EE-503 Linear System Theory

Lecture-1:

Chapter 1:

Spring 2015

Feb. 06, 2015

INTRODUCTION

1.1 INTRODUCTION
Study and Design of Physical Systems
Empirical Methods
Analytical Methods
Analytical Study of Physical Systems
Modeling
Development of Mathematical Descriptions
Analysis
Design
Distinction between Physical Systems and Models
Physical SystemModelMathematical equations using physical laws
Analysis
Quantitative:
Response of the system excited by certain
inputs.
Qualitative: General properties of systems (stability,
controllability, observability)
Design techniques often evolve from this study.
If the Response of a system is unsatisfactory, the system must be
modified;
By adjusting system parameters.
By introducing compensators.
Success is based on the selection of a model.
Most difficult and important task is the selection of a model close
enough to a physical system and yet simple enough to be
studies analytically.

1.2 OVERVIEW
The systems to be studied in this course are limited to linear systems.
Every linear system can be expressed by
Linear Systems
(Continuous-Time)

Linear and Time-Invariant

Systems
(Continuous-Time)

Input-output or
external
description

Linear System Theory Spring-2015

Internal
description
State-space
equations

Transfer Function Approach (input-output or external description)

LTI systems with zero initial conditions

y ( s ) g( s ) u ( s )
this textbook uses circumflex denotes the Laplace Transform of
the variable
Y(s)=G(s)U(s)
bold-case letters are used to represent matrix variables
Note: SISO Systems: Transfer Function MIMO Systems: Transfer
Matrix
1. The transfer function of a system is a mathematical model in
that it is an operational method of expressing the differential
equation that relates the output variable to the input variable.
2. The transfer function is a property of a system itself,
independent of the magnitude and nature of the input or driving
function.
3. The transfer function includes the units necessary to relate the
input to the output; however, it does not provide any information
concerning the physical structure of the system. (The transfer
functions of many physically different systems can be identical.)
4. If the transfer function of a system is known, the output or
response can be studied for various forms of inputs with a view
toward understanding the nature of the system.
5. If the transfer function of a system is unknown, it may be
established experimentally by introducing known inputs and
studying the output of the system. Once established, a transfer
function gives a full description of the dynamic characteristics of
the system, as distinct from its physical description.
Modern Control Theory
multiple-input, multiple-output.
linear or nonlinear systems.
time invariant or time varying
systems.
Linear System Theory Spring-2015

Conventional Control Theory

only linear time-invariant (LTI), singleinput single-output (SISO) systems.
Transfer Matrix for MIMO.
Non-linear systems are addressed
2

Systems are modeled in the

form of state-space equations.

Modern control theory is

essentially a time-domain
approach and frequency domain
approach in certain cases.

Linear System Theory Spring-2015

using linearization.
Systems are modeled in the form of
Transfer Function.

Conventional control theory is a

complex frequency-domain
approach.

Chapter 2:
SYSTEMS

MATHEMATICAL DESCRIPTION OF

2.1 INTRODUCTION
Continuous-Time Systems
A system is continuous-time if it accepts continuous-time signals as its
input and generates continuous-time signal as its output.
Discrete-Time Systems
A system is discrete-time if it accepts discrete-time signals as its input
and generates discrete-time signal as its output.

2.1.1 Causality and Lumpedness

Memoryless system: output y(t0) depends only on the input u(t)
at t0.
Example: resistive circuit.

Causal or nonanticipatory system: output y(t0) depends on past

input u(t), t<t0, and current input u(t0) but not on the future
input, u(t), t>t0

Noncausal or anticipatory system: the current output also

depends on future input. System can anticipate what will be
applied in the future.

No physical system can anticipate the future input. Therefore, every

physical system is causal.
Definition 2.1 The state x(t0) of a system at t0 is the information at t0
that, together with the input u(t), for t t0, determines uniquely the
output y(t) for all t t0.
If we know the state at t0, there is no more need to know the input u(t)
applied before t0 in determining the output y(t) after t0.

t0 t
Linear System Theory Spring-2015

Output is partly excited by the initial state x(t0) and partly by the input
u(t) applied at and after t0. It means that there is no need to know the
input applied before t0.
Lumped System: if its number of state variables is finite or its state is
a finite vector. A system is called distributed system if its state has
infinitely many state variables (for example, a transmission line).
Network shown below is a lumped system: its state consists of three
states x1, x2 and x3.

2.2 LINEAR SYSTEMS

Additivity and Homogeneity

Zero-Input Response
Zero-State Response
Input-output description (external description)
State-space equations (internal description)

x(t) = Ax(t) +
Bu(t)
y(t) = Dx(t) +
Du(t)

Linear System Theory Spring-2015

n: states
inputs

p:
q:

outputs

A: nn
np
C: qn
qp

B:
D:

Irrational function of s.
Unit-delay, as shown below, is a distributed system, infinitely many
state variables.
Exponential function e-s has infinite terms; hence it requires
infinitely many states.

