Full Ebook of Elements of Applied Bifurcation Theory 4Th Edition Yuri A Kuznetsov Online PDF All Chapter
Full Ebook of Elements of Applied Bifurcation Theory 4Th Edition Yuri A Kuznetsov Online PDF All Chapter
Full Ebook of Elements of Applied Bifurcation Theory 4Th Edition Yuri A Kuznetsov Online PDF All Chapter
https://ebookmeta.com/product/elements-of-sociology-a-critical-
canadian-introduction-4th-edition-oxford/
https://ebookmeta.com/product/elements-of-category-theory-1st-
edition-emily-riehl/
https://ebookmeta.com/product/group-theory-of-chemical-
elements-1st-edition-abram-i-fet/
https://ebookmeta.com/product/megatrends-of-world-politics-1st-
edition-marina-m-lebedeva-editor-denis-a-kuznetsov-editor/
https://ebookmeta.com/product/theory-of-applied-robotics-
kinematics-dynamics-and-control-reza-n-jazar/
https://ebookmeta.com/product/energy-transition-and-the-local-
community-a-theory-of-society-applied-to-hazleton-pennsylvania-
dan-rose/
https://ebookmeta.com/product/identical-relations-in-lie-
algebras-2nd-edition-yuri-a-bahturin/
https://ebookmeta.com/product/revolution-and-the-rebirth-of-
inequality-a-theory-applied-to-the-national-revolution-in-
bolivia-johathan-kelley/
Applied Mathematical Sciences
Yuri A. Kuznetsov
Elements
of Applied
Bifurcation
Theory
Fourth Edition
Applied Mathematical Sciences
Founding Editors
F. John
J. P. LaSalle
L. Sirovich
Volume 112
Series Editors
Anthony Bloch, Department of Mathematics, University of Michigan, Ann Arbor,
MI, USA
C. L. Epstein, Department of Mathematics, University of Pennsylvania,
Philadelphia, PA, USA
Alain Goriely, Department of Mathematics, University of Oxford, Oxford, UK
Leslie Greengard, New York University, New York, NY, USA
Advisory Editors
J. Bell, Center for Computational Sciences and Engineering, Lawrence Berkeley
National Laboratory, Berkeley, CA, USA
P. Constantin, Department of Mathematics, Princeton University, Princeton, NJ,
USA
R. Durrett, Department of Mathematics, Duke University, Durham, CA, USA
R. Kohn, Courant Institute of Mathematical Sciences, New York University,
New York, NY, USA
R. Pego, Department of Mathematical Sciences, Carnegie Mellon University,
Pittsburgh, PA, USA
L. Ryzhik, Department of Mathematics, Stanford University, Stanford, CA, USA
A. Singer, Department of Mathematics, Princeton University, Princeton, NJ, USA
A. Stevens, Department of Applied Mathematics, University of Münster, Münster,
Germany
S. Wright, Computer Sciences Department, University of Wisconsin, Madison, WI,
USA
The mathematization of all sciences, the fading of traditional scientific boundaries,
the impact of computer technology, the growing importance of computer modeling
and the necessity of scientific planning all create the need both in education and
research for books that are introductory to and abreast of these developments. The
purpose of this series is to provide such books, suitable for the user of mathematics,
the mathematician interested in applications, and the student scientist. In particular,
this series will provide an outlet for topics of immediate interest because of the
novelty of its treatment of an application or of mathematics being applied or lying
close to applications. These books should be accessible to readers versed in
mathematics or science and engineering, and will feature a lively tutorial style, a
focus on topics of current interest, and present clear exposition of broad appeal.
A compliment to the Applied Mathematical Sciences series is the Texts in Applied
Mathematics series, which publishes textbooks suitable for advanced undergraduate
and beginning graduate courses.
Yuri A. Kuznetsov
Elements of Applied
Bifurcation Theory
Fourth Edition
Yuri A. Kuznetsov
Department of Mathematics
Utrecht University
Utrecht, The Netherlands
Department of Applied Mathematics
University of Twente
Enschede, The Netherlands
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
To my family
Preface to the Fourth Edition
After almost two decades since the appearance of the third edition of this book, it
is time for a new revision of the text. While the original structure and selection of
material proved to be reasonable and can be preserved, the progress in research and
education asked for additions, improvements, polishing, and—unfortunately—error
corrections. Apparently, many people read and actually used the previous editions.
Moreover, by an agreement between Springer and Science Press (Beijing), a Chinese
translation has been published in 2010 that remained until now the most up-to-date
version.
As before, the book structure has been kept intact. Most of the changes reflect
recent theoretical and software developments in which the author was involved.
Hopefully, the book will remain readable for students and applied researchers with
a moderate mathematical background. Major changes in the fourth edition can be
summarized as follows.
Chapter 1. Introduction to Dynamical Systems. Appendix A is added that
discusses the general properties of orbits in planar autonomous ODEs, including
proofs of the classical Poinaré-Bendixson Theorem and Bendixson-Dulac Criterion.
