Computer Controlled Systems

Istanbul Technical University, FEE

Fall, February-May, 2024-2025

Yaprak YALÇIN
[email protected]
Room:D4202

1
Evaluation :

• Homework-projects (3) : % 25
• Final test: %5
• Midterm Exam (1): %25 (15 nisan 2025)
• Final Exam :%45 (Enterence Prerequisite: Weighted average
of midterm exam and homework-projects should be > =20)
Books :

[1] K. Ogata 1995, Discrete-Time Control Systems, PH

[2] B. C. KUO, 1992, Digital Control Systems, Saunders
[3] B. Wittenmark, K.J.Astrom, 2002, Computer Controlled Systems,
Theory and Design, P.H.

2
OUTLINE

Introduction to computer controlled systems.

Types of Signals, Sampling and Reconstruction operations
Z-transformations
Signal holders
Discrete-time models of continuous systems; Impulse transfer
functions, Discrete transfer functions
Stability and performance analysis of discrete time systems.
Main direct digital controller design methods
• Analytical method
• Designs based on Root-Locus method
• Design in frequency domain criterias
Sensitivity Analysis

3
Model matching based design methods
Emulation Design – (Designing discrete-time controllers
by using continuous time controllers)
State Space Representation, Discrete-time State-Space
Models, Stability, Lyapunov Theory
Controlability, Observability, Pole-Placement

4
INTRODUCTION
Recall: What is a Control System ?

Open loop control

Desired System input, Plane, automobile, chemical System response
behavior, Manupulated processes, social systems, (behavior)
Setpoint, value, robots, motor, etc. Process value
REFERANS Control signal Output Signal

Controller System(Plant, Process)

r
u y

Closed loop control

System input,
Desired
Manupulated Plane, automobile, chemical System response
behavior, processes, social systems, (behavior)
(r-y) value, Process value
Setpoint, robots, motor, etc.
ERROR Control signal Output Signal
REFERANS

Controller System(Plant, Process)

r e
y
y u

5
INTRODUCTION

Double tank system

6
Example control systems

7
Lane Warning or Changing

8
9
Example: Speed control in automobiles

ACTAUTOR
Gas pedal MEASUREMENT
(Throttle) Vehicle Speed

CALCULATIONS
Control law

Input
Control
Reference Signal Output
Controller

10
 Input
Control
Reference Signal Output
Controller

Speed
?

Driver? Speed
Auto?

11
Control problems

▪ Stabilization ▪ Disturbance attenuation

▪ Changing the behaviour of system ▪ Minimization of cost functions
as desired;
✓ Transient response
✓ Steady state response
Why Feedback?

uncertainties
disturbances noises

Non-linear terms

Reasons for using feedback

➢ The need to decrease effects of ➢ The need to decrease affects of
disturbances and noises modelling errors and uncertainties

12
 Analysis and Design of Control Systems
❖ Step 1: Modelling
• Using phsical law
• Using identification methods
❖ Step 2: Analysis– Open Loop
• Stability, controllability, observability, information on model
uncertainties
❖ Step 3: Construction of control rule
• Selection of control structure: 1-DOF, 2-DOF
• Selection of design method:Clasical, modern, post-modern
❖ Step 4: Analysis– Closed Loop
❖ Step 5: Simulation
• Matlab, simulink, Mathematica etc….
❖ Step 6: Implementation

13
 First Examples of Control Systems

❖ One of the earliest control systems known is the

water clock invented by Ktesibios (300 BC)-
Liquid level control.

❖ In 1681, Denis Papin introduced the steam

pressure control systems, where he invented the
safety valve.

❖ In 1745, speed of windmills are controlled.

❖ In 1769, James Watt fly-ball governer: Rotational

velocity stabilization of the steam engine

14
 Short History of Control

▪ 1840- The first article in the control “The Theory of Escapements” G.B.AIRY
▪ 1868 The second article in the control ‘‘on governor’s ’’ – Maxwell
▪ 1877 Routh stability criteria
▪ 1892 Liyapunov stability theory
▪ 1895 Hurwitz stability criteria
▪ 1932 Nyquist- Frequency response methods
▪ 1945 Bode -Frequency response methods Phase &Gain margins
(robust!)
▪ 1947 Hurewicz introduced the concepts of pulse transfer functions and Z-
transforms for the analysis of sampled-data systems.
▪ 1947 Nichols - Frequency response methods
▪ 1948 Evans– Design with Zero and Poles and Root Locus
▪ 1949 Wiener optimal control studies- Minimal variance problem

