Professional Documents
Culture Documents
ebook download (eBook PDF) Linear System Theory and Design 4th Edition all chapter
ebook download (eBook PDF) Linear System Theory and Design 4th Edition all chapter
https://ebooksecure.com/download/power-electronic-system-design-
linking-differential-equations-linear-algebra-and-implicit-
functions-ebook-pdf/
http://ebooksecure.com/product/ebook-pdf-digital-control-system-
analysis-design-4th-edition/
http://ebooksecure.com/product/ebook-pdf-power-system-analysis-
and-design-6th-edition/
https://ebooksecure.com/download/thermal-system-design-and-
simulation-ebook-pdf/
(eBook PDF) The Analysis and Design of Linear Circuits,
8th Edition
http://ebooksecure.com/product/ebook-pdf-the-analysis-and-design-
of-linear-circuits-8th-edition/
http://ebooksecure.com/product/ebook-pdf-organization-theory-
design-an-international-perspective-4th-edition/
http://ebooksecure.com/product/ebook-pdf-linear-systems-theory-
second-edition-2nd-edition/
http://ebooksecure.com/product/ebook-pdf-differential-equations-
and-linear-algebra-4th-edition-2/
http://ebooksecure.com/product/ebook-pdf-linear-algebra-and-its-
applications-4th-edition/
Contents
Preface xi
CHAPTER 1
Introduction 1
1.1 Introduction 1
1.2 Overview 2
1.2.1 A Brief History 4
CHAPTER 2
vii
viii CONTENTS
CHAPTER 3
Linear Algebra 55
3.1 Introduction 55
3.2 Basis, Representation, and Orthonormalization 56
3.3 Linear Algebraic Equations 61
3.4 Similarity Transformation 66
3.5 Diagonal Form and Jordan Form 68
3.6 Functions of a Square Matrix 75
3.7 Lyapunov Equation 84
3.8 Some Useful Formulas 86
3.9 Quadratic Form and Positive Definiteness 87
3.10 Singular Value Decomposition 91
3.11 Norms of Matrices 93
Problems 95
CHAPTER 4
CHAPTER 5
Stability 149
5.1 Introduction 149
5.2 Input–Output Stability of LTI Systems 149
5.3 Discrete-Time Case 158
CONTENTS ix
CHAPTER 6
CHAPTER 7
CHAPTER 8
CHAPTER 9
References 369
Answers to Selected Problems 371
Index 381
Preface
T
his text is intended for use in senior/first-year graduate courses on linear systems in
electrical, mechanical, bioengineering, chemical, and aeronautical departments. It
may also be useful to practicing engineers because it contains many design proce-
dures. The mathematical background assumed is a working knowledge of linear algebra
and the Laplace transform and an elementary knowledge of differential equations. A
knowledge of the sophomore/junior subject area Signals and Systems such as the one
in Reference 10 is helpful but not essential.
Linear system theory is a vast subject area. This text studies mainly linear time-
invariant lumped systems which are describable by state-space equations and rational
transfer functions. The former is an internal description, and the latter is an external
description. We study their structures, relationships, and implications in design. As this
is an engineering text, we aim to achieve two objectives. The first one is to use simple and
efficient methods to develop results and design procedures. Thus the presentation is not
exhaustive. For example, we skip many multivariable companion-forms in state-space
equations and the Smith–McMillan form in transfer matrices. The second objective is
to enable the reader to employ the results to carry out design. Thus most results are
discussed with an eye toward numerical computation. All design procedures in the text
can be carried out using MATLAB.1 We adopt the theorem–proof format to cultivate
the reader’s ability to think critically and to develop ideas logically.
1 MATLAB is a registered trademark of the MathWorks, Inc., 24 Prime Park Way, Natick, MA 01760-1500.
http://www.mathworks.com.
xi
xii PREFACE
This text also touches upon linear time-invariant distributed systems and linear
time-varying lumped systems. We use examples to show that some results in this text
are not applicable to those systems. Even so, this text will provide a foundation and a
benchmark for studying distributed, time-varying, or nonlinear systems.
The first edition of this text, entitled Introduction to Linear System Theory, was
published in 1970. Its second edition, renamed Linear System Theory and Design, was
published in 1984. It expanded to 662 pages from 431 pages of the original edition. The
third edition, published in 1999, cut the second edition in half to 332 pages by skipping
many topics that are either of only academic interests or of limited practical use. The
third edition also introduced the two-parameter (feedforward/feedback) configuration,
which is more suitable for practical application.
Before carrying out this revision, Oxford University Press sent out the third edition
for reviews. Two reviewers suggested to drop Chapters 7 and 9 because they were
not covered in their courses. Indeed, the multi-input and multi-output (MIMO) parts
of those two chapters, which are included for completeness, should not be covered;
they are more suitable for advanced courses. It is, however, suggested to cover the
single-input single-output (SISO) parts because they relate the concept of coprimeness
in transfer functions and the concepts of controllability and observability in state-
space equations and establish the equivalence of the two descriptions. They also use
simpler mathematics to develop results that are more general than those obtained using
state-space equations.
