Control System Engineering

(Fall 2008)

Instructors: Prof. Bin Jiang

Dr. Ruiyun QI
Email: [email protected]
[email protected]
Basic Info.
Course units: 3
Course hours:
51 hrs = 45 hrs lectures + 6 hrs labs
Grading:
homework - 15%
lab - 15%
attendance - 10%,
final - 60%
Basic Info.
Text book and references
Available from bookstore and library

“Automatic Control Systems”, Benjamin C. Kuo

& F. Golnaraghi

“Modern Control Engineering”, Katsuhiko Ogata

“Feedback Control of Dynamic Systems”, Gene

F. Franklin et al
Basic Info.
 Online Caltech resource (for
personal use only)
“Feedback System” (PDF files), Karl J. Astrom &
Richard M. Murry
http://www.cds.caltech.edu/~murray/amwiki/

Computer tool: MATLAB with control toolbox

a high-level language and interactive environment
enables you to perform computationally intensive tasks
faster than C, C++, and Fortran
a powerful tool for control system designers
Goals of this course
Understand the basic concepts and
disciplines of automatic control

Know how to conduct design and analysis of

linear control system with the following
techniques:
Mathematical modeling techniques
Time-domain analysis techniques
Root-locus analysis techniques
Frequency-domain analysis techniques
Goals of this course

Introduction to other topics in modern

control engineering

Use Matlab to design and analyze control

systems
NUAA-Control System Engineering

Chapter 1
Introduction to Control Systems
Outline
What is a control system?
A brief history of control
Basic components of a control system
Open-loop control vs. closed-loop control
Classification of control systems
Basic requirements of control systems
Summary
What is a control system?
Generally speaking, a control system is a
system that is used to realize a desired output
or objective.

Control systems are everywhere

 They appear in our homes, in cars, in industry, in
scientific labs, and in hospital…
 Principles of control have an impact on diverse fields as
engineering, aeronautics ,economics, biology and
medicine…
 Wide applicability of control has many advantages (e.g.,
it is a good vehicle for technology transfer)
A brief history of control
 Two of the earliest examples
 Water clock (270 BC)
 Self-leveling wine vessel (100BC)

The idea is
still used
today, i.e.
flush toilet
A brief history of control
 Fly-ball governor (James Watt,1769)

• the first modern controller

• regulated speed of steam engine
• reduced effects of variances in load
• propelled Industrial Revolution
A brief history of control
 Fly-ball governor (James Watt,1789)
A brief history of control
 Birth of mathematical control theory
G. B. Airy (1840)
 the first one to discuss instability in a feedback control
system
 the first to analyze such a system using differential
equations
J. C. Maxwell (1868)
 the first systematic study of the stability of feedback
control
E. J. Routh (1877)
 deriving stability criterion for linear systems
A. M. Lyapunov (1892)
 deriving stability criterion that can be applied to both
linear and nonlinear differential equations
 results not introduced in control literature until about 1958
A brief history of control
Birth of classical control design method
H. Nyquist (1932)
 developed a relatively simple procedure to determine stability
from a graphical plot of the loop-frequency response.
H. W. Bode (1945)
 frequency-response method
W. R. Evans (1948)
core of classical
 root-locus method
control design

With the above methods, we can design control

systems that are stable, acceptable but not optimal in
any meaningful sense.
A brief history of control
Development of modern control design
Late 1950s: designing optimal systems in some
meaningful sense
1960s: digital computers help time-domain
analysis of complex systems, modern control
theory has been developed to cope with the
increased complexity of modern plants
1960s~1980s: optimal control of both
deterministic and stochastic systems; adaptive
control and learning control
1980s~present: robust control, H-inf control…
A brief history of control
Recent applications of modern control theory
include such non-engineering systems as
biological, biomedical, economic and
socioeconomic systems…
Basic components of a control system
Basic concepts of a control system
1.Plant: a physical object to be
Plant controlled such as a mechanical device,
a heating furnace, a chemical reactor or
a spacecraft.

2.Controlled variable: the variable

Controlled controlled by Automatic Control
variable
System , generally refers to the
system output.