While this material is treated in many standard textbooks, a brief summary of the
main ideas is included here for completeness. Another new Appendix C is added that
introduces an important class of infinite-dimensional dynamical systems, i.e., DDEs
(ordinary differential equations with finite time delays). Such equations describe
many important natural and industrial processes and appear, e.g., in the mathematical
modeling of epidemics or lasers with optical feedback. It is explained why such
systems have infinite-dimensional phase spaces, and how to determine the stability
of their equilibria.
Chapter 2. Topological Equivalence, Bifurcations, and Structural Stability
of Dynamical Systems. More results on the topological equivalence of scalar maps
are given, including a condition for the equivalence of strictly increasing maps, with
a proof based on the fundamental domains. This result is then used several times in
the other chapters.
Chapter 3. One-Parameter Bifurcations of Equilibria in Continuous-Time
Dynamical Systems. The construction of the local conjugating homeomorphism for
vii
viii Preface to the Fourth Edition
Bibliographical notes in all chapters and References are updated. The text,
including some additional material provided to students as online notes, is used
many times in the Bachelor’s and Master’s courses on nonlinear dynamical systems
and their numerical analysis, both at the Utrecht University and the University of
Twente. The new approaches emerged in a joint work and numerous discussions
with many colleagues, in particular, with O. Diekmann and H. Hanssmann (Utrecht
University), W. Govaerts (Ghent University), S. van Gils and H. G. E. Meijer (Univer-
sity of Twente), and E. Doedel (Concordia University). The translator C. F. Jin of
the third edition into Chinese has found a number of unnoticed misprints in the third
edition, which are now corrected. Other remarks and suggestions from all over the
world have helped to improve the text.
There is one global change in the notation that is useful to mention here: An
element x ∈ Rn is written in one of the following equivalent forms:
⎛ ⎞
x1
⎜ ⎟
⎜ x2 ⎟
(x1 , x2 , . . . , xn ) = (x1 x2 · · · xn )T = ⎜ ⎟
⎜ .. ⎟,
⎝. ⎠
xn
The years that have passed since the publication of the first edition of this book proved
that the basic principles used to select and present the material made sense. The idea
was to write a simple text that could serve as a serious introduction to the subject. Of
course, the meaning of “simplicity” varies from person to person and from country
to country. The word “introduction” contains even more ambiguity. To start reading
this book, only a moderate knowledge of linear algebra and calculus is required.
Other preliminaries, qualified as “elementary” in modern mathematics, are explicitly
formulated in the book. These include the Fredholm Alternative for linear systems
and the multidimensional Implicit Function Theorem. Using these very limited tools,
a framework of notions, results, and methods is gradually built that allows one to read
(and possibly write) scientific papers on bifurcations of nonlinear dynamical systems.
Among other things, progress in the sciences means that mathematical results and
methods that once were new become standard and routinely used by the research and
development community. Hopefully, this edition of the book will contribute to this
process.
The book’s structure has been kept intact. Most of the changes introduced reflect
recent theoretical and software developments in which the author was involved.
Important changes in the third edition can be summarized as follows. A new section
devoted to the fold-flip bifurcation for maps has appeared in Chapter 9. We derive
there a parameter-dependent normal form for the map having at critical parameter
values a fixed point with eigenvalues ±1, approximate this normal form by a flow,
present generic bifurcation diagrams, and discuss their relationships with the original
map. The treatment of the strong resonance 1:1 is extended considerably along similar
lines.
A modern technique, due to Coullet and Spiegel, of the simultaneous center mani-
fold reduction and normalization on the center manifold is introduced in Chapters 8
and 9. Using this technique, explicit formulas for the normal form coefficients for
codim 2 local bifurcations of n-dimensional ODEs and maps are derived, including
the fold-flip bifurcation. The resulting formulas are independent of n and equally
suitable for both symbolic and numerical evaluation in the original basis. The center
xi
xii Preface to the Third Edition
manifold computations for codim 1 local bifurcations in Chapter 5 were also recti-
fied. The reader should be warned that the scaling of some coefficients differs slightly
from that in the previous editions.
The Hénon map now plays a prominent role in the book. It is used to illustrate
the Smale horseshoe in Chapter 1 and the codim 1 bifurcation curves in Chapter 4.
In Chapter 7, it is shown that this map approximates rescaled return maps defined
near a nontransversal homoclinic orbit to a saddle limit cycle. Using these facts, we
sketch a proof of the classical results by Gavrilov and Shil’nikov concerning the
accumulation of folds and flips on this bifurcation.
Following numerous suggestions, two new appendices are included. Appendix
B to Chapter 3 gives a rather standard introduction to the theory of the Poincaré
normal forms of ODEs near an equilibrium. This provides the reader with a uniform
point of view on various normal form computations in the book. Appendix B to
Chapter 10 is devoted to an elementary treatment of the bialternate matrix product.
This product has proved to be an important tool in the numerical analysis of Hopf
and Neimark-Sacker bifurcations but it is not considered in the standard courses on
Linear Algebra.
While preparing this edition, I have corrected misprints and errors found in the
second edition. The bibliography was extended and checked against the AMS Math-
SciNet reference database; if known, the references to English translations are added.
All figures were revised, and in some cases relevant data were recomputed.
I am thankful to all colleagues for comments, suggestions, and discussions that
led to text improvements. Finally, may the constant support by my wife, Lioudmila,
and my daughters, Elena and Ouliana, be acknowledged.
