WELCOME
TO
CONTROL SYSTEMS
ENGINEERING
DISCLAIMER
We should not expect to solve some real world control engineering
problems in one/two hours in a class room!
We shall mostly discuss analytical techniques to solve variety of
problems which are significantly simplified or even hypothetical!
(to maintain analytical tractability and the multidisciplinary nature)
Books:
1. Nagrath and Gopal (3rd/5th Ed.)
2. K. Ogata
3. B.C. Kuo
4. Norman Nise
5. Dorf and Bishop
Chamber Consultation: Thursday 5:00 p.m. to 6:00 p.m.
Grading Scheme:
Mid-sem: 90
Compre: 120
Quiz: 18*4=72 (best 4 out of 5)
MATLAB Assignment: 18
(Tut. Classes to commence from 28th Jan.)
What is Control Systems?
Why Control Systems?
How to control a system?
The major application areas
Relevance to Electrical & Electronics Engineering
A brief historical account
What is ‘Control Systems Engg.’ ?
Control:
- To have power/authority over something/somebody to
ensure that they behave in a particular way
- to limit/regulate somebody/something
System:
- A group of things/parts working together as a whole
towards a specific goal
(the goal cannot be met by them individually)
Engineering: Science + Art + Economics
Roughly speaking, the objective is
' to adjust a system so that it functions as desired '
Why ‘Control Systems’ ?
Because often systems don’t naturally behave the way we
want them to behave !
How to control a system ?
CONTROL
Manual Automatic
- less accurate - arrangement of components doing the job
- less precise (with min./zero human intervention)
- slow
- infeasible/ inhuman
Passive (crude, limited, simple)
Active (wider, finer, efficient)
Open loop
Closed loop or Feedback
Manual control Vs. Automatic control
A simple example:
Water level control Water level control
(manual) (automatic)
Infeasibility of manual control
Passive Control:
Example: spacecraft with gravity gradient stabilization
Inverted pendulum:-
Common in Mechanical systems
Electrical Engineering
Electronics Power Communication Control
Manipulation of Transmission, Long distance Regulating the
electric currents distribution & transfer of low behaviour of a system
through conversion of bulk power electrical • Control
semiconductor electrical energy signal through or • Instrumentation
devices • Power Systems without wire
• Electric Machines
Electronics and Instrumentation
Electrical Engg.
Communication Engg. Engg.
How to Control a System Elaborated
The Principal Goal of Control Systems is
to look for a technically and economically viable way to make
a system behave as closely as possible to the desired behavior
even in the presence of ‘modest’ plant uncertainties and
external disturbances
• the system behavior is changed by feeding an ‘appropriate’ signal
Signal
A function representing a physical quantity or variable
usually varying with time.
Voltage across a capacitor in an RC circuit is a signal
System
A combination of components that act together and perform
a certain objective. In other words, a system is a device or
a collection of devices that process signals.
An RC circuit is a system
Different Types of Signals:
Signal
Continuous time Discrete time
defined at each defined at distinct
instant of time instants of time
e.g. capacitor voltage e.g. tiger population in
in an RC circuit a forest
Signal
Deterministic Stochastic or Random
value of a signal at any instant signals taking random values
is completely specified
• e.g. wind gust acting on the
• e.g. thrust generated by an aircraft
aircraft (statistical treatment)
(mathematical treatment)
Different Types of Systems:
System
Continuous-time Discrete-time
input & output Input & output signals
signals are continuous-time are discrete-time
• e.g. an RC circuit •e.g. tiger population as
a function of cattle
• Differential Eqn. population in a forest
•Difference Eqn.
System
Causal Non-causal
present value of output depends present value of the output
on present and/or past values of depends on future values of
input input
•all naturally occurring systems •share market dynamics
System
Static Dynamic
Present output depends only Present output depends on
on present input present and past/future input
•algebraic equations •differential equations
•no energy storing element •energy storing element present
•e.g. a resistive circuit •e.g. an RC circuit
System
Linear Nonlinear
Obeys superposition theorem Does not obey superposition theorem
•linear algebraic/differential •nonlinear algebraic/differential
equation equation
•spring-mass system with •spring-mass system with static
viscous friction friction
•linear inductor •nonlinear inductor
System
Time-invariant Time-varying
System parameters don’t change System parameters change
over the duration of interest over the duration of interest
•time-invariant equations •time-varying equations
•an aircraft flying for a short •an aircraft flying for
duration a long duration
System
Lumped parameter Distributed parameter
Physical size of the system is small Physical size is comparable to the
compared to the wavelength wavelength corresponding to the
corresponding to its normal normal frequency of operation
frequency of operation
•ordinary differential equations •partial differential equations
•e.g. an RLC circuit •e.g. electric power transmission
system
System
SISO MIMO
• Single tank water level • Coupled tank water level
control system control system
System
Stable Unstable
Given a small perturbation, the Given a small perturbation, the
system comes back to its original system does not come back to its
state original state
How to Control a System Elaborated Contd. …
Open loop Control
Control action u(t) is independent of the actual value of
the output (i.e. the variable being controlled)
Uncontrolled system System with open loop control
Closed loop Control (Feedback Control)
A system, which measures the output and compares it with the
reference input (i.e. the desired profile of the output) and uses the
difference to generate the appropriate control action so as to remove/
reduce the difference
Open loop control Vs. Closed loop control
Advantages
1. Reduces the effects of external disturbances
2. Reduces the effects of plant uncertainties
3. Can stabilize an unstable system
Disadvantages
1. Can destabilize an otherwise stable system
2. Enhances cost and complexities
3. Invites sensor noise
Two Simple Illustrative Examples:
1. Room temperature control
2. Formal Education System
Common Terminology
1. Plant or Process
Any physical system (or abstract operation) that is to be controlled
e.g. a chemical reactor, an electric motor, an aircraft etc.