( s) ( s)
e s 1 ( s)

2!
3!
2

Rational function of s
A lumped system has finite states and therefore the transfer
function is a rational function of s.
Every rational function of s can be expressed as

G( s)

N (s)
D( s ) .

G(s) is proper if deg D(s) deg N(s) and G() = zero or

nonzero constant.

G(s) is strictly proper if deg D(s) > deg N(s) and G() = 0.

G(s) is biproper if deg D(s) = deg N(s) and G() = nonzero

constant.

G(s) is improper if deg D(s) < deg N(s) and |G()| = .

Linear System Theory Spring-2015

Example of a Transfer Matrix of a system with 2-inputs and 2-outputs.

Linear System Theory Spring-2015

Lecture-2:

Feb. 07, 2015

State-space equations

x(t) = Ax(t) +
Bu(t)
y(t) = Dx(t) +
Du(t)
X(s) = [sI-A]-1 x(0) + [sI-A]-1 B U(s)
Y(s) = C[sI-A]-1 x(0) + C[sI-A]-1 BU(s) + DU(s)
zero-input
zero-state response

Transfer Function
with

MATLAB

x(0)=0:

G(s) = Y(s ) / U(s) = C[sI-A]-1 BU(s) + D

[A, B, C, D] = tf2ss(num, den)

[num, den] = ss2tf(A, B, C, D)
step(num, den)
sys = tf(num, den)
step(sys)
[y, t] = step(sys)
[y, x, t] = step(A, B, C, D)

Laplace transform is not used in studying linear time-varying systems

because
L [A(t) x(t)] L [A(t)] L [x(t)]
2.3.1 Op-Amp Circuit Implementation
A system represented in state-space equations.

Draw simulation diagram from state equations

Linear System Theory Spring-2015

Simulation
diagram

Op-Amp implementation

Note: With appropriate position of inverters, a system can be

implemented with minimum number of Op-Amps.

Linear System Theory Spring-2015

Lecture-3:

Feb. 13, 2015

2.4 LINEARIZATION
The principle of superposition does not apply to nonlinear systems.
Thus, for a nonlinear system the response to two inputs cannot be
calculated by treating one input at a time and adding the results.
Although many physical relationships are often represented by linear
equations, in most cases actual relationships are not quite linear.
In fact, a careful study of physical systems reveals that even so-called
linear systems are really linear only in limited operating ranges.
In practice, many electromechanical systems, hydraulic systems,
pneumatic systems, and so on, involve nonlinear relationships among
the variables.
For example, the output of a component may saturate for large input
signals. There may be a dead space that affects small signals. (The
dead space of a component is a small range of input variations to
which the component is insensitive.)
Linearization of Nonlinear Systems
In control engineering a normal operation of the system may be
around an equilibrium point, and the signals may be considered small
signals around the equilibrium. (It should be pointed out that there are
many exceptions to such a case.)
However, if the system operates around an equilibrium point and if the
signals involved are small signals, then it is possible to approximate
the nonlinear system by a linear system.
Such a linear system is equivalent to the nonlinear system considered
within a limited operating range. Such a linearized model (linear, timeinvariant model) is very important in control engineering.
Linear Approximation of Nonlinear Mathematical Models
Taylor series expansion
Input:
x(t)
Output: y(t)
Inputs: x1(t) &
x2(t)
Output: y(t)

______________________
Linearize
Linear System Theory Spring-2015

10

Linear System Theory Spring-2015

11

2.5 Examples
Mechanical Systems:

Spring-mass-damper systems

Characteristic of damper

Characteristic of spring

Suspension system of an automobile

Input: xi, the vertical displacement due to road condition

Output: xo, vertical displacement of vehicles body

Transfer function: Xo(s) / Xi(s) = ?

State equations: Assume x1= x0
(displacement)
energy
x2= d(x0)/dt
(velocity)
kinetic energy
u = xi
y = x0

Linear System Theory Spring-2015

potential

12

Example 2.7
Example 2.8
Electrical Systems:

RLC Networks

Transfer function

Write differential equations using KVL (loop equations) and KCL

(node equations).
Take Laplace transform of differential equations and find transfer
function. OR
Convert RLC network in s-domain (C 1/Cs, L Ls) and find
transfer function using loop/node equations.

Transfer function

State variable: inductor current and capacitor voltage.

Write state equations using loop and node equations.
One state variable for capacitors in parallel, one state variable
for inductors in series.