The favorable reaction to the first edition of this book confirmed that the publication
of such an application-oriented text on bifurcation theory of dynamical systems
was well timed. The selected topics indeed cover major practical issues of applying
the bifurcation theory to finite-dimensional problems. This new edition preserves
the structure of the first edition while updating the context to incorporate recent
theoretical developments, in particular, new and improved numerical methods for
bifurcation analysis. The treatment of some topics has been clarified.
Major additions can be summarized as follows: In Chapter 3, an elementary proof
of the topological equivalence of the original and truncated normal forms for the fold
bifurcation is given. This makes the analysis of codimension-one equilibrium bifur-
cations of ODEs in the book complete. This chapter also includes an example of the
Hopf bifurcation analysis in a planar system using MAPLE, a symbolic manipulation
software. Chapter 4 includes a detailed normal form analysis of the Neimark-Sacker
bifurcation in the delayed logistic map. In Chapter 5, we derive explicit formulas
for the critical normal form coefficients of all codim 1 bifurcations of n-dimensional
iterated maps (i.e., fold, flip, and Neimark-Sacker bifurcations). The section on homo-
clinic bifurcations in n-dimensional ODEs in Chapter 6 is completely rewritten and
introduces the Melnikov integral that allows us to verify the regularity of the mani-
fold splitting under parameter variations. Recently proved results on the existence
of center manifolds near homoclinic bifurcations are also included. By their means
the study of generic codim 1 homoclinic bifurcations in n-dimensional systems is
reduced to that in some two-, three-, or four-dimensional systems. Two- and three-
dimensional cases are treated in the main text, while the analysis of bifurcations in
four-dimensional systems with a homoclinic orbit to a focus-focus is outlined in the
new appendix. In Chapter 7, an explicit example of the “blue sky” bifurcation is
discussed. Chapter 10, devoted to the numerical analysis of bifurcations, has been
changed most substantially. We have introduced bordering methods to continue fold
and Hopf bifurcations in two parameters. In this approach, the defining function
for the bifurcation used in the minimal augmented system is computed by solving
a bordered linear system. It allows for explicit computation of the gradient of this
xiii
xiv Preface to the Second Edition
function, contrary to the approach when determinants are used as the defining func-
tions. The main text now includes BVP methods to continue codim 1 homoclinic
bifurcations in two parameters, as well as all codim 1 limit cycle bifurcations. A
new appendix to this chapter provides test functions to detect all codim 2 homoclinic
bifurcations involving a single homoclinic orbit to an equilibrium. The software
review in the last appendix to this chapter is updated to present recently developed
programs, including AUTO97 with HomCont, DsTool, and CONTENT providing
the information on their availability by ftp.
A number of misprints and minor errors have been corrected while preparing
this edition. I would like to thank many colleagues who have sent comments and
suggestions, including E. Doedel (Concordia University, Montreal), B. Krauskopf
(VU, Amsterdam), S. van Gils (TU Twente, Enschede), B. Sandstede (WIAS,
Berlin), W.-J. Beyn (Bielefeld University), F.S. Berezovskaya (Center for Ecological
Problems and Forest Productivity, Moscow), E. Nikolaev and E.E. Shnoll (IMPB,
Pushchino, Moscow Region), W. Langford (University of Guelph), O. Diekmann
(Utrecht University), and A. Champneys (University of Bristol).
I am thankful to my wife, Lioudmila, and to my daughters, Elena and Ouliana,
for their understanding, support, and patience, while I was working on this book and
developing the bifurcation software CONTENT.
Finally, I would like to acknowledge the Research Institute for Applications of
Computer Algebra (RIACA, Eindhoven) for the financial support of my work at CWI
(Amsterdam) in 1995–1997.
During the last few years, several good textbooks on nonlinear dynamics have
appeared for graduate students in applied mathematics. It seems, however, that the
majority of such books are still too theoretically oriented and leave many practical
issues unclear for people intending to apply the theory to particular research prob-
lems. This book is designed for advanced undergraduate or graduate students in math-
ematics who will participate in applied research. It is also addressed to professional
researchers in physics, biology, engineering, and economics who use dynamical
systems as modeling tools in their studies. Therefore, only a moderate mathemat-
ical background in geometry, linear algebra, analysis, and differential equations is
required. A brief summary of general mathematical terms and results, which are
assumed to be known in the main text, appears at the end of the book. Whenever
possible, only elementary mathematical tools are used. For example, we do not try
to present normal form theory in full generality, instead developing only the portion
of the technique sufficient for our purposes.
The book aims to provide the student (or researcher) with both a solid basis
in dynamical systems theory and the necessary understanding of the approaches,
methods, results, and terminology used in the modern applied mathematics literature.
A key theme is that of topological equivalence and codimension, or “what one may
expect to occur in the dynamics with a given number of parameters allowed to vary.”
