Control of Discrete Time Descriptor

Systems Alexey A. Belov

Visit to download the full and correct content document:
https://textbookfull.com/product/control-of-discrete-time-descriptor-systems-alexey-a-
belov/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Linear Discrete-Time Systems Zoran M. Buchevats

https://textbookfull.com/product/linear-discrete-time-systems-
zoran-m-buchevats/

Stability and Control of Nonlinear Time varying Systems

Shuli Guo

https://textbookfull.com/product/stability-and-control-of-
nonlinear-time-varying-systems-shuli-guo/

Fixed Time Cooperative Control of Multi Agent Systems

Zongyu Zuo

https://textbookfull.com/product/fixed-time-cooperative-control-
of-multi-agent-systems-zongyu-zuo/

Fundamentals of Electronics 3 Discrete time Signals and

Systems and Quantized Level Systems 1st Edition Pierre
Muret

https://textbookfull.com/product/fundamentals-of-
electronics-3-discrete-time-signals-and-systems-and-quantized-
level-systems-1st-edition-pierre-muret/
Cooperative Control of Nonlinear Networked Systems
Infinite time and Finite time Design Methods Yongduan
Song

https://textbookfull.com/product/cooperative-control-of-
nonlinear-networked-systems-infinite-time-and-finite-time-design-
methods-yongduan-song/

Dynamic Systems with Time Delays Stability and Control

Ju H. Park

https://textbookfull.com/product/dynamic-systems-with-time-
delays-stability-and-control-ju-h-park/

Discrete Biochronological Time Scales 1st Edition Jean

Guex

https://textbookfull.com/product/discrete-biochronological-time-
scales-1st-edition-jean-guex/

Models of Technologies Mikhail V. Belov

https://textbookfull.com/product/models-of-technologies-mikhail-
v-belov/

Nonlinear Dynamics of Discrete and Continuous Systems

Andrei K. Abramian

https://textbookfull.com/product/nonlinear-dynamics-of-discrete-
and-continuous-systems-andrei-k-abramian/
Studies in Systems, Decision and Control 157

Alexey A. Belov · Olga G. Andrianova

Alexander P. Kurdyukov

Control of
Discrete-Time
Descriptor
Systems
An Anisotropy-Based Approach
Studies in Systems, Decision and Control

Volume 157

Series editor
Janusz Kacprzyk, Polish Academy of Sciences, Warsaw, Poland
e-mail: kacprzyk@ibspan.waw.pl
The series “Studies in Systems, Decision and Control” (SSDC) covers both new
developments and advances, as well as the state of the art, in the various areas of
broadly perceived systems, decision making and control- quickly, up to date and
with a high quality. The intent is to cover the theory, applications, and perspectives
on the state of the art and future developments relevant to systems, decision
making, control, complex processes and related areas, as embedded in the fields of
engineering, computer science, physics, economics, social and life sciences, as well
as the paradigms and methodologies behind them. The series contains monographs,
textbooks, lecture notes and edited volumes in systems, decision making and
control spanning the areas of Cyber-Physical Systems, Autonomous Systems,
Sensor Networks, Control Systems, Energy Systems, Automotive Systems,
Biological Systems, Vehicular Networking and Connected Vehicles, Aerospace
Systems, Automation, Manufacturing, Smart Grids, Nonlinear Systems, Power
Systems, Robotics, Social Systems, Economic Systems and other. Of particular
value to both the contributors and the readership are the short publication timeframe
and the world-wide distribution and exposure which enable both a wide and rapid
dissemination of research output.

More information about this series at http://www.springer.com/series/13304

Alexey A. Belov Olga G. Andrianova
•

Alexander P. Kurdyukov

Control of Discrete-Time
Descriptor Systems
An Anisotropy-Based Approach

123
Alexey A. Belov Olga G. Andrianova
Department of Control Systems Laboratory of Dynamics of Control
and Informatics Systems
ITMO University V. A. Trapeznikov Institute of Control
St. Petersburg Sciences of Russian Academy of Sciences
Russia Moscow
Russia
and
and
Laboratory of Dynamics of Control
Systems School of Applied Mathematics
V. A. Trapeznikov Institute of Control HSE Tikhonov Moscow Institute
Sciences of Russian Academy of Sciences of Electronics and Mathematics
Moscow Moscow
Russia Russia

Alexander P. Kurdyukov
Laboratory of Dynamics of Control
Systems
V. A. Trapeznikov Institute of Control
Sciences of Russian Academy of Sciences
Moscow
Russia

ISSN 2198-4182 ISSN 2198-4190 (electronic)

Studies in Systems, Decision and Control
ISBN 978-3-319-78478-6 ISBN 978-3-319-78479-3 (eBook)
https://doi.org/10.1007/978-3-319-78479-3
Library of Congress Control Number: 2018935859

Mathematics Subject Classification (2010): 93E20, 93E25, 93C05, 93C55

© Springer International Publishing AG, part of Springer Nature 2018

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt from
the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, express or implied, with respect to the material contained herein or
for any errors or omissions that may have been made. The publisher remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.