Example 2.11

Linear System Theory Spring-2015

13

Example 2.12

Transformation of Mathematical Models with MATLAB

[A,B,C,D] = tf2ss(num,den)

% transfer function to state-space

conversion
[num,den] = ss2tf(A,B,C,D)

% state-space to transfer function

conversion
[num,den] = ss2tf(A,B,C,D,iu)% system with more than one inputs
RLC network of Figure 2.16
A = [-1/6 0 -1/3; 0 0 1; 1.2 -1/2 -1/2];
0];
C = [1 -1 -1; -0.5 0 0]; D = 0;

B = [1/6 1/3; 0 0; 0

Find transfer function matrix of the system using MAT LAB

Find the step response using MATLAB
From: In(1)

0.3

Step Response
From: In(2)

To: Out(1)

0.2
0.1
0

To: Out(2)

Amplitude

-0.1
-0.2
0.5
0
-0.5
-1

20

Example 2.13

Linear System Theory Spring-2015

40

60

Time (seconds)

20

40

60

14

End of Lecture-3 Feb 13, 2015

Lecture-4:
Feb. 14, 2015
Lecture-4 cancelled due to PEC visit to Karachi. Makeup will be
arranged later.
Lecture-5:

Feb. 20, 2015

2.6 DISCRETE-TIME SYSTEMS

2.7 CONCLUDING REMARKS

2.8 PROBLEMS

Chapter-3

LINEAR ALGEBRA
Lecture-6:

Feb. 21, 2015

Lecture-7:

Feb. 27, 2015

Lecture-8:

Feb. 28, 2015

Lecture-9:

Mar. 06, 2015

Lecture-10:

Mar. 07, 2015

Linear System Theory Spring-2015

15

Lecture-11:

Mar13, 2015

Lecture-12

Mar. 14, 2015

Lecture-13:

Mar. 20, 2015

Lecture-14:

Mar. 21, 2015

Midterm Examination

Linear System Theory Spring-2015

16

Force

b1>b2
b1
Viscous

Static

b2
Coulomb
Velocity

Linear System Theory Spring-2015

17

EE-503
Weeks

Linear System Theory

Spring 2015

Topics

Readings

Introduction to Linear Systems: LTI systems, Modeling in

Continuous-Time and Discrete-Time, System Realization of
continuous-time and discrete-time systems

Linear Algebra: Basis, Representation and

Orthonormlaization, Similarity Transformation, Diagonal Form
and Jordan Form, Functions of a Square Matrix, Lyapnov
Equation, Singular Value Decomposition

Ch. 3

Solution of LTI State Equations, Discretization and Solution of

Discrete-Time State Equations, Equivalent State Equations,
Canonical Forms, Magnitude Scaling, State-Space Realization,
Examples

Ch. 4

Stability: Input-Output Stability of LTI Systems, Internal

Stability, Lyapnov Theorem, Examples

Ch. 5

Ch. 1, 2

M I D T E R M EXAMINATION
2

Controllability and Observability: Introduction to

Controllability, Observability, Controllability Canonical Form,
Observability Canonical Form, Examples

Ch. 6

Canonical Decomposition, Conditions in Jordan-Form

Equations, Discrete-Time State Equations, Examples

Ch. 6

State Feedback and State Estimation: State Feedback,

Regulation and Tracking, , State Estimation, Reduced Order
State Estimators, Examples

Ch. 8

Pole Placement, Compensator Equations using Classical

Method, Poles Placement of Unity-Feedback Systems,
Implementation of Compensators

Ch. 9

FINAL T E R M EXAMINATION

Textbook:
Press, 1999.

C. T. Chen, Linear System Theory and Design, Oxford University

Linear System Theory Spring-2015

18

C. T. Chen, Linear System Theory and Design, Oxford University

Press, 1999.
Table of Contents
Chapter 1: INTRODUCTION
1.1 Introduction

Wee
k
1

Lectu
re
1

Date
Feb. 09,
2015

1.2 Overview
Chapter 2: MATHEMATICAL DESCRIPTION OF
SYSTEMS
2.1 Introduction
2.1.1 Causality and Lumpedness
2.2 Linear Systems
2.3 Linear Time-Invariant (LTI) Systems
2.3.1 Op-Amp Circuit Implementation
2.4 Linearization

Feb. 10,
2015

2.5 Examples
2.5.1 RLC Networks
2.6 Discrete-Time Systems

Feb. 15,
2015

2.7 Concluding Remarks

Chapter 3: LINEAR ALGEBRA
3.1 Introduction
3.2 Basis, Representation and Orthonormalization
3.3 Linear Algebraic Equations
3.4 Similarity Transformation
3.5 Diagonal Form and Jordan Form
3.6 Functions of Square Matrix
3.7 Lyapunov Equation
3.8 Some Useful Formulas
3.9 Quadratic Form and Positive Definiteness
3.10 Singular Value Decomposition
3.11 Norms of Matrices
CHAPTER 4: STATE-SPACE SOLUTIONS AND
REALIZATIONS
CHAPTER 5: STABILITY
Chapter 6: CONTROLLABILITY AND
OBSERVABILITY
Chapter 8: STATE-FEEDBACK AND STATE
ESTIMATORS

Linear System Theory Spring-2015

19