Actually, the material covered is sufficient to perform quite complex bifurcation
analysis of dynamical systems arising in applications. The book examines the basic
topics of bifurcation theory and could be used to compose a course on nonlinear
dynamical systems or systems theory. Certain classical results, such as Andronov-
Hopf and homoclinic bifurcation in two-dimensional systems, are presented in great
detail, including self-contained proofs. For more complex topics of the theory, such
as homoclinic bifurcations in more than two dimensions and two-parameter local
bifurcations, we try to make clear the relevant geometrical ideas behind the proofs
but only sketch them or, sometimes, discuss and illustrate the results but give only
references of where to find the proofs. This approach, we hope, makes the book
readable for a wide audience and keeps it relatively short and able to be browsed. We
also present several recent theoretical results concerning, in particular, bifurcations
xv
xvi Preface to the First Edition
we hope the reader will find the book useful, especially as an interface between
undergraduate and postgraduate studies.
This book would have never appeared without the encouragement and help from
many friends and colleagues to whom I am very much indebted. The idea of such
an application-oriented book on bifurcations emerged in discussions and joint work
with A.M. Molchanov, A.D. Bazykin, E.E. Shnol, and A.I. Khibnik at the former
Research Computing Centre of the USSR Academy of Sciences (Pushchino). S.
Rinaldi asked me to prepare and give a course on nonlinear systems at the Politec-
nico di Milano that would be useful for applied scientists and engineers. O. Diekmann
(CWI, Amsterdam) was the first to propose the conversion of these brief lecture notes
into a book. He also commented on some of the chapters and gave friendly support
during the whole project. S. van Gils (TU Twente, Enschede) read the manuscript
and gave some very useful suggestions that allowed me to improve the content and
style. I am particularly thankful to A.R. Champneys of the University of Bristol, who
reviewed the whole text and not only corrected the language but also proposed many
improvements in the selection and presentation of the material. Certain topics have
been discussed with J. Sanders (VU/RIACA/CWI, Amsterdam), B. Werner (Univer-
sity of Hamburg), E. Nikolaev (IMPB, Pushchino), E. Doedel (Concordia Univer-
sity, Montreal), B. Sandstede (IAAS, Berlin), M. Kirkilonis (CWI, Amsterdam), J.
de Vries (CWI, Amsterdam), and others, whom I would like to thank. Of course,
the responsibility for all remaining mistakes is mine. I would also like to thank A.
Heck (CAN, Amsterdam) and V.V. Levitin (IMPB, Pushchino/CWI, Amsterdam) for
computer assistance. Finally, I thank the Nederlandse Organisatie voor Wetenschap-
pelijk Onderzoek (NWO) for providing financial support during my stay at CWI,
Amsterdam.
xxi
xxii Contents
This chapter introduces some basic terminology. First, we define a dynamical system
and give several examples, including symbolic dynamics. Then we introduce the
notions of orbits, invariant sets, and their stability. As we shall see while analyzing
the Smale horseshoe, invariant sets can have very complex structures. This is closely
related to the fact discovered in the 1960s that rather simple dynamical systems may
behave “randomly,” or “chaotically.” Finally, we discuss how differential equations
can define dynamical systems in both finite- and infinite-dimensional spaces.
All possible states of a system are characterized by the points of some set X . This set
is called the state space of the system. Actually, the specification of a point x ∈ X
must be sufficient not only to describe the current “position” of the system but also
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 1
Yu. A. Kuznetsov, Elements of Applied Bifurcation Theory, Applied Mathematical
Sciences 112, https://doi.org/10.1007/978-3-031-22007-4_1
2 1 Introduction to Dynamical Systems
m
ϕ̇
(q1 , q2 , . . . , qs , p1 , p2 , . . . , ps ),
where qi are the generalized coordinates, while pi are the corresponding generalized
momenta. Therefore, in this case, X = R2s . If k coordinates are cyclic, X = Sk ×
R2s−k . In the case of the pendulum, s = k = 1, q1 = ϕ, and we can take p1 = ϕ̇. ♦
|a1 |2 + |a2 |2 = 1.
1.1 Definition of a Dynamical System 3
The probability of finding the system in the ith state is equal to pi = |ai |2 , i = 1, 2.
♦
c = (c1 , c2 , . . . , cn ).
X = {c : c = (c1 , c2 , . . . , cn ) ∈ Rn , ci ≥ 0}. ♦
N = (N1 , N2 , . . . , Nn ) ∈ Rn .
Here Ni is the number (or density) of the ith species or other group (e.g., predators
or prey). ♦
where ωi ∈ {1, 2}. Note that the zero position in a sequence must be pointed out; for
example, there are two distinct periodic sequences that can be written as
ω = {. . . , 1, 2, 1, 2, 1, 2, . . .},
one with ω0 = 1, and the other with ω0 = 2. The space 2 will play an important
role in the following.
Sometimes, it is useful to identify two sequences that differ only by a shift of the
origin. Such sequences are called equivalent. The classes of equivalent sequences
form a set denoted by 2 . The two periodic sequences mentioned above represent
the same point in 2 . ♦
In all the above examples, the state space has a certain natural structure, allowing
for comparison between different states. More specifically, a distance ρ between two
states is defined, making these sets metric spaces.