Chi-Tsong Chen
July 2012
This page intentionally left blank
1
CHAPTER
Introduction
1.1 Introduction
The study and design of physical systems can be carried out using empirical methods.
We can apply various signals to a physical system and measure its responses. If the
performance is not satisfactory, we can adjust some of its parameters or connect to
it a compensator to improve its performance. This approach relies heavily on past
experience and is carried out by trial and error, and it has succeeded in designing many
physical systems.
Empirical methods may become unworkable if physical systems are complex, or
they may become too expensive or too dangerous to be experimented on. In these cases,
analytical methods become indispensable. The analytical study of physical systems
consists of four parts: modeling, development of mathematical descriptions, analysis,
and design. We briefly introduce each of these tasks.
The distinction between physical systems and models is important in engineering.
For example, circuits or control systems studied in any textbook are models of physical
systems. A resistor with a constant resistance is a model; it will burn out if the applied
voltage is over a limit. This power limitation is often disregarded in its analytical study.
An inductor with a constant inductance is again a model; in reality, the inductance
may vary with the amount of current flowing through it. Modeling is a very important
problem, for the success of the design depends upon whether the physical system is
properly modeled.
A physical system may have different models, depending on the questions asked.
It may also be modeled differently in different operational ranges. For example, an
electronic amplifier is modeled differently at high and low frequencies. A spaceship
can be modeled as a particle in investigating its trajectory; however, it must be modeled
as a rigid body in maneuvering. A spaceship may even be modeled as a flexible body
1
2 INTRODUCTION
1.2 Overview
The study of systems consists of four parts: modeling, setting up mathematical
equations, analysis, and design. Developing models for physical systems requires
knowledge of the particular field and some measuring devices. For example, to develop
models for transistors requires a knowledge of quantum physics and some laboratory
setup. Developing models for automobile suspension systems requires actual testing
and measurements; it cannot be achieved by use of pencil and paper. Computer sim-
ulation certainly helps but cannot replace actual measurements. Thus the modeling
problem should be studied in conjunction with the specific field and cannot be properly
covered in this text. In this text, we shall assume that models of physical systems are
available to us.
The systems to be studied in this text are mostly limited to linear (L), time-
invariant (TI), and lumped. This class of systems with input u(t) and output y(t) can
be described by
OVERVIEW 3
1. Convolution:
t
y(t) = g(t − τ )u(τ ) dτ (1.1)
τ =0
start with the continuous-time (CT) case because most physical systems are continuous
time and then discuss their discrete-time (DT) counterparts.
We will also touch upon LTI distributed systems described by irrational transfer
functions, as well as linear and lumped systems described by time-varying state-space
equations. We show that some results for LTI lumped systems may not be applicable to
linear time-varying or distributed systems. Thus their study is much more complicated.
The study of nonlinear systems is even more so. To study them, this text however will
provide a foundation and a benchmark.
2
Mathematical
Descriptions of Systems
2.1 Introduction
This text models a system as a black box with one or more input terminals and one or
more output terminals as shown in Fig. 2.1. We assume that if an excitation or input is
applied to the input terminal, a unique response or output signal can be measured at the
output terminal. This unique relationship between the excitation and response, input
and output, or cause and effect is essential in defining a system. A system with only one
input terminal and only one output terminal is called a single-input single-output (SISO)
system. A system with two or more input terminals and two or more output terminals
is called a multi-input multi-output (MIMO) system. Likewise, a single-input multi-
output (SIMO) system has only one input terminal and two or more output terminals.
A MISO system has multi-input terminals and single-output terminal.
A signal is called a continuous-time (CT) signal if it is defined at every instant of
time. A system is called a CT system if it accepts CT signals as its input and generates
CT signals as its output. The input will be denoted by lowercase italic u(t) for single
input or by boldface u(t) for multiple inputs. If a system has p input terminals, then
u(t) is a p × 1 vector or u = [u1 u2 . . . up ] , where the prime denotes the transpose.
Similarly, the output will be denoted by y(t) or y(t). The time t is mostly assumed to
range from −∞ to ∞.
A signal is called a discrete-time (DT) signal if it is defined only at discrete
instants of time. A system is called a DT system if it accepts DT signals as its input and
generates DT signals as its output. All DT signals in a system will be assumed to have
the same sampling period T . The input and output will be denoted by u[k] := u(kT ) and
y[k] := y(kT ), where the integer k, ranging mostly from −∞ to ∞, is called the time
index and kT denotes discrete time instant. For multiple inputs and multiple outputs,
we use boldface u[n] and y[n].
6
CAUSALITY, LUMPEDNESS, AND TIME INVARIANCE 7
Definition 2.1 The state x(t0 ) of a system at time 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 .