15
 Short History of Control

▪ 1955 Kalman filter and controllability and observability

analysis-LQG
▪ 1956 Artificial Intelligence
▪ 1957 Bellman, optimal and adaptive control
▪ 1962 Pontryagin, optimal control
▪ 1972 Vidyasagar, multi-variable optimal control ve Robust
control
▪ 1978 Zames, Proposed the H∞ norm as a design criterion
▪ 1981 Doyle, “Robust control theory”

16
A brief history of control sicence and engineering

coefficients of differential equations

17
18
• Development of post modern control design
After 1980: Combination of clasical and modern
approaches
H-infinity Control
Robust Control

19
Clasical Control Modern Control

✓ Developed between 1920-50 ✓ Developed between 1950-80

✓ Frequency domain analysis and ✓ Time domain analysis and
design (transfer function based) design (based on differential
✓ SISO systems based equation)
✓ Interested with input and output ✓ MIMO systems based
variables ✓ Interested with Input, output
✓ Robust thanks to phase and gain and state variables
margins ✓ Being robust is controversial
✓ Not being concerned with ✓ Controllability and
controllability and observability observability are important
✓ Not being concerned about being ✓ Being optimal is your primary
optimal concern
✓ Common and known in the ✓ Interest of industry is very slow
industry

20
 World Wars and Classical Control
➢SHIP CONTROL,
At 1910, E.A. Sperry
“gyroscope”, invented for stability and control of ships,
later on it is used on aeroplanes as well.
1922’de N. Minorsky
uses the three-term control rule,
proportional-integral-derivative (PID) controller.

➢ "Theory of Servomechanisms" H.L. Házen [1934]

 Milestones

In1920s and 1930s

Analysis of dynamical systems are done in time domain with

differential equations.

P.-S. de Laplace (1749-1827)

J. Fourier (1768-1830),

Solution of dif. Equations in Frequency Domain

 Classical Era of Control,
Frequency response analysis

H. Nyquist [1932],
at Bell Telephone Laboratories

Regeneration Theory,
Nyquist stability criterion
H.W. Bode [1938],
Closed loop stability,
Gain and phase margin concepts.
MODERN ERA of CONTROL
(1960-1980)

R. Kalman, Bertram, 1960,

Stability analysis of nonlinear systems
R. Kalman, Bertram, 1960a,1960b,
LQR, Optimal Control,
Discrete-time Optimal Filtering
R. Kalman, Bucy, 1961,
Continuous Kalman Filter
 APOLLO 11, 1969

The cruise route provided with Kalman

Filter; provided the first landing to the
moon.

➢ 1965, mini computer era with Digital

Eqp. Co. PDP-8
➢ 1969, W. Hoff, mikroprocessor

✓ Computer usage era is started, 1980’ler

 Computer usage in control

➢ Basic aims and functionalities of control and instrumentation

systems is almost the same since the beginning
➢ As a result of technological innovations and increasing demands
a paradigm change happened in measurement and control
methodology
➢ Fast and accurate digital technology and analog to digital
converters,
➢ Revolutionary advances in microprocessors and communication
technology
➢ Manual monitoring and control replaced by extremely advanced
automatic process monitoring and control systems

26
 Computer usage in control

➢ The need for monitoring

➢ The desire to adapt to changing conditions (more flexible
design)
➢ Desire to convey and response fast to the changing production
requirements
➢ ...

27
 Computer usage in control

✓ Computers provide continuity in product quality that necessary in

production facilities

✓ To enable the production of a different product and to be able to

adapt the frequent changes in the product specification, it allows
flexibility in changing the control procedures

✓ Keeps a database containing the product recipe, thus it makes esier

to pass a new recipe.

✓ Computer use, increase the availibitiy of the factory therefore

significiantly increases the productivity of the factory.

✓ Factories first used the computers to process data.

28
 Computer usage in control

✓ Digital electronic systems in 1950s planned and scheduled the

production

✓ Big leap; transition from planning to monitoring and controling the

processes in the production stages

✓ As a result, computer control changed the way of production, resulting

in unprecedented productivity and efficiency

✓ At first, digital computers are used to supervise the analog controllers

✓ The next step was to remove the analog intermediary

✓ In 1962, in Monsanto's ethylene distillation plant in Louisiana,

England, individual thermostats and pressure controllers are removed
and the computer is allowed to perform these control tasks directly.

29
First Industrial Computer Controlled Systems

History; 1959
Ramo-Wooldridge; Refinery;
Closed loop

The First Industrial Computer Control

System

in 1959 A close cooperation between

Texaco and Ramo Wooldridge, using an
RW-300 computer “Closed-loop control”
of industrial polymerization process is
provided.