Printed on acid-free paper

This Springer imprint is published by the registered company Springer International Publishing AG
part of Springer Nature
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface

Descriptor linear systems theory is an important part of control systems theory and
has attracted much attention in the last decades. Many researchers pay a great
attention to solving analysis and control design of descriptor systems. In the last 30
years, differential-algebraic equations have become a widely accepted tool for the
modeling and simulation of constrained dynamical systems in numerous applica-
tions, such as mechanical multibody systems, electrical circuit simulation, chemical
engineering, control theory, fluid dynamics, and many other areas.
Problems of sensitivity reduction or external disturbance attenuation are
well-known in modern control theory. The mostly studied ones are LQG/H2 and
H1 control problems. In LQG/H2 optimal theory the Gaussian white noise
sequence is considered as the input disturbance. In discrete-time H1 control
approach input disturbances are considered as sequences with limited power, i.e. the
sequences are square summable. The discrete-time LQG/H2 and H1 control
problems were successfully generalized on the class of descriptor systems.
Anisotropy-based approach deals with the stationary random Gaussian signals with
known mean anisotropy level a  0, which has a sense of “spectral color” of the
signal. Similar to H2 and H1 norms, anisotropic norm defines a performance
index of the system from the input to output. The key feature of anisotropy-based
approach is that anisotropic norm of the system lies between the scaled H2 -norm
and H1 -norm. Anisotropy-based control theory allows to develop unified
theoretical framework to performance analysis and control synthesis, which covers
popular H2 - and H1 -approaches as limiting cases. From the practical point of
view additional information about the input disturbance allows to expend less
energy for control, and, at the same time, remove strong assumption that the input
disturbance is white noise sequence.
This book addresses the original research on anisotropy-based analysis and
control design theory for discrete-time descriptor systems. The book consists of
seven chapters. The first chapter illustrates a variety of practical applications of
descriptor systems. The aim of the second chapter is to provide a background
material on linear discrete-time descriptor systems. The rest part of the book

v
vi Preface

consists of authors’ results on analysis and control of discrete-time descriptor

systems in presence of colored noise.
First of all, we are grateful to a number of people who helped in reviewing and
improving of the book, especially, Alexander B. Kurzhanski, Alexander S.
Poznyak, and Boris T. Polyak for useful comments and remarks.
We also would like to acknowledge the financial support kindly provided by the
Government of the Russian Federation through ITMO Postdoctoral Fellowship
program (grant 074-U01) and the Russian Foundation for Basic Research (grant
16-38-00216, grant 17-08-00185).

St. Petersburg, Russia Alexey A. Belov

Moscow, Russia Olga G. Andrianova
Moscow, Russia Alexander P. Kurdyukov
December 2017
Contents

1 Practical Application of Descriptor Systems . . . . . . . . ........... 1

1.1 Chemistry and Biology . . . . . . . . . . . . . . . . . . . . . ........... 3
1.1.1 Isothermal Reaction in an Isothermal Batch
Reactor System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.1.2 Chemical Reaction of Urethane . . . . . . . . . . . . . . . . . . . . 6
1.1.3 Evaporator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.1.4 Multispecies Food Chain . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.2 Economic Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.3 Large-Scale Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
1.4 Constrained Mechanical Systems . . . . . . . . . . . . . . . . . . . . . . . . . 14
1.4.1 Two Connected One-Mass Oscillators . . . . . . . . . . . . . . . . 15
1.4.2 Cart Pendulum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1.4.3 Planar Crane Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
1.5 Robotics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
1.6 Electrical Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
1.7 Discretization of Partial Differential Equations . . . . . . . . . . . . . . . 29
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
2 Basics of Discrete-Time Descriptor Systems Theory . . . . . . . . . . . . . 33
2.1 Equivalent Forms of Descriptor Systems . . . . . . . . . . . . . . . . . . . 33
2.1.1 Weierstrass Canonical Form . . . . . . . . . . . . . . . . . . . . . . . 35
2.1.2 SVD Equivalent Form