By definition, 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 . Thus in some sense, the
state summarizes the effect of past input on future output. For the circuit shown in
Fig. 2.2, if we know the voltages x1 (t0 ) and x2 (t0 ) across the two capacitors and the
current x3 (t0 ) passing through the inductor, then for any input applied on and after t0
8 MATHEMATICAL DESCRIPTIONS OF SYSTEMS
we can determine uniquely the output for t ≥ t0 . Thus the state of the circuit at time t0 is
⎡ ⎤
x1 (t0 )
x(t0 ) = ⎣ x2 (t0 ) ⎦
x3 (t0 )
It is a 3 × 1 vector. The entries of x are called state variables. Thus, in general, we may
consider the state simply as a set of initial conditions. We call x(t0 ) the initial state.
Using the initial state at t0 , we can express the input and output of a system as
x(t0 )
→ y(t), t ≥ t0 (2.1)
u(t), t ≥ t0
It means that the output is partly excited by the initial state at t0 and partly by the input
applied at and after t0 . In using (2.1), there is no more need to know the input applied
before t0 all the way back to −∞. Thus (2.1) is easier to track and will be called a
state-input–output pair.
A system is said to be lumped if its number of state variables is finite or its state
is a finite vector. The circuit in Fig. 2.2 is clearly a lumped system; its state consists of
three numbers. A system is called a distributed system if its state has infinitely many
state variables. The transmission line is the most well known distributed system. We
give one more example.
y(t) = u(t − 1)
The output is simply the input delayed by one second. In order to determine {y(t), t ≥ t0 }
from {u(t), t ≥ t0 }, we need the information {u(t), t0 − 1 ≤ t < t0 }. Therefore, the
initial state of the system is {u(t), t0 − 1 ≤ t < t0 }. There are infinitely many points in
{t0 − 1 ≤ t < t0 }. Thus the unit-time delay system is a distributed system.
be any pair of a system. If the system is time invariant, then we have, for any t1 ,
x(t0 + t1 )
→ y(t − t1 ), t ≥ t0 + t1 (time shifting)
u(t − t1 ), t ≥ t0 + t1
where u(t − t1 ) and y(t − t1 ) are shifting, respectively, of u(t) and y(t) from t0 to
t0 + t1 . The equation means that if the initial state is shifted to time t0 + t1 and the same
input waveform is applied from t0 + t1 instead of from t0 , then the output waveform
will be the same except that it starts to appear from time t0 + t1 . In other words, if the
initial state and the input are the same, no matter at what time they are applied, the
output waveform will always be the same. Therefore, for time-invariant systems, we
can always assume, without loss of generality, that t0 = 0. Note that the initial time
t0 = 0 is not absolute; it can be selected by us. It may be the instant we start to study a
system. If we select t0 = 0, then the time interval of interest will be [0, ∞). If a system
is not time invariant, it is said to be time varying.
Some physical systems must be modeled as time-varying systems. For example,
a burning rocket is a time-varying system, because its mass decreases rapidly with
time. Although the performance of an automobile or a TV set may deteriorate over a
long period of time, its characteristics do not change appreciable in the first couple of
years. Thus a large number of physical systems can be modeled as time invariant over
a limited time period.
2.2.1 Impulses
We need the concept of impulses to develop some mathematical equations. Consider
the pulse defined by
1/ t1 ≤ t < t0 +
δa (t − t1 ) :=
0 t < t1 and t ≥ t1 +
and shown in Fig. 2.3. It is located at time t1 and has width and height 1/. Its area
or any integration covering the pulse equals 1 for any > 0. The impulse at t = t1 is
then defined as
δ(t − t1 ) := lim δ (t − t1 )
→0
It equals 0 for all t = t1 and ∞ at t = t1 . Its area or any integration covers the impulse
equals 1 such as
∞ t1 + t1
δ(t − t1 ) dt = δ(t − t1 ) dt = δ(t − t1 ) dt = 1
t=−∞ t=t1 t=t1
where the impulse is located at t = 2.5. The impulse has the following sifting property
∞ t1
f (t)δ(t − t1 ) dt = f (t)δ(t − t1 ) dt
t=−∞ t=t1
= f (t)|t−t1 =0 = f (t)|t=t1 = f (t1 )
for any function f (t) that is continuous at t1 . The impulse at t1 sifts out the value of f (t)
at t = t1 . In general, whenever a function is multiplied by an impulse in an integration,
we simply move the function outside the integration and then replace the integration
variable by the variable obtained by equating the argument of the impulse to zero.
We discuss an application. Consider the signal shown in Fig. 2.4. It can be approx-
imated by a staircase function formed from a sequence of pulses as shown. We call
the step size. The pulse in Fig. 2.3 has height 1/; its multiplication by or δ (t − ti )
has height 1. The leftmost pulse in Fig. 2.4 has height u(ti ), thus it can be expressed as
LA MORAL Y EL ARTE