4 1 Introduction to Dynamical Systems
In the examples from mechanics and in the examples from chemistry and ecol-
ogy, the state space was a real vector space Rn of some finite dimension n, or a
(sub-)manifold (hypersurface) in this space. The Euclidean norm can be used to mea-
sure the distance between two states parametrized by the points x, y ∈ Rn , namely
n
ρ(x, y) = x − y = x − y, x − y = (xi − yi )2 , (1.1)
i=1
If necessary, the distance between two (close) points on a manifold can be measured as
the minimal length of a curve connecting these points within the manifold. Similarly,
the distance between two states ψ, ϕ of the quantum system from Example 1.3 can
be defined using the standard scalar product in Cn ,
n
T
ψ, ϕ = ψ ϕ = ψ i ϕi ,
i=1
where
0 if ωk = θk ,
δ ωk θ k =
1 if ωk = θk .
According to this formula, two sequences are considered to be close if they have a
long block of coinciding elements centered at position zero (check!).
Using the previously defined distances, the introduced state spaces X are com-
plete metric spaces. Loosely speaking, this means that any sequence of states, all of
1.1 Definition of a Dynamical System 5
1.1.2 Time
The evolution of a dynamical system means a change in the state of the system
with time t ∈ T , where T is a number set. We will consider two types of dynamical
systems: those with continuous (real) time T = R1 , and those with discrete (integer)
time T = Z. Systems of the first type are called continuous-time dynamical systems,
while those of the second are termed discrete-time dynamical systems. Discrete-time
systems appear naturally in ecology and economics when the state of a system at a
certain moment of time t completely determines its state after a year, say at t + 1.
The main component of a dynamical system is an evolution law that determines the
state xt of the system at time t, provided the initial state x0 is known. The most
general way to specify the evolution is to assume that for given t ∈ T a map ϕt is
defined in the state space X ,
ϕt : X → X,
xt = ϕt x0 .
The map ϕt is often called the evolution operator of the dynamical system. It might
be known explicitly; however, in most cases, it is defined indirectly and can be
computed only approximately. In the continuous-time case, the family {ϕt }t∈T of
evolution operators is called a flow.
Note that ϕt x may not be defined for all pairs (x, t) ∈ X × T . Dynamical systems
with evolution operator ϕt defined for both t ≥ 0 and t < 0 are called invertible. In
such systems the initial state x0 completely defines not only the future states of the
system, but its past behavior as well. However, it is useful to consider also dynamical
systems whose future behavior for t > 0 is completely determined by their initial
state x0 at t = 0, but the history for t < 0 can not be unambigously reconstructed.
Such (noninvertible) dynamical systems are described by evolution operators defined
only for t ≥ 0 (i.e., for t ∈ R1+ or Z+ ). In the continuous-time case, they are called
semiflows.
6 1 Introduction to Dynamical Systems
It is also possible that ϕt x0 is defined only locally in time, for example, for
0 ≤ t < t0 , where t0 depends on x0 ∈ X . An important example of such a behavior
is a “blow-up,” when a continuous-time system in X = Rn approaches infinity within
a finite time, i.e.,
ϕt x0 → +∞,
for t → t0 .
The evolution operators have two natural properties that reflect the deterministic
character of the behavior of dynamical systems. First of all,
(DS.0) ϕ0 = id,
where id is the identity map on X, id (x) = x for all x ∈ X . The property (DS.0)
implies that the system does not change its state “spontaneously.” The second prop-
erty of the evolution operators reads
(DS.1) ϕt+s = ϕt ◦ ϕs .
It means that
ϕt+s x = ϕt (ϕs x)
for all x ∈ X and t, s ∈ T , such that both sides of the last equation are defined.1
Essentially, the property (DS.1) states that the result of the evolution of the system
in the course of t + s units of time, starting at a point x ∈ X , is the same as if the
system were first allowed to change from the state x over only s units of time and
then evolved over the next t units of time from the resulting state ϕs x (see Fig. 1.2).
This property means that the law governing the behavior of the system does not
change in time: The system is “autonomous.”
For invertible systems, the evolution operator ϕt satisfies the property (DS.1) for t
and s both negative and nonnegative. In such systems, the operator ϕ−t is the inverse
to ϕt , (ϕt )−1 = ϕ−t since
ϕ−t ◦ ϕt = id.
ϕ2 = ϕ1 ◦ ϕ1 = f ◦ f = f 2 ,
ϕk = f k
ϕt
xt+s
ϕs xs
x0 X
for all k > 0. If the discrete-time system is invertible, the above equation holds for
k ≤ 0, where f 0 = id.
Finally, let us point out that, for many systems, ϕt x is a continuous function of
x ∈ X , and if t ∈ R1 , it is also continuous in time. Here, the continuity is supposed
to be defined with respect to the corresponding metric or norm in X . Furthermore,
many systems defined on Rn , or on smooth manifolds in Rn , are such that ϕt x is
smooth as a function of (x, t). Such systems are called smooth dynamical systems.
ω = {. . . , ω−2 , ω−1 , ω0 , ω1 , ω2 , . . .} ∈ X
θ = {. . . , θ−2 , θ−1 , θ0 , θ1 , θ2 , . . .} ∈ X,
where
θk = ωk+1 , k ∈ Z.