2. Sensor
A device that measures a physical quantity like pressure, temperature,
acceleration etc. Usually, sensors produce an electrical signal whose
voltage is proportional to the physical quantity being measured.
3. Controller
Devices that implement a control law.
Control law is a strategy that processes the error signal and generates
the appropriate control signal.
Implementation can be through electronic circuits or hydraulic or
pneumatic or mechanical arrangements or a computer program.
4. Actuator
A device that has the capacity to finally execute the control action
by generating enough power to be able to influence the system
being controlled.
e.g. electric motors
5. Comparator or Error Detector
A device that calculates the algebraic difference between the
reference input and the measured signal
6. Reference Input
An electrical signal (or a number) that represents the desired profile
of the variable being controlled
7. Output (Plant Output)
The variable being controlled (Temperature, pressure, position,
velocity …)
8. Disturbance Input
External inputs that adversely affect the behavior of the plant being
controlled
Disturbances are often induced by the environment, and often
random in nature and immeasurable
9. Model or Plant Uncertainty
Refers to changes in the plant behavior
May be caused by approximations in its mathematical description or
parameter variations due to environmental/operational effects
10. Sensor Noise
Usually a random signal representing the errors and inaccuracies
occurring in the measurement process
General Features of Different Components
Sensor (The Eye)
Features
- High accuracy (low error)
- High precision (repeatability)
- Very fast dynamics
- High sensitivity (ratio of incremental o/p and
incremental i/p)
- Linearity (liner IO characteristic)
- High signal to noise ratio (SNR)
Controller (The Brain)
Different Types (implementation method wise)
- Analog Controller (usually an electronic ckt., sometimes
a hydraulic or pneumatic or mechanical arrangement)
- Digital Controller (a computer program)
Different Types (algorithm wise)
- PID controller, Lead-Lag controller
- On-off controller
- State feedback controller
- Several other more advanced algorithms
(H-infinity controller, fuzzy controller…)
Actuator (The Muscle)
Different Types
1. Hydraulic Actuators
high power/weight ratio, lubricating and cooling,
nonlinear, messy, fire hazards, costlier
2. Pneumatic Actuators
low power/weight ratio, compressible, less messy (no
return pipes)
3. Electrical Actuators
Easier to control and interface, faster, less cumbersome,
less nonlinear
Major Engineering Applications
1. Electrical Engineering
a) Power System
- Equipments are designed to handle certain voltage & frequency
- Terminal voltage depends on reactive power demand
- Frequency depends on active power demand
- Power demand is changing continuously and arbitrarily
Block diagram of LFC Loop Block diagram of the AVR Loop
1.4
1.2
0.8
Vt
0.6
0.4
0.2
0
0 1 2 3 4 5
time (sec)
1500
b) Electric Motor Drives
1000
speed (rpm)
Large number of modern applications
require variable speed operation e.g. 500
- Steel mills 0
0 1 2
t (sec)
3 4 5
- Textile mills
- Paper mills 1600
1400
- Pumps, fans, compressors 1200
1000
- Electric propulsion system
speed (rpm)
800
600
- Plant automation 400
200
-200
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Advantages of drives with t (sec)
closed loop control :
- Reduced running costs
- Energy efficiency
- Higher reliability
- Improved quality
2. Aerospace Engineering
a) Aircraft Flight Control
p 0.5v 2 const.
- Attitude control to maintain roll, pitch and yaw
- Executing maneuvers
- Automatic landing system
Actuators: Aileron, Elevator, Rudder, Thrust vector
Sensors: Gyroscope, Accelerometer, Altimeter
b) Satellite Attitude/Orbit Control
Control Tasks:
- Orbit maneuvering
- Station keeping
- Attitude control
Actuators:
- Gas-jet Thrusters
- Momentum Wheel
- Magnetic Torquer etc.
