1

Digital Control Systems (DCS)

Lecture-3
Modeling of Digital Control
Systems
Lecture Outline
2
 Sampling Theorem

 ADC Model

 DAC Model

 Combined Models
 Sampling Theorem
3
 Sampling is necessary for the processing of
analog data using digital elements.

 Successfuldigital data processing requires that

the samples reflect the nature of the analog
signal and that analog signals be recoverable
from a sequence of samples.
 Sampling Theorem
4
 Following figure shows two distinct waveforms
with identical samples.

 Obviously, faster sampling of the two

waveforms would produce distinguishable
sequences.
 Sampling Theorem
5
 Thus, it is obvious that sufficiently fast
sampling is a prerequisite for successful digital
data processing.

 The sampling theorem gives a lower bound on

the sampling rate necessary for a given band-
limited signal (i.e., a signal with a known
finite bandwidth)
 Sampling Theorem
6
 The band limited signal with

 canbe reconstructed from the discrete-time

∞
waveform ∗
𝑓 ( 𝑡 )= ∑ 𝑓 ( 𝑡 ) 𝛿(𝑡 −𝑘𝑇 )
𝑘=− ∞

 ifand only if the sampling angular frequency

satisfies the condition 𝜔 𝑠 > 2 𝜔𝑚
 Selection of Sampling
7
Frequency
 A given signal often has a finite “effective bandwidth”
beyond which its spectral components are negligible.

 This allows us to treat physical signals as band limited

and choose a suitable sampling rate for them based on
the sampling theorem.

 In practice, the sampling rate chosen is often larger than

the lower bound specified in the sampling theorem.

 A rule of thumb is to choose as

𝜔 𝑠 =𝑘 𝜔 𝑚 ,5 ≤ 𝑘 ≤10
 Selection of Sampling
8
Frequency
𝜔 =𝑘 𝜔 ,5 ≤ 𝑘 ≤10
𝑠 𝑚

 The choice of depends on the application.

 In many applications, the upper bound on the

sampling frequency is well below the capabilities
of state-of-the-art hardware.

 A closed-loop control system cannot have a

sampling period below the minimum time
required for the output measurement; that is, the
sampling frequency is upper-bounded by the
sensor delay.
 Selection of Sampling
9
Frequency
 For example, oxygen sensors used in
automotive air/fuel ratio control have a sensor
delay of about 20 ms, which corresponds to a
sampling frequency upper bound of 50 Hz.
 Anotherlimitation is the computational time
needed to update the control.

 This is becoming less restrictive with the

availability of faster microprocessors but must
be considered in sampling rate selection.
 Selection of Sampling
1
Frequency 0
 For a linear system, the output of the system has
a spectrum given by the product of the frequency
response and input spectrum.

 Because the input is not known a priori, we must

base our choice of sampling frequency on the
frequency response.
Selection of Sampling Frequency (1st Order
Systems) 1
 The frequency response of first order system1is
𝐾
𝐻 ( 𝑗 𝜔 )=
𝑗 𝜔 /𝜔 𝑏 +1
 where K is the DC gain and is the system
bandwidth.

 Time constant and 3db bandwidth relationship

1
𝜔𝑏=
𝑇
1
𝑓 3 𝑑𝑏 =
2𝜋𝑇
Selection of Sampling Frequency (1st Order
Systems) 1
1 2
𝐺 ( 𝑠) =
3 𝑠 +1
Bode Diagram
0

-5
1 System: sys
𝜔 𝑏 = =0.33 𝑟𝑎𝑑/ 𝑠𝑒𝑐 Frequency (rad/s): 0.33
Magnitude (dB)

-10 𝑇 Magnitude (dB): -2.97

-15

-20 7 𝜔𝑏 =2.31 𝑟𝑎𝑑/𝑠𝑒𝑐 System: sys

Frequency (rad/s): 2.31
Magnitude (dB): -16.9
-25

-30
0
Phase (deg)

-45

-90
-2 -1 0 1
10 10 10 10
Frequency (rad/s)
Selection of Sampling Frequency (1st Order
Systems) 1
3
 The frequency response amplitude drops below
the DC level by a factor of about 10 at the
frequency .

