Control System Modelling Case Studies

Purpose
The case study is a Research Paper and is to be used as a tool to assist you in developing
new skills in research, analysis and technical communication. The case study report is
intended to encourage the development of your ability to critically review a chosen topic
relevant to the unit of Control Systems and your ability to prepare a formal research report
on an assigned topic. These skills will be very valuable for your professional development
and will greatly assist you in taking on the responsibilities of a professional engineer.

You are provided with an opportunity to practice and develop your skills in critical thinking,
develop a better understanding of the subject content in the unit being studied, and to
develop ways of improving your study methods. You are expected to write a technical report
which expresses these ideas in a formal way. As a result you will also be able to develop
and practice your technical writing skills.


At the end of the case study you will submit a completed report on the Blackboard that will
be assessed by your lecturer and this assessment will form a part of the assessment for the
unit.

What is expected in the case study report
To assist you with the preparation of the case study, there are sample case studies attached
to this instruction sheet. These case study examples are copied from the textbook Control
Systems Engineering by Norman Nise. You dont need this text book. Simply use the
attached document as a guide. You will prepare a report:
On a topic assigned to you.
You will explain the problem with the help of a block diagram and develop a transfer
function fully documenting all aspects of the mathematical modelling.
Provide a block diagram of multiple loops if necessary and show a single block or
closed-loop system which is simplified from the multi-loop diagram.
Explain the above process with the help of references and list the references at the
end of the report.
You are not required to create an original mathematical model. You can use
available resources (textbook, online research articles or published industrial case
studies) and summarize the modelling process in your own words.
12 Chapter 1 Introduction
Other Considerations
The three main objectives of control system analysis and design have already been
enumerated. However, other important considerations must be taken into account. For
example, factors affecting hardware selection, such as motor sizing to fulfill power
requirements and choice of sensors for accuracy, must be considered early in the design.
Finances are another consideration. Control system designers cannot create
designs without considering their economic impact. Such considerations as budget
allocations and competitive pricing must guide the engineer. For example, if your
product is one of a kind, you may be able to create a design that uses more expensive
components without appreciably increasing total cost. However, if your design will be
used for many copies, slight increases in cost per copy can translate into many more
dollars for your company to propose during contract bidding and to outlay before sales.
Another consideration is robust design. System parameters considered con-
stant during the design for transient response, steady-state errors, and stability
change over time when the actual system is built. Thus, the performance of the
system also changes over time and will not be consistent with your design. Un-
fortunately, the relationship between parameter changes and their effect on per-
formance is not linear. In some cases, even in the same system, changes in parameter
values can lead to small or large changes in performance, depending on the system's
nominal operating point and the type of design used. Thus, the engineer wants to
create a robust design so that the system will not be sensitive to parameter changes.
We discuss the concept of system sensitivity to parameter changes in Chapters 7 and
8. This concept, then, can be used to test a design for robustness.
Introduction to a Case Study
Now that our objectives are stated, how do we meet them? In this section we will
look at an example of a feedback control system. The system introduced here will
be used in subsequent chapters as a running case study to demonstrate the
objectives of those chapters. A colored background like this will identify the
case study section at the end of each chapter. Section 1.5, which follows this first
case study, explores the design process that will help us build our system.
Antenna Azimuth: An Introduction to Position Control Systems
A position control system converts a position input command to a position output
response. Position control systems find widespread applications in antennas, robot
arms, and computer disk drives. The radio telescope antenna in Figure 1.8 is one
example of a system that uses position control systems. In this section, we will look in
detail at an antenna azimuth position control system that could be used to position a
radio telescope antenna. We will see how the system works and how we can effect
changes in its performance. The discussion here will be on a qualitative level, with the
objective of getting an intuitive feeling for the systems with which we will be dealing.
An antenna azimuth position control system is shown in Figure 1.9(a), with a
more detailed layout and schematic in Figures 1.9(6) and 1.9(c), respectively.
Figure 1.9(d) shows a functional block diagram of the system. The functions are
shown above the blocks, and the required hardware is indicated inside the blocks.
Parts of Figure 1.9 are repeated on the front endpapers for future reference.
FIGURE 1.8 The search for
extraterrestrial life is being
carried out with radio antennas
like the one pictured here. A
radio antenna is an example of
a system with position
controls.
Case Study
The purpose of this system is to have the azimuth angle output of the antenna,
9
0
(t), follow the input angle of the potentiometer, 0,-(f). Let us look at Figure 1.9(d)
and describe how this system works. The input command is an angular displace-
ment. The potentiometer converts the angular displacement into a voltage.