3.Expected value : the desired

Expected
value value of controlled variable based on
requirement, often it is used as the
reference input
Controller
4.Controller: an agent that can
calculate the required control signal.

5.Actuator: a mechanical device that

Actuator takes energy, usually created by air,
electricity, or liquid, and converts that
into some kind of motion.
6.Sensor : a device that measures a
Sensor physical quantity and converts it into a
signal which can be read by an observer
or by an instrument.

7.Disturbance: the unexpected factors

disturbing the normal functional
Disturbance
relationship between the controlling and
controlled parameter variations.
Block diagram of a control system
n Disturbance

r e u y
Controller Actuator Plant
Expected Controlled
value - Error variable

Sensor
lead-out point:
comparison component Here, the signal is
(comparison point) : transferred along
its output equals the two separate routes.
algebraic sum of all input The Block represents
signals. the function and name of its
corresponding mode, we don’t
“+”: plus; “-”: minus need to draw detailed structure,
and the line guides for the transfer route.
Open-loop control systems
 Open-loop control systems: those systems in which
the output has no effect on the control action.

System Control System

input signal output
CONTROLLER PLANT

 The output is neither measured nor fed back for

comparison with the input.
 For each reference input, there corresponds a fixed
operating conditions; the accuracy of the system depends
on calibration.
 In the presence of disturbances, an open-loop system will
not perform the desired task.
Open-loop control systems
Examples
Washing machine

Traffic signals

Note that any control systems

that operates on a time basis
are open-loop.
Open-loop control systems
Some comments on open-loop control
systems
Simple construction and ease of
maintenance.
Less expensive than a closed-loop Good
system.
No stability problem.
Recalibration is necessary from
time to time. Bad
Sensitive to disturbances, so less
accurate.
Open-loop control systems
When should we apply open-loop
control?
The relationship between the input and
output is exactly known.
There are neither internal nor external
disturbances.
Measuring the output precisely is very
hard or economically infeasible.
Closed-loop control systems
 Closed-loop control systems are often referred to as
feedback control systems.
 The idea of feedback:
Compare the actual output with the expected value.
Take actions based on the difference (error).

Expected Control System

value Error signal output
CONTROLLER PLANT

This seemingly simple idea is tremendously powerful.

Feedback is a key idea in the discipline of control.
Closed-loop control systems
In practice, feedback control system and
closed-loop control system are used
interchangeably

Closed-loop control always implies the use

of feedback control action in order to
reduce system error
Example 1 : flush toilet
Plant: water tank water q1(t)
Input: water flow piston
Output: water level h(t ) lever
Expected value: h0 float
Sensor: float h0
Controller: lever h(t)
Actuator: piston

Controller Actuator Plant

h0 q1 (t ) Water h(t )
Lever Piston threshold
Tank

q2(t)
Float
Sensor
Example 2: Cruise control
mv  bv  ueng  uhill
ueng  k (vdes  v)
Disturbance
Road grade uhill
Desired Control Actual
velocity vdes signal
Calculation Auto velocity v
Engine
element body
Reference Error ueng Controlled
input variable
Controller Actuator Plant

Sensor

Speedometer
Measured
velocity
Sensor noise
Disturbance
Example 2: Cruise control

mv  bv  uengine  uhill

uengine  k (vdes  v)
Stability/performance
vss  vdes as k  
􀁹 Steady state velocity approaches desired velocity as k → ∞;
􀁹 Smooth response: no overshoot or oscillations
Disturbance rejection
􀁹 Effect of disturbances (eg, hills) approaches zero as k → ∞
Robustness
􀁹 Results don’t depend on the specific values of b, m or k, for k
sufficiently large
Note
In this example, we ignore the dynamic
response of the car and consider only the
steady behavior.
Dynamics will play a major role in later chapters.