The map σ merely shifts the sequence by one position to the left. It is called a shift
map. The shift map defines a discrete-time dynamical system {Z, 2 , σ k } called the
symbolic dynamics, that is invertible (find σ −1 ). Notice that two sequences, θ and ω,
are equivalent if and only if θ = σ k0 (ω) for some k0 ∈ Z. ♦
Later on in the book, we will encounter many different examples of dynamical
systems and will study them in detail.
x0
(f 2 (a), f 3 (a))
(f (a), f 2 (a))
(f 2 (a), f 2 (a))
(a, f (a))
(f (a), f (a))
(a, a), (a, f (a)), (f (a), f (a)), (f (a), f 2 (a)), (f 2 (a), f 2 (a)), . . .
by “jumping” vertically and horizontally between the graph of f and the line x̃ = x.
Now notice that the sequence obtained by taking the first coordinate of all odd points
composes the forward part
a, f (a), f 2 (a), . . . ,
. . . , a, a, a, a, . . .
correspond to those points, where the graph of f intersects the line x̃ = x: f (a) = a.
♦
Orbits are often also called trajectories. If y0 = ϕt0 x0 for some t0 , the sets Or(x0 )
and Or(y0 ) coincide. For example, two equivalent sequences θ, ω ∈ 2 generate the
10 1 Introduction to Dynamical Systems
same orbit of the symbolic dynamics {Z, 2 , σ k }. Thus, all different orbits of the
symbolic dynamics are represented by points in the set 2 introduced in Example
1.6.
The simplest orbits are equilibria.
Definition 1.3 A point x0 ∈ X is called an equilibrium (fixed point) if ϕt x0 = x0
for all t ∈ T .
The evolution operator maps an equilibrium onto itself. Equivalently, a system
placed at an equilibrium remains there forever. Thus, equilibria represent the sim-
plest mode of behavior of the system. We will reserve the name “equilibrium” for
continuous-time dynamical systems, while using the term “fixed point” for corre-
sponding objects of discrete-time systems. The system from Example 1.7 obviously
has a single equilibrium at the origin, x0 = (0, 0)T . If λ, μ < 0, all orbits converge
to x0 as t → +∞ (this is the simplest mode of asymptotic behavior for large time).
The symbolic dynamics from Example 1.7 have only two fixed points, represented
by the sequences
ω 1 = {. . . , 1, 1, 1, . . .}
and
ω 2 = {. . . , 2, 2, 2, . . .}.
Clearly, the shift σ does not change these sequences: σ(ω 1,2 ) = ω 1,2 .
Another relatively simple type of orbit is a cycle.
Definition 1.4 A cycle is a periodic orbit, namely a nonequilibrium orbit L 0 , such
that each point x0 ∈ L 0 satisfies ϕt+T0 x0 = ϕt x0 with some T0 > 0, for all t ∈ T .
The minimal T0 with this property is called the period of the cycle L 0 . If a system
starts its evolution at a point x0 on the cycle, it will return exactly to this point after
every T0 units of time. The system exhibits periodic oscillations. In the continuous-
time case a cycle L 0 is a closed curve (see Fig. 1.5(a)).
Definition 1.5 A cycle of a continuous-time dynamical system, in a neighborhood
of which there are no other cycles, is called a limit cycle.
f 2(x0)
f (x0)
x0 x0
f N −1(x0)
L0 L0
(a) (b)
Fig. 1.5 Periodic orbits in (a) a continuous-time and (b) a discrete-time system
1.3 Invariant Sets 11
If the state space X is endowed with a metric ρ, we could consider closed invari-
ant sets in X . Equilibria (fixed points) and cycles are clearly the simplest exam-
ples of closed invariant sets. There are other types of closed invariant sets. The
next more complex are invariant manifolds, that is, finite-dimensional hypersurfaces
in some space R K . Figure 1.6 sketches an invariant two-dimensional torus T2 of
a continuous-time dynamical system in R3 and a typical orbit on that manifold.
One of the major discoveries in dynamical systems theory was the recognition that
very simple, invertible, differentiable dynamical systems can have extremely com-
plex closed invariant sets containing an infinite number of periodic and nonperiodic
orbits. Smale constructed the most famous example of such a system. It provides an
invertible discrete-time dynamical system on the plane possessing an invariant set
, whose points are in one-to-one correspondence with all the bi-infinite sequences
of two symbols. The invariant set is not a manifold. Moreover, the restriction of
the system to this invariant set behaves, in a certain sense, as the symbolic dynamics
specified in Example 1.8. That is how we can verify that it has an infinite number of
cycles. Let us explore Smale’s example in some detail.
Consider the geometrical construction in Fig. 1.7.Take a square S on the plane (Fig.
1.7(a)). Contract it in the horizontal direction and expand it in the vertical direction
(Fig. 1.7(b)). Fold it in the middle (Fig. 1.7(c)) and place it so that it intersects the
original square S along two vertical strips (Fig. 1.7(d)). This procedure defines a map
f : R2 → R2 . The image f (S) of the square S under this transformation resembles
a horseshoe. That is why it is called a horseshoe map. The exact shape of the image
f (S) is irrelevant; however, let us assume for simplicity that both the contraction and
expansion are linear and that the vertical strips in the intersection are rectangles. The
map f can be made invertible and smooth together with its inverse (a diffeomorphism).