Sensors:
- Gyroscope
- Sun sensor
- Earth sensor
c) Missile Guidance
Missile
Tactical Strategic
Homing Non-homing
(continuously tracks target) (follows a preset path) Ballistic Missile
Actuators:
- Fins (control surfaces)
- Divert Thrusters (gas jet)
Sensors:
- Ground Radar (long range sensing of target)
- Seeker (short range target sensing, mounted on missile)
- GPS (for sensing the missile’s position)
- Gyroscope & Accelerometer (missile’s attitude)
Block Diagram of a Typical Homing Missile Control System
3. Chemical Engineering
Process Control: Control of process variables such as Temperature,
Pressure, Level, Flow etc.
Servomechanism: Control of Position, Velocity and Acceleration
Distinctive Features of Process Control:
- Systems are slower
- Systems have significant time delays
- Disturbances are sometimes measurable
- Distillation column: Separating a mixture of liquids or vapors into its
constituents with desired level of purity by application and
removal of heat
- pH level control
4. Mechanical Engineering
a) Industrial Automation
A set of technologies that result in operation of machines
and systems without significant human intervention and
achieve performance superior to manual operation
A Typical Example: Rolling Mill
Other Examples:
- Cement kiln
- CNC machines
- Paper mills
b) Robotics
• Application in 4D Environments
– Dangerous
– Dirty
– Dull
Robotic Manipulators
– Difficult
Sojourner Rover
Mobile Robots
Sensors: position, velocity, force, torque, camera etc.
Actuators: DC Motor, Stepper Motor
Applications: materials handling, assembly task, arc welding etc.
Single-link manipulator
(inverted pendulum) 5 DOF manipulator
5. Biomedical Engineering
Implantable drug delivery system: Blood Glucose Regulation
Oral & Injectable Medication
• Is delivered to the body system-wide.
• Affects both dysfunctional and healthy organs and cells.
Site Specific Drug Delivery
• Allows for higher drug concentrations at the site of clinical disorder.
• Medication directly into the blood stream is more effective.
Implantable Drug Delivery Systems
• Continuous and controlled drug delivery is important for chronic medical
conditions requiring long-term treatment.
• Optimum rate of infusion at proper site using implantable devices
exposes the patient to only 1% of the equivalent oral dosage to
experience the same level of relief.
Causes of Diabetes:
- Inadequate production of insulin (Type-1)
- Inadequate (resistive) insulin action (Type-2)
Treatment:
• Diet Control therapy
• Oral drugs for enhancing insulin production
• Direct Insulin administration into blood stream
Implantable Sensors:
- Bioluminescent Chips , consisting of a coating of bioluminescent bacteria on top of
a light-sensitive integrated circuit
- A microchip incorporating an optical sensor
Disturbance inputs:
meal and exercise
Controller:
Closed-loop Micro-Electro-Mechanical-System (MEMS) and wafer-bonding to
create a micro-miniature pump (size of a contact lens).
6. Consumer Electronics
- Washing Machines
- Digital Camera
- Hard disc drive
- Air Conditioner . . .
7. Non-engineering Applications
- Economics
- Management & Finance
A Brief History
Pre-classical Period (1789 - 1930)
- 1789: James Watt’s Governor
- 1868: Maxwell’s explanation
- 1877: Routh’s stability criterion
- 1892: Lyapunov’s stability analysis
- 1922: Mironosky’s PID control law
- 1927: Harold Stephen Black invents Watt’s Fly Ball Governor :
Simplified Schematic Diagram
negative feedback amplifier
Classical Period (1930 - 1960)
- 1932: Harry Nyquist’s frequency domain techniques
- 1934: Describing function method
- 1940: Hendrick Bode’s frequency domain analysis
- 1942: Zeigler and Nichols propose PID controller tuning rules
- 1948: Root locus method proposed by W.R. Evans
- Early 1950’s: CNC machine and industrial robot developed
- ~ 1955: Study of discrete-time systems by Tustin, Jury
- ~1950’s: Introductory studies of nonlinear systems (Popov)
Modern Period (1960 - 1980)
- Early 1960’s: Kalman’s state space approach
- Mid 1960’s: Optimal control theory of Kalman, Bellman
and Pontryagin
- Late 1960’s: Focus was on Aerospace applications
- 1960’s & 1970’s: Adaptive and Nonlinear control techniques
emerge
- 1970’s: Zadeh & Mamdani’s Fuzzy logic control
Post-modern Period (1980 - present)
- Early 1980’s: Emergence of Robust control theory due to
Zames, Doyle and Kharitonov (model uncertainty)
- Throughout 1980’s and 1990’s: Intelligent control matures
- Extensive use of digital computers for implementation
- Powerful and cheap simulation software come up
- Present trend: hybridization of different methods with emphasis
on nonlinear systems
Typical Control Systems Design Cycle
1. Mathematical Modeling
- describing system behavior by mathematical equations
2. Analysis of the Uncontrolled system
- diagnosis
3. Controller Design
- specs selection, algorithm selection, design and tuning
4. Validation
- on a pilot plant or on a software platform