Antenna
Desired
azimuth ang
input
(a)
Potentiometer
Antenna
Differential amplifier
and power amplifier
Potentiometer
Potentiometer
Amplifiers Motor
Differential
and
power
amplifier
K
Armature
resistance
Armature
Fixed field
Gear
Gear
0,,(0
Potentiometer-.
Inertia Viscous
damping
Gear
(0
FIGURE 1.9 Antenna azimuth
position control system:
a. system concept; b. detailed
layout; c. schematic;
{figure continues)
14
Chapter 1 Introduction
FIGURE 1.9 (Continued)
d. functional block diagram
Input
transducer
Angular
input
Potentiometer
Voltage Error
proportional Summing or
t 0
junction Actuating
input + ,<~x signal
Voltage
proportional
to
output
Sensor
(output transducer)
Potentiometer
Similarly, the output angular displacement is converted to a voltage by the potentiome-
ter in the feedback path. The signal and power amplifiers boost the difference between
the input and output voltages. This amplified actuating signal drives the plant.
The system normally operates to drive the error to zero. When the input and output
match, the error will be zero, and the motor will not turn. Thus, the motor is driven only
when the output and the input do not match. The greater the difference between the input
and the output, the larger the motor input voltage, and the faster the motor will turn.
If we increase the gain of the signal amplifier, will there be an increase in the
steady-state value of the output? If the gain is increased, then for a given actuating
signal, the motor will be driven harder. However, the motor will still stop when the
actuating signal reaches zero, that is, when the output matches the input. The
difference in the response, however, will be in the transients. Since the motor is
driven harder, it turns faster toward its final position. Also, because of the increased
speed, increased momentum could cause the motor to overshoot the final value and
be forced by the system to return to the commanded position. Thus, the possibility
exists for a transient response that consists of damped oscillations (that is, a sinusoidal
response whose amplitude diminishes with time) about the steady-state value if the
gain is high. The responses for low gain and high gain are shown in Figure 1.10.
FIGURE 1.10 Response of a
position control system,
showing effect of high and low
controller gain on t he out put
response
, Output with
high gain
Time
1.5 The Design Process
We have discussed the transient response of the position control system. Let us
now direct our attention to the steady-state position to see how closely the output
matches the input after the transients disappear.
We define steady-state error as the difference between the input and the output
after the transients have effectively disappeared. The definition holds equally well
for step, ramp, and other types of inputs. Typically, the steady-state error decreases
with an increase in gain and increases with a decrease in gain. Figure 1.10 shows
zero error in the steady-state response; that is, after the transients have disap-
peared, the output position equals the commanded input position. In some systems,
the steady-state error will not be zero; for these systems, a simple gain adjustment
to regulate the transient response is either not effective or leads to a trade-off
between the desired transient response and the desired steady-state accuracy.
To solve this problem, a controller with a dynamic response, such as an electrical
filter, is used along with an amplifier. With this type of controller, it is possible to
design both the required transient response and the required steady-state accuracy
without the trade-off required by a simple setting of gain. However, the controller
is now more complex. The filter in this case is called a compensator. Many systems
also use dynamic elements in the feedback path along with the output transducer to
improve system performance.
In summary, then, our design objectives and the system's performance revolve
around the transient response, the steady-state error, and stability. Gain adjust-
ments can affect performance and sometimes lead to trade-offs between the
performance criteria. Compensators can often be designed to achieve performance
specifications without the need for trade-offs. Now that we have stated our
objectives and some of the methods available to meet those objectives, we describe
the orderly progression that leads us to the final system design.
l . 5 The Design Process
In this section, we establish an orderly sequence for the design of feedback control
systems that will be followed as we progress through the rest of the book. Figure 1.11
shows the described process as well as the chapters in which the steps are discussed.
The antenna azimuth position control system discussed in the last section is
representative of control systems that must be analyzed and designed. Inherent in
Step Step 2 Step 3 Step 4 Step 5
Determine
a physical
system and
specifications
from the
requirements.
Draw a
functional
block
diagram.
Transform
the physical
system into
a schematic.
Analog: Chapter 1
Digital:
FIGURE 1.11 The cont rol system design process
Use the
schematic
to obtain a
block diagram,
signal-flow
diagram,
or state-space
representation.
Chapters 2, 3
Chapter 13
If multiple
blocks, reduce
the block
diagram to a
single block or
closed-loop
system.
Chapter 5
Chapter 13
Step 6
Analyze,
design, and test
to see that
requirements
and
specifications
are met.