 If we consider , the sampling frequency is chosen

as 𝜔 =7 𝑘 𝜔 ,5 ≤ 𝑘 ≤ 10
𝑠 𝑏

 OR 𝜔 𝑠 =𝑘 𝜔 𝑏 ,35 ≤𝑘 ≤ 70
Selection of Sampling Frequency (2nd Order
Systems) 1
4
 The frequency response of second order system
is
K
H ( j ) 
j 2 / n  1  ( / n ) 2

The bandwidth of the system isapproximated

d  n 1  
2
 by
the damped natural frequency

 Using a frequency of as the maximum significant

frequency, we choose the sampling frequency as
𝜔 𝑠 =𝑘 𝜔 𝑑 , 35 ≤ 𝑘 ≤ 70
 Example-1 1
5
 Given a first-order system of bandwidth 10 rad/s,
select a suitable sampling frequency and find the
corresponding sampling period.
Solution

𝜔 𝑏 =10 𝑟𝑎𝑑/𝑠𝑒𝑐

 We know
𝜔 𝑠 =𝑘 𝜔 𝑏 ,35 ≤𝑘 ≤ 70

 Choosing k=60
𝜔 𝑠 =60 𝜔𝑏 =600 𝑟𝑎𝑑/𝑠𝑒𝑐
 Example-1 1
6
 Corresponding sapling period is calculated as

2 𝜋 2 ×3141
𝑇= = =0.01 𝑠𝑒𝑐
𝜔𝑠 600
 Example-2 1
7
 Fort he following first-order system select a
suitable sampling frequency and find the
corresponding sampling period.
10
𝐺 ( 𝑠) =
𝑠+ 1
 Example-3 1
8
 Consider the following second order transfer
function. Select a suitable sampling period for
the system. 16
G (s)  2
s  8s 16
 Example-4 1
9
 A closed-loop control system must be designed
for a damping ratio of about 0.7, and an
undamped natural frequency of 10 rad/s. Select
a suitable sampling period for the system if the
system has a sensor delay of 0.02 sec.
Solution

 Let the sampling frequency be

𝜔 𝑠 ≥ 35 𝜔𝑑
 s 35n 1   2

 s 35 10 1  0.7 2

 Example-4 2
 s 249.95 rad / s 0
 The corresponding sampling period is
2 ×3.141
𝑇≤
249.95

A suitable choice is T = 20 ms because this is

equal to the sensor delay.
 Home Work 2
1
 A closed-loop control system must be designed
for a damping ratio of about 0.7, and an
undamped natural frequency of 10 rad/s. Select
a suitable sampling period for the system if the
system has a sensor delay of 0.03 sec.
 Home Work 2
2
 The following open-loop systems are to be digitally feedback-
controlled. Select a suitable sampling period for each if the
closed-loop system is to be designed for the given
specifications.
 Digital Control Systems
2
3
 A common configuration of digital control system
is shown in following figure.
 ADC Model
2
 Assume that 4
 ADC outputs are exactly equal in magnitude to their
inputs (i.e., quantization errors are negligible)
 The ADC yields a digital output instantaneously
 Sampling is perfectly uniform (i.e., occur at a fixed rate)

 Then the ADC can be modeled as an ideal sampler

with sampling period T.
T

u(t) u*(t)

t t
0 0
Sampling Process
T
u(t) u*(t)

u*(t)
u(t)
δT(t)

× ＝

t t t
0 0 0
Modulation modulating modulated
signal pulse(carrier) wave


u (t )  u (t ) (t  kT )
*

k 0
 DAC Model
2
 Assume that 6
 DAC outputs are exactly equal in magnitude to their inputs.
 The DAC yields an analog output instantaneously.
 DAC outputs are constant over each sampling period.
u(k)
u(t)
uh(t)

 Then the input-output relationship of the DAC is given by

𝑢 ( 𝑘 ) 𝑍 𝑂𝐻 𝑢 h (𝑡 )=𝑢 ( 𝑘 ) , 𝑘𝑇 ≤ 𝑡 ≤ ( 𝑘+1 ) 𝑇
→
 DAC Model
2
 Unit impulse response of ZOH 7

 Thetransfer function can then be obtained by

Laplace transformation of the impulse
response.
 DAC Model
2
 As shown in figure the impulse response is 8
a unit
pulse of width T.

 A pulse can be represented as a positive step at

time zero followed by a negative step at time T.

 Using the Laplace transform of a unit step and the

time delay theorem 1 for Laplace transforms, − 𝑇𝑠
𝑒
L { 𝑢 (𝑡 ) } = L { − 𝑢(𝑡 −𝑇 ) }=−
𝑠 𝑠
 DAC Model
2
− 𝑇𝑠
1 𝑒 9
L { 𝑢 (𝑡 ) } = L { − 𝑢(𝑡 −𝑇 ) }=−
𝑠 𝑠
 Thus, the transfer function of the ZOH is
−𝑇𝑠
1 −𝑒
𝐺 𝑍𝑂𝐻 ( 𝑠)=
𝑠
 DAC, Analog Subsystem, and ADC Combination
Transfer Function
3
0
 The cascade of a DAC, analog subsystem, and
ADC is shown in following figure.