There are limits on how high the gain k can

be made.
when dynamics are introduced, the feedback can
make the response worse than before, or even
cause the system to be unstable.
Comments on feedback control
Main advantages of feedback:

reduce disturbance effects

make system insensitive to variations

stabilize an unstable system

create well-defined relationship between

output and reference

Comments on feedback control
Potential drawbacks of feedback:

cause instability if not used properly

couple noise from sensors into the

dynamics of a system
increase the overall complexity of a

system
Comments on feedback control
Feedback control design:
how to
get the gain as large as possible to reduce
the error
without
making the system become unstable.
Other examples of feedback

Feedback systems
are not limited to
engineering but can
be found in various
non-engineering
fields as well.
Other examples of feedback
The human body is highly advanced feedback
control system.

Body temperature and blood pressure are kept

constant by means of physiological feedback.

Feedback makes the human body relatively

insensitive to external disturbance. Thus we can
survive in a changing environment.
Open-loop vs. closed-loop
Open-loop control Closed-loop control

Simple structure, Ability to correct error

low cost

High accuracy and

Easy to regulate resistance of disturbance

Complex structure,
Low accuracy and high cost
resistance to
disturbance Selecting parameter is
critical (may cause
stability problem)

Open-loop ＋ Closed-loop ＝ Composite control system

Thinking time…
Examples of open-loop
control and closed-loop
control systems ?

For each system, could

you identify the sensor,
actuator and controller?
Classification of control systems
Classification of control systems

• the reference input • the reference input may • the input changes
(expected value) is a constant be unknown or varying according to a program
value • the controller works to • the controller works
• the controller works to make the output track according to predefined
keep the output around the the varying reference command
constant value e.g. automatic e.g. numerical control
e.g. constant-temperature navigation systems on machine
control, liquid level control boats and planes,
and constant-pressure satellite-tracking
control. antennas
Classification of control systems
f ( x1 )  y1 f ( x2 )  y 2
superposition principle
f ( x1  x2 )  f ( x1 )  f ( x2 )  y1  y2

• superposition principle applies

• described by linear
differential equation

• described by nonlinear
differential equation
Remark on nonlinear systems
Quite often, nonlinear characteristics are
intentionally introduced in a control system to
improve its performance or provide more
effective control.
For instance, to achieve minimum-time control, an
on-off (bang-bang or relay) type controller is used
in many missile or spacecraft control systems
There are no general methods for solving a
wide class of nonlinear systems
Classification of control systems

All the signals are functions

of continuous time variable t

Signals are in the form of

either a pulse train or a
digital code
e.g. digital control system
Remark on digital control systems
A digital control system refers to the use of a
digital computer or controller in the system, so
that the signals are digitally coded, such as in
binary code.
Digital computers provide many advantages in
size and flexibility.
The expensive equipment used in a system may
be shared simultaneously among several control
channels.
Digital control systems are usually less sensitive
to noise.
Classification of control systems

The parameters of a control

system are stationary with
respect to time

System contain elements that

drift or vary with time
e.g. Guided-missile control system, time-
varying mass results in time-varying
parameters of the control system
Basic requirements for control systems
Stability: refer to the ability of a system
to recover equilibrium
Quickness: refer to the duration of
transient process before the control
system to reach its equilibrium
Accuracy: refer to the size of steady-
state error when the transient process
ends
(Steady-state error=desired output –
actual output)
Note
 For a control system, the above three
performance indices (stability, quickness,
accuracy) are sometimes contradictory.

 In design of a practical control system, we

always need to make compromise.
Review questions
1. A closed-loop control system is usually more accurate than
an open-loop system. (T) (F)
2. Feedback is sometimes used to improve the sensitivity of a
control system. (T) (F)
3. If an open-loop system is unstable, then applying feedback
will always improve its stability. (T) (F)
4. Feedback can cause instability. (T)
(F)
5. Nonlinear elements are sometimes intentionally introduced
to a control system to improve its performance. (T) (F)

6. Discrete-data control systems are more susceptible to noise

due to the nature of its signals. (T) (F)
Summary
What is a
control system?
Open-loop control

A brief history
of control Closed-loop
control
Feedback:
• reduce disturbance effects
• provide extra freedom for control design
• may cause instability

Classificatio
n of control
systems
Requirements: Stability, Quickness, Accuracy
Some suggestions on course study
Pay enough attention
Control theory is very interesting, very
useful, but sometimes, very difficult.

Practice, practice and practice