EE 312/ lecture no.

5/ Time domain performance specifications

University of Benghazi
Faculty of Engineering
Electrical and Electronic Engineering Department
----------------------------------------------------------------------------------------------------------------------------
Control systems EE312 Fall 2011/2012
Course Format and Assessment
The course will be delivered in 56 hours, with 4 home works. Final exam will be in Feb 2012.
Course material will include handouts.
----------------------------------------------------------------------------------------------------------------------------
Instructor:
Dr. Awad Shamekh
----------------------------------------------------------------------------------------------------------------------------

Lecture no. (5) Date: 25/12/2011

Learning Outcomes of this lecture

The student will understand:
• Test signal concepts
• Time domain performance specifications
----------------------------------------------------------------------------------------------------------------------------
References:
- Modern control systems, R.C. Dorf
- Control systems Engineering, Norman Nise
- Linear control systems engineering, Morris Driels
----------------------------------------------------------------------------------------------------------------------------

Time Domain Performance Specifications

Due to the fact that the actual input signal of the system is usually unknown, a
standard test input signal is normally chosen. The standard signals commonly used
are; the step input, the ramp input, and the parabolic input. See Table (1) & Figure (1).

Table (1)

1
 EE 31
12/ lecture no. 5/ Tim
me domain performan
p nce specificaations

Figure (1)

Thee step inputt signal is the

t easiest to t generate and evaluaate and usuually chosen
n for
me domain performancce specificattions are im
perfformance teests. The tim mportant inddices
becaause controol system aree inherentlyy time domaain system.

Let us considerr a single looop second order system

m and deterrmine its response to a unit
stepp. A closed loop
l feedbaack system is
i shown in figure (2)

Figure (2)
Thee closed loopp output is:

G(s)
C (s) = R(s) (1)
1 + G(s)
or
k
C (s) = R(s) (2)
s + ps
2
p +k
Utillizing the geeneralized notation,
n equuation (1) can
c be rewriitten as:

ωn
2

C (s) = 2 R( s ) (3)
s + 2ζω n s + ω n
2

In equation
e (3)
ζ denotes
d the damping
d rattion, and ω n signifies the
t natural frequency.
f W
With a unit step
equation (3) beecomes:

ωn
2

C ( s) =
s ( s + 2ζω n s + ω n )
2 2

For which the transient

t ouutput, as obttained from the Laplacee transform table is:

2
 EE 312/ lecture no. 5/ Time domain performance specifications

1
C ( s) = 1 − e −ζω t sin(ω n β t + θ )
n
(4)
β
Where

β
β = 1− ζ 2 , θ = tan −1
ζ
The transient response of this second order system for various values of damping
ration is shown in Figure (3).

Figure(3)
As ζ decreases the closed loop roots approach the imaginary axis and the response
becomes increasingly oscillatory.

Response of second order system

Consider the general form of the second order system which is given in equation (5):

(bm s m + bm−1 s m−1 + ... + b1 s + bo )

Tf = (5)
s 2 + a1 s + ao
The roots of equation (5), which represent the system poles, are given by:

− a1 ± a12 − 4ao
s1, 2 = (6)
2

If the general notations are applied, the following parameters are obtained:

3
 EE 312/ lecture no. 5/ Time domain performance specifications

a1 = 2ζω n , a 0 = ω n2 , thus equation (6) can be given as:

− 2ζω n ± (2ζω n ) 2 − 4ω n2
s1, 2 = (7)
2
If the characteristic roots are:
• different real numbers, the system is said to be overdamped, Figure (4.a).
• equal real numbers, the system is said to be critically damped, Figure (4.b).
• complex (conjugate) numbers, the system said to be underdamped, Figure
(4.c).
• purely imaginary, the system said to be undamped, Figure (4.d).

Figure (4.a), overdamped system

Figure (4.b), critically damped system

Figure (4.c), underdamped system

4
 EE 312/ lecture no. 5/ Time domain performance specifications

Figure (4.d), undamped system

Standard performance measures are usually defined in terms of the step response of a
system as shown in Figure(5).

Figure (5)

The swiftness of the response is measured by the rise time, Tr , and the peak time, T p .
For underdamped systems with an overshoot, 0 − 100 % rise time is a useful index. If
the system is overdamped, then the peak time is not defined and 10 − 90 % rise time
is normally used. The similarity with which the actual response matches the step input
is measured by the percent overshoot, P.O. and settling time Ts .

For a unit step input, the peak value of time response is maximum overshoot, M pt .
Meanwhile the settling time, Ts , is defined as the time required for the time required
for the system to settle within a certain percentage δ of the input amplitude. This band
of ± δ is shown in Figure (5).

5
 EE 312/ lecture no. 5/ Time domain performance specifications

ζπ
−
1−ζ 2
M pt = 1 + e (7)

ζπ
−
1−ζ 2
P.O. = 100e (8)

For the second order system with closed loop damping ratio (ζ ) and natural
frequency (ωn ) , the response remains within 2% after four time constants or:
4
Ts = 4τ = (9)
ζω n

Remark : the equivalent time constant of a second order system can be defined as:

1
τ= (10)
ζω n

Therefore the settling time is defined as four time constant of the dominant response.
The steady state error of the system can be measured on the step response of the
system as shown (5). The peak time relationship for this second order system is:

π π
Tp = = (11)
ωn 1 − ζ 2
ωd

Where ω d is the damped natural frequency and given by:

ωd = ωn 1 − ζ 2 (12)

Dominant poles:

Consider the transfer function given below:

C ( s) 100
=
R ( s ) ( s + 1)( s + 100)
In this system, one pole is located close to the imaginary axis at s = −1 while the
other is located far from the imaginary axis at s = −100 . Assuming the system
subjected to a unit impulse, the output may be written as:

100 A B
c( s ) = = +
( s + 1)( s + 100) ( s + 1) ( s + 100)

6
 EE 312/ lecture no. 5/ Time domain performance specifications

100 100
Calculating A = &B=− , the output maybe rewritten as:
99 99
C ( s ) = 1.01e − t − 1.01e −100 t = 1.01(e − t − e −100 t )
For a given value of t, the term e −100t will be considerably smaller than the term e − t
and for all but t → 0 the contribution from e −100t maybe neglected. This means that
closed loop poles close to the imaginary axis dominate those far away, and in many
cases the response may be approximated by the response of the closed loop poles
closest to the imaginary axis. This argument applies equally to complex poles as it
does to real poles.