The inverse map f −1 transforms the horseshoe f (S) back into the square S through
stages (d)–(a). This inverse transformation maps the dotted square S shown in Fig.
1.7(d) into the dotted horizontal horseshoe in Fig. 1.7(a), which we assume intersects
the original square S along two horizontal rectangles.
1.3 Invariant Sets 13
B C
f B
B C
S S
A D
D CA B
f −1
A D
A D
H12
H1
H11
S ∩ f (S) = V1 ∪ V2
(see Fig. 1.8(a)). Now make the most important step: Perform the second iteration
of the map f . Under this iteration, the vertical strips V1,2 will be transformed into
two “thin horseshoes” that intersect the square S along four narrow vertical strips:
V11 , V21 , V22 , and V12 (see Fig. 1.8(b)). We write this as
H21
H22 f (H12) f (H22 )
H12 f (H21 )
f (H11)
H11
V11 V21V22 V12
f f
Similarly,
S ∩ f −1 (S) = H1 ∪ H2 ,
where H1,2 are the horizontal strips shown in Fig. 1.8(c), and
with four narrow horizontal strips Hij (Fig. 1.8(d)). Notice that f (Hi ) = Vi , i = 1, 2,
as well as f 2 (Hij ) = Vij , i, j = 1, 2 (Fig. 1.9).
Iterating the map f further, we obtain 2k vertical strips in the intersection S ∩
f (S), k = 1, 2, . . .. Similarly, iteration of f −1 gives 2k horizontal strips in the
k
intersection S ∩ f −k (S), k = 1, 2, . . ..
Most points leave the square S under iteration of f or f −1 . Forget about such
points, and instead consider a set composed of all points in the plane that remain in
the square S under all iterations of f and f −1 :
(any point x ∈ must belong to each of the involved sets). It is clear from this
representation that the set has a peculiar shape. Indeed, it should be located within
f −1 (S) ∩ S ∩ f (S),
(a) (b)
which is the union of sixteen smaller squares (Fig. 1.10(b)), and so forth. In the limit,
we obtain a Cantor (fractal) set.
Lemma 1.1 There is a one-to-one correspondence h : → 2 , between points of
and all bi-infinite sequences of two symbols.
Proof:
For any point x ∈ , define a sequence of two symbols {1, 2}
ω = {. . . , ω−2 , ω−1 , ω0 , ω1 , ω2 , . . .}
by the formula
1 if f k (x) ∈ H1 ,
ωk = (1.3)
2 if f k (x) ∈ H2 ,
for k = 0, ±1, ±2, . . .. Here, f 0 = id, the identity map. Clearly, this formula defines
a map h : → 2 , which assigns a sequence to each point of the invariant set.
To verify that this map is invertible, take a sequence ω ∈ 2 , fix m > 0, and
consider a set Rm (ω) of all points x ∈ S, not necessarily belonging to , such that
f k (x) ∈ Hωk ,
Remark:
The map h : → 2 is continuous together with its inverse (a homeomorphism)
if we use the standard metric (1.1) in S ⊂ R2 and the metric given by (1.2) in 2 . ♦
Consider now a point x ∈ and its corresponding sequence ω = h(x), where h
is the map previously constructed. Next, consider a point y = f (x), that is, the image
of x under the horseshoe map f . Since y ∈ by definition, there is a sequence that
corresponds to y : θ = h(y). Is there a relation between these sequences ω and θ?
As one can easily see from (1.3), such a relation exists and is very simple. Namely,
θk = ωk+1 , k ∈ Z,
since f k (f (x)) = f k+1 (x). In other words, the sequence θ can be obtained from the
sequence ω by the shift map σ, defined in Example 1.8:
θ = σ(ω).
f | = h−1 ◦ σ ◦ h.
Combining Lemmas 1.1 and 1.2 with obvious properties of the shift dynamics
on 2 , we get a theorem giving a rather complete description of the behavior of the
horseshoe map.
Theorem 1.1 (Smale (1963)) The horseshoe map f has a closed invariant set
that contains a countable set of periodic orbits of arbitrarily long period, and
an uncountable set of nonperiodic orbits, among which there are orbits passing
arbitrarily close to any point of .
The dynamics on have certain features of “random motion.” Indeed, for any
sequence of two symbols we generate “randomly,” thus prescribing the phase point
to visit the horizontal strips H1 and H2 in a certain order, there is an orbit showing
this feature among those composing .
The next important feature of the horseshoe example is that we can slightly perturb
the constructed map f without qualitative changes to its dynamics. Clearly, Smale’s
construction is based on a sufficiently strong contraction/expansion, combined with
a folding. Thus, a (smooth) perturbation f˜ will have similar vertical and horizontal
strips, which are no longer rectangles but curvilinear regions. However, provided the
perturbation is sufficiently small (see the next chapter for precise definitions), these
strips will shrink to curves that deviate only slightly from vertical and horizontal
lines. Thus, the construction can be carried through verbatim, and the perturbed map
1.3 Invariant Sets 17
6
2
fα,β (R)
y 0 R
−3
fα,β (R)
−6
−4 −2 0 2 4
x
Fig. 1.11 Smale horseshoe in the Hénon map
Remark:
One can precisely specify the contraction/expansion properties required by the
horseshoe map in terms of expanding and contracting cones of the Jacobian matrix
fx (see the literature cited in the bibliographical notes in Appendix D to this chapter).