 Because both the input and the output of the

cascade are sampled, it is possible to obtain its z-
domain transfer function in terms of the transfer
functions of the individual subsystems.
 DAC, Analog Subsystem, and ADC Combination
Transfer Function
3
1
 Using the DAC model, and assuming that the
transfer function of the analog subsystem is G(s),
the transfer function of the DAC and analog
subsystem cascade is

𝐺 𝑍𝐴 ( 𝑠 )=𝐺 𝑍𝑂𝐻 (𝑠)𝐺(𝑠)

−𝑇𝑠
1 −𝑒
𝐺 𝑍𝐴 ( 𝑠 )= 𝐺( 𝑠)
𝑠
 DAC, Analog Subsystem, and ADC
Combination 3
2
Transfer Function
( ) 1 −𝑒
−𝑇𝑠
𝐺 𝑍𝐴 𝑠 = 𝐺( 𝑠)
 𝑠
The corresponding impulse response is
−𝑇𝑠
𝐺(𝑠) − 𝐺(𝑠) 𝑒
𝐺 𝑍𝐴 ( 𝑠 )=
𝑠
−𝑇𝑠
𝐺(𝑠) 𝐺(𝑠) 𝑒
𝐺 𝑍𝐴 ( 𝑠 )= −
𝑠 𝑠

 The impulse response is the analog system step

response minus a second step response delayed
by one sampling period.
DAC, Analog Subsystem, and ADC
Combination 3
3
Transfer Function −𝑇𝑠
𝐺(𝑠) 𝐺(𝑠) 𝑒
𝐺 𝑍𝐴 ( 𝑠 )= −
𝑠 𝑠
 DAC, Analog Subsystem, and ADC
Combination 3
4
Transfer Function −𝑇𝑠
𝐺(𝑠) 𝐺(𝑠) 𝑒
𝐺 𝑍𝐴 ( 𝑠 )= −
𝑠 𝑠
 Inverse Laplace yields
𝑔 𝑍𝐴 ( 𝑡 ) =𝑔 𝑠 (𝑡 ) − 𝑔 𝑠 (𝑡 −𝑇 )

 Where
 DAC, Analog Subsystem, and ADC
Combination 3
5
Transfer Function
𝑔 ( 𝑡 ) =𝑔 (𝑡 ) − 𝑔 (𝑡 −𝑇 )
𝑍𝐴 𝑠 𝑠
 The analog response is sampled to give the
sampled impulse response

𝑔 𝑍𝐴 ( 𝑘𝑇 ) =𝑔 𝑠 ( 𝑘𝑇 )− 𝑔 𝑠 (𝑘𝑇 −𝑇 )

 By z-transforming, we can obtain the z-transfer

function of the DAC (zero-order hold), analog
subsystem, and ADC (ideal sampler) cascade.
 DAC, Analog Subsystem, and ADC
Combination 3
6
Transfer𝑔Function
( 𝑘𝑇 ) =𝑔 ( 𝑘𝑇 )− 𝑔 (𝑘𝑇 −𝑇 )
𝑍𝐴 𝑠 𝑠
 Z-Transform is given as
𝐺 𝑍𝐴𝑆 ( 𝑧 )=(1− 𝑧 −1 ) Z {𝑔❑∗
𝑠 (𝑡 ) }

[ { }]
∗
−1 −1 𝐺( 𝑠)
𝐺 𝑍𝐴𝑆 ( 𝑧 )=(1− 𝑧 )Z L
𝑠

 The * in above equation is to emphasize that sampling of

a time function is necessary before z-transformation.

 Having made this point, the equation can be rewritten

more concisely as
𝐺 𝑍𝐴𝑆 ( 𝑧 )=(1− 𝑧 ) Z
𝐺(𝑠 )
−1
𝑠 [ ]
 Example-3
3
7 the
 Find GZAS(z) for the cruise control system for
vehicle shown in figure, where u is the input force,
v is the velocity of the car, and b is the viscous
friction coefficient.

Solution

 The transfer function 𝑉 (system

of 𝑠) 1is given as
𝐺 ( 𝑠) = =
𝑈 ( 𝑠) 𝑀𝑠+𝑏

𝐾 𝐾 /𝜏
𝐺 ( 𝑠) = =
𝜏 𝑠+1 in𝑠+1/
 Re-writing transfer function 𝜏
standard form
 Example-3
3
𝐾 /𝜏 8
𝐺 ( 𝑠) =
𝑠+ 1/ 𝜏
 Where and
 Now we know
−1
𝐺 𝑍𝐴𝑆 ( 𝑧 )=(1− 𝑧 ) Z [
𝐺(𝑠 )
𝑠 ]
 Therefore, 𝐺 (𝑠 ) 𝐾 /𝜏
=
𝑠 𝑠 ( 𝑠+1/ 𝜏 )
 The corresponding partial fraction expansion is