Chapters 4, 6-12
Chapter 13
94 Chapter 2 Modeling in the Frequency Domain
Case Studies
Antenna Control: Transfer Functions
This chapter showed that physical systems can be modeled mathematically with
transfer functions. Typically, systems are composed of subsystems of different
types, such as electrical, mechanical, and electromechanical.
The first case study uses our ongoing example of the antenna azimuth position
control system to show how to represent each subsystem as a transfer function.
PROBLEM: Find the transfer function for each subsystem of the antenna
azimuth position control system schematic shown on the front endpapers. Use
Configuration 1.
SOLUTION: First, we identify the individual subsystems for which we must find
transfer functions; they are summarized in Table 2.6. We proceed to find the
transfer function for each subsystem.
TABLE 2.6 Subsystems of the antenna azimuth position control system
Subsystem
Input potentiometer
Preamp
Power amp
Motor
Output potentiometer
Input
Angular rotation from user, #,(*)
Voltage from potentiometers,
v
e
(t) = v,{t) - v
0
{t)
Voltage from preamp, v
p
{t)
Voltage from power amp, e
a
(t)
Angular rotation from load, 0Q(()
Output
Voltage to preamp, Vj(t)
Voltage to power amp, v
p
(t)
Voltage to motor, e
(l
(t)
Angular rotation to load,
0o(t)
Voltage to preamp, VQ(0
Input Potentiometer; Output Potentiometer
Since the input and output potentiometers are configured in the same way, their
transfer functions will be the same. We neglect the dynamics for the potentiometers
and simply find the relationship between the output voltage and the input angular
displacement. In the center position the output voltage is zero. Five turns toward
either the positive 10 volts or the negative 10 volts yields a voltage change of 10
volts. Thus, the transfer function, V,-(s)/0,;(s), for the potentiometers is found by
dividing the voltage change by the angular displacement:
Vt(s)
m
10
lOJr
(2.202)
Preamplifier; Power Amplifier
The transfer functions of the amplifiers are given in the problem statement. Two
phenomena are neglected. First, we assume that saturation is never reached.
Second, the dynamics of the preamplifier are neglected, since its speed of response
is typically much greater than that of the power amplifier. The transfer functions of
both amplifiers are given in the problem statement and are the ratio of the Laplace
transforms of the output voltage divided by the input voltage. Hence, for the
Case Studies
95
preamplifier,
and for the power amplifier,
E
a
(s)
V
P
(s)
I)
100
5 + 100
(2.203)
(2.204)
Motor and Load
The motor and its load are next. The transfer function relating the armature displace-
ment to the armature voltage is given in Eq. (2.153). The equivalent inertia, /,, is
J - ^ +' r f J g ) = 0.02 + 1 ^ = 0.03 (2.205)
where JL = lis the load inertia at 9$. The equivalent viscous damping, D
m
, at the
armature is
D

B f l
"
+ 1 > i
( p ) = 0.01 + 1 ^ = 0.02 (2.206)
where D
L
is the load viscous damping at 9Q. From the problem statement, K
t
= 0.5
N-m/A, Kb = 0.5 V-s/rad, and the armature resistance R
a
= 8 ohms. These quantit-
ies along with J
m
and D
m
are substituted into Eq. (2.153), yielding the transfer
function of the motor from the armature voltage to the armature displacement, or
9
m
{s) _ K
t
/{R
a
J
m
) 2.083
E
a
(s)
_L * (n
K
*
Kb
s + [D
m
+
5(5 + 1.71)
T l J?
To complete the transfer function of the motor, we multiply by the gear ratio to
arrive at the transfer function relating load displacement to armature voltage:
Oo(s) 6
m
(s) 0-2083
EW) =
0A
EM-4^vn)
(2
-
208)
The results are summarized in the block diagram and table of block diagram
parameters (Configuration 1) shown on the front endpapers.
CHALLENGE: We now give you a problem to test your knowledge of this chapter's
objectives; Referring to the antenna azimuth position control system schematic
shown on the front endpapers, evaluate the transfer function of each subsystem.
Use Configuration 2. Record your results in the table of block diagram parameters
shown on the front endpapers for use in subsequent chapters' case study challenges.
Transfer Function of a Human Leg
In this case study we find the transfer function of a biological system. The system is
a human leg, which pivots from the hip joint. In this problem, the component of
weight is nonlinear, so the system requires linearization before the evaluation of
the transfer function.
96
Chapter 2 Modeling in the Frequency Domain
Hip joint
FIGURE 2. 51 Cyl i nder model of a
human leg.
