 / 41 Lecture 7
Skills check 20 - From 2018 midterm examination

The inverted pendulum position control system has been designed to maintain
the angular position ω(s) of the pendulum at zero.

Calculate the steady state error for a ramp disturbance T (t) = t excitation1 .

1
Answer: ess = 1/(K1 K ) 38/41
MECE 3350 - C. Rossa 38 / 41 Lecture 7
Skills check 21

Find the value of k so that the steady-state error of the the system is 0.1 for a
unit ramp input of r (t) that is r (t) = t.

Answer: k = 672 39/41

MECE 3350 - C. Rossa 39 / 41 Lecture 7
Next episode...

• Transient performance

40/41
MECE 3350 - C. Rossa 40 / 41 Lecture 7