𝑠
=( )[
𝐺 (𝑠 ) 𝐾
𝜏
𝜏
−
𝜏
𝑠 𝑠+1 /𝜏 ]
 Example-3
3
9
−1
𝐺 𝑍𝐴𝑆 ( 𝑧 )=(1− 𝑧 ) Z
[( ) {
𝐾
𝜏
𝜏
−
𝜏
𝑠 𝑠+1/ 𝜏 }]
 Using
the z-transform table, the desired z-
domain transfer function is
−1
𝐺 𝑍𝐴𝑆 ( 𝑧 )=(1− 𝑧 ) Z 𝐾
1
−
1
[ {
𝑠 𝑠+ 1/ 𝜏 }]
𝐺 𝑍𝐴𝑆 ( 𝑧 )=
𝑧 −1
𝑧
𝐾
[ {𝑧
−
𝑧
𝑧 − 1 𝑧 −𝑒 −𝑇 /𝜏 }]
[ {
𝐺 𝑍𝐴𝑆 ( 𝑧 )= 𝐾 1−
𝑧 −1
𝑧 − 𝑒− 𝑇 / 𝜏 }]
Example-3
4
0
[ {
𝐺 𝑍𝐴𝑆 ( 𝑧 )= 𝐾 1−
𝑧 −1
𝑧 − 𝑒− 𝑇 / 𝜏 }]
𝑇
−
𝜏
𝑧 − 𝑒 − 𝑧+1
𝐺 𝑍𝐴𝑆 ( 𝑧 )=𝐾 −𝑇 /𝜏
𝑧 −𝑒

𝑇
−
𝜏
1 −𝑒
𝐺 𝑍𝐴𝑆 ( 𝑧 )=𝐾 − 𝑇 /𝜏
𝑧−𝑒
 Example-4
4
1
 Find GZAS(z) for the vehicle position control system,
where u is the input force, y is the position of the
car, and b is the viscous friction coefficient.
𝑏 𝑦˙ 𝑦

Solution

 The transfer function of system is given as

𝑌 ( 𝑠) 1
𝐺 ( 𝑠) = =
𝑈 ( 𝑠) 𝑠( 𝑀𝑠+ 𝑏)

 Re-writing transfer function in standard form

𝐾 𝐾 /𝜏
𝐺 ( 𝑠) = =
𝑠 (𝜏 𝑠+ 1) 1
𝑠( 𝑠+ )
𝜏
 Example-4
4
𝐾 /𝜏 2
𝐺 ( 𝑠) =
1
𝑠 (𝑠 + )
 𝜏
Where and
 Now we know
−1
𝐺 𝑍𝐴𝑆 ( 𝑧 )=(1− 𝑧 ) Z [
𝐺(𝑠 )
𝑠 ]
 Therefore, 𝐺 (𝑠 ) 𝐾 /𝜏
= 2
𝑠 𝑠 ( 𝑠+1/ 𝜏 )
 The corresponding partial fraction expansion is
𝐺 (𝑠 )
𝑠 [1 𝜏
=𝐾 2 − +
𝑠
𝜏
𝑠 𝑠 +1/ 𝜏 ]
 Example-4
4
3 be
 The desired z-domain transfer function can
obtained as −1 1 𝜏
𝐺 𝑍𝐴𝑆 ( 𝑧 )=(1− 𝑧 ) Z 𝐾 2 − +
[
𝜏
𝑠 𝑠 𝑠 +1/ 𝜏 ]
𝐺 𝑍𝐴𝑆 ( 𝑧 )=
𝑧 −1
𝑧
𝐾
𝑧
[
( 𝑧 − 1)
2
−
𝜏𝑧
𝑧 − 1
+
𝜏𝑧
𝑧 −𝑒
− 𝑇 /𝜏
]
𝐺 𝑍𝐴𝑆 ( 𝑧 )=𝐾
1
𝑧 −1 [
−𝜏 +
𝜏 ( 𝑧 − 1)
𝑧−𝑒
− 𝑇 /𝜏 ]
[ ( 1− 𝜏 +𝜏 𝑒 ) 𝑧 +[ 𝜏 − 𝑒 ]
]
𝑇 𝑇
− −
𝜏 𝜏
(𝜏 +1)
𝐺 𝑍𝐴𝑆 ( 𝑧 )= 𝐾 − 𝑇 /𝜏
( 𝑧 − 1)( 𝑧 − 𝑒 )
 Example-5
4
4 in
 Find GZAS(z) for the series R-L circuit shown
Figure with the inductor voltage as output.