M
D
Tt
Tntt)
PROBLEM: The transfer function of a human leg rel at es the out put angul ar
rot at i on about t he hip joint t o the i nput t or que supplied by t he leg muscle. A
simplified model for t he leg is shown in Figure 2. 51. The model assumes an
applied muscul ar t orque, T
m
(t), viscous dampi ng, D, at t he hi p joint, and
inertia, J, around t he hip joint.
15
Also, a component of t he weight of the leg,
Mg, wher e M is the mass of t he leg and g is t he acceleration due t o gravity,
creat es a nonl i near t orque. If we assume t hat the leg is of uniform density,
the weight can be applied at L/2, wher e L is t he l engt h of the leg (Milsum,
1966). Do t he following:
a. Eval uat e the nonl i near torque.
b. Find the transfer function, 9(s)/T
m
(s), for small angles of rot at i on,
wher e 9{s) is t he angular rot at i on of t he leg about t he hip joint.
SOLUTION: First, calculate t he t orque due t o the weight. The total weight of
t he leg is Mg acting vertically. The component of t he weight in the direction
of rot at i on is Mg sin 9. This force is applied at a distance L/ 2 from the hip
joint. Hence t he t orque in t he direction of rot at i on, Tw(t), is Mg(L/2) sin 9.
Next , draw a free-body diagram of t he leg, showing t he applied t orque,
T
m
{t), t he t orque due t o the weight, T
w
(t), and the opposing t orques due t o
inertia and viscous dampi ng (see Figure 2.52).
Summi ng torques, we get
T
w
(t)
J
&9_
dt
2
d9 I
D-^ + Mg-sm9 = T
m
(t)
(2.209)
FIGURE 2.52 Free-body diagram of We linearize the system about t he equilibrium point, 9 = 0, the vertical
leg model position of t he leg. Usi ng Eq. (2.182), we get
s i n # - s i n 0 = (cos0)<5# (2.210)
from which, sin 9 = 89. Al so, J d
2
9/dt
2
= J d
2
89/dt
2
and D d9/dt = D d89/dt.
Hence Eq. (2.209) becomes
r
d
2
89
n
d89
mr
L _, , ,
(2.211)
Not i ce t hat t he t or que due to the weight approxi mat es a spring t orque on t he leg.
Taking t he Laplace t ransform with zero initial conditions yields
Js
2
+ Ds + Mg^\89{s) = T
m
(s) (2.212)
from which the transfer function is
89(s) 1/7
T
m
(s) 9
2
+%S +
MgL
2/
(2.213)
15
For emphasis, J is not around the center of mass, as we previously assumed for inertia in mechanical
rotation.
Review Questions 97
for small excursions about the equilibrium point, 9 = 0.
CHALLENGE: We now introduce a case study challenge to test your
knowledge of this chapter's objectives. Although the physical
system is different from a human leg, the problem demonstrates
the same principles: linearization followed by transfer function
evaluation.
Given the nonlinear electrical network shown in Figure 2.53, find
the transfer function relating the output nonlinear resistor voltage,
V
r
(s), to the input source voltage, V(s).
v
r
O = 2/:(/)
FIGURE2.53 Nonlinear electric circuit
^ Summary ^
In this chapter, we discussed how to find a mathematical model, called a transfer
function, for linear, time-invariant electrical, mechanical, and electromechanical
systems. The transfer function is defined as G(s) = C(s)/R(s), or the ratio of the
Laplace transform of the output to the Laplace transform of the input. This relation-
ship is algebraic and also adapts itself to modeling interconnected subsystems.
We realize that the physical world consists of more systems than we illustrated
in this chapter. For example, we could apply transfer function modeling to hydraulic,
pneumatic, heat, and even economic systems. Of course, we must assume these
systems to be linear, or make linear approximations, in order to use this modeling
technique.
Now that we have our transfer function, we can evaluate its response to a
specified input. System response will be covered in Chapter 4. For those pursuing the
state-space approach, we continue our discussion of modeling in Chapter 3, where
we use the time domain rather than the frequency domain.
( Review Questions ^
1. What mathematical model permits easy interconnection of physical systems?
2. To what classification of systems can the transfer function be best applied?
3. What transformation turns the solution of differential equations into algebraic
manipulations?
4. Define the transfer function.
5. What assumption is made concerning initial conditions when dealing with
transfer functions?
6. What do we call the mechanical equations written in order to evaluate the
transfer function?
7. If we understand the form the mechanical equations take, what step do we avoid
in evaluating the transfer function?
8. Why do transfer functions for mechanical networks look identical to transfer
functions for electrical networks?