The computer control system increased

productivity ve decreased the cost of
renewment of catalyzer

30
 Computer usage in control

✓ Most of the computer that are use in process control was not
designed and produced by computer companies.

✓ The engineering firms that design manufacturing equipments

developed also the systems that is needed to control those
equipments.

✓ One important exception was a military company Ramo Wooldridge.

✓ As mini computers and micro-computers are presented to use,

engineers integrate these computers to process control systems.

✓ Computer control systems is an example of special use of a general

purpose tehcnology.

31
Computer usage in control
Date. 1971-01
Keywords: Fortran IV (Software); FOX 1
System (Computer); Business applications;
Software; Digital computer: minicomputer;
Scientific applications; promotional
materials; Foxboro; Wallace, P.R.; Fixed application
digital computer: industrial

✓ Foxboro company, established in the

early 20th century, is specializes in industrial
monitoring and control devices.
✓ FOX 1 computer system is relied on to its
own design.
✓ Later on, onshelf mini computers also
used in Foxboro systems.

32
Computer usage in control

33
34
35
36
37
 Converting to Digital: ADDAVERTER Sistem

A system for interconnection of large-scale analog and digital computers

Tarih: 1956
© Epsco Incorporated
Keywords
Analog; ENIAC (1945), UNIVAC(1951)
(Universal Automatic Computer);
Gordon,
Bernie

✓ Former UNIVAC engineer Bernie Gordon

founded Epsco in 1953

✓ A year later, Epsco introduced Datrac,

https://www.computerhistory.org/revol one of the oldest commercial analog /
ution/story/130 digital converters
✓ This equipment was also used to
create hybrid analog-digital systems in
the mid-1950s
38
 Real Time Computing

✓ Real-time computing responds to events as they happen, something

even early computers were able to do.

✓ Demand for real-time computing began with the military, but swiftly
expanded to industrial, medical and soon, everyday uses.

✓ Fear of nuclear-armed Soviet bombers terrified 1950s America. SAGE, a

massive real-time control and communications system developed for
the Air Force by MIT’s Lincoln Laboratory, offered a solution.

✓ SAGE (Semi-Automatic Ground Environment) linked 23 sites across the

U.S. and Canada, coordinating weapons systems and processing radar,
weather reports and other data. By the time it became fully
operational in 1963, however, the principal threat had shifted from
aircraft to missiles, making SAGE’s value questionable. Nevertheless,
some sites remained in service until 1982.

39
SAGE weapons director console with light gun

✓ This user console showed all activity in

the air space assigned to it.

✓ Operators could request information

about objects that appeared and use
the light gun to assign identification
numbers to displayed aircraft.

✓ This user console showed all activity in

the air space assigned to it.

✓ Operators could request information

about objects that appeared and use
the light gun to assign identification
numbers to displayed aircraft.

40
How Big Was SAGE?

SAGE model
✓ IBM built this model to help
visualize the layout of a SAGE
installation’s four floors: Telephone
and Test Equipment, Computer
Floor, Command Post and Display
Floor

Istanbul Technical University, 11 January

2012 41
Real Time Computing

✓ When you step away from your computer you’re probably

still using a computer.

✓ Faster, more powerful processors—and the shrinking cost and

size of microelectronics—make it ever easier to integrate
computers into our daily lives and household products.

✓ Computerized monitoring and control systems bring new

efficiencies and capabilities.

✓ They also bring new risks as we grow more dependent on

them

42
A tipical computer controlled system

43
Industrial Control
Systems

PLC
(Digital
Controller)

44
What is a computer-controlled system?

Computer-controlled system (Discrete-time Control Systems, Digital Control systems):

Continuous time systems controlled by digital computers via interfaces.

Computer: PLC, Special-purpose

computer (Industrial PC),
microcontrollers, DSP, PC.

Interface: A/D (Sampling) and

D/A converters

Computer + interfaces

r(t) + u(t) y(t)

e(t)
ADC Micro-processor DAC Plant
(Real-time processing)
-

45
Örnek

46
Continuous-time / Discrete-time control systems

47
 Signals in Computer Controlled Systems

Continuous analog
Continuous digital
signal
signal

Discrete
Discrete digital
analog signal
signal

48
Sampling

49
Holding

50
Signal processing in ADC and DAC

51
Signals in Computer Controlled Systems

52
Discrete-time control systems general scheme

53
Design Approaches

54