♦
An equivalent map was introduced by Hénon (1976) as the simplest map with “ran-
dom dynamics.” The map fα,β is invertible if β = 0 and has the essential stretching
and folding properties of the horseshoe map in a certain parameter range.
For example, fix α = 4.5, β = 0.2, and consider the first two images fα,β (R)
2
and fα,β (R) of a rectangle R, shown in Fig. 1.11. The similarity to Fig. 1.9(c) is
obvious. ♦
18 1 Introduction to Dynamical Systems
x → f (x), x ∈ Rn ,
S0
S0 V
U U0
(a) (b)
det(A − μIn ) = 0,
x → x̃ = f (x), x ∈ R1 ,
with some 0 < λ < 1. Then the discrete-time dynamical system {Z+ , X, f k } has a
stable fixed point x0 ∈ X . Moreover, f k (x) → x0 as k → +∞, starting from any
point x ∈ X .
2 Notice that a repelling fixed point for f is an attracting fixed point for f −1 .
20 1 Introduction to Dynamical Systems
x
x
x
x
=x x
= f (x) x
= f (x)
x
=x x
=x
x
= f (x)
x0 x0 x0
x0 x x0 x x0 x
(a) (b) (c)
Fig. 1.13 Fixed points of monotone maps: attracting (a), repelling (b), and (positive) semistable (c)
The proof of this fundamental theorem can be found in any text on mathematical
analysis or differential equations. Notice that there is no restriction on the dimension
of the space X : It can be, for example, an infinite-dimensional function space. Another
important difference from Theorem 1.2 is that Theorem 1.3 guarantees the existence
and uniqueness of the fixed point x0 , whereas this has to be assumed in Theorem
1.2. Actually, the map f from Theorem 1.2 is a contraction near x0 , provided an
appropriate metric (norm) in Rn is introduced. The Contraction Mapping Principle
is a powerful tool: Using this principle, we can prove the Implicit Function Theorem,
the Inverse Function Theorem, as well as Theorem 1.4 ahead. We will apply the
Contraction Mapping Principle in Chap. 4 to prove the existence, uniqueness, and
stability of a closed invariant curve that appears under parameter variation from
a fixed point of a generic planar map. Notice also that Theorem 1.3 gives global
asymptotic stability: Any orbit of {Z+ , X, f k } converges to x0 .
Finally, let us point out that the invariant set of the horseshoe map is not stable.
However, there are invariant fractal sets that are stable. Such objects are called strange
attractors.
ẋi = fi (x1 , x2 , . . . , xn ), i = 1, 2, . . . , n,
ϕ̈ = −k 2 sin ϕ,
with g
k2 = ,
l
where l is the pendulum length, and g is the gravity acceleration constant. If we
introduce ψ = ϕ̇, so that (ϕ, ψ) represents a point in the state space X = S1 × R1 ,
the above differential equation can be rewritten in the form of equation (1.5):
ϕ̇ = ψ,
(1.6)
ψ̇ = −k 2 sin ϕ.
Here
ϕ
x= ,
ψ
while
ϕ ψ
f = .♦
ψ −k 2 sin ϕ
∂H ∂H
q̇i = , ṗi = − , (1.7)
∂pi ∂qi
II.
Updated editions will replace the previous one—the old editions will
be renamed.
1.D. The copyright laws of the place where you are located also
govern what you can do with this work. Copyright laws in most
countries are in a constant state of change. If you are outside the
United States, check the laws of your country in addition to the terms
of this agreement before downloading, copying, displaying,
performing, distributing or creating derivative works based on this
work or any other Project Gutenberg™ work. The Foundation makes
no representations concerning the copyright status of any work in
any country other than the United States.
• You pay a royalty fee of 20% of the gross profits you derive from
the use of Project Gutenberg™ works calculated using the
method you already use to calculate your applicable taxes. The
fee is owed to the owner of the Project Gutenberg™ trademark,
but he has agreed to donate royalties under this paragraph to
the Project Gutenberg Literary Archive Foundation. Royalty
payments must be paid within 60 days following each date on
which you prepare (or are legally required to prepare) your
periodic tax returns. Royalty payments should be clearly marked
as such and sent to the Project Gutenberg Literary Archive
Foundation at the address specified in Section 4, “Information
about donations to the Project Gutenberg Literary Archive
Foundation.”
• You comply with all other terms of this agreement for free
distribution of Project Gutenberg™ works.
1.F.
1.F.4. Except for the limited right of replacement or refund set forth in
paragraph 1.F.3, this work is provided to you ‘AS-IS’, WITH NO
OTHER WARRANTIES OF ANY KIND, EXPRESS OR IMPLIED,
INCLUDING BUT NOT LIMITED TO WARRANTIES OF
MERCHANTABILITY OR FITNESS FOR ANY PURPOSE.
Please check the Project Gutenberg web pages for current donation
methods and addresses. Donations are accepted in a number of
other ways including checks, online payments and credit card
donations. To donate, please visit: www.gutenberg.org/donate.
Most people start at our website which has the main PG search
facility: www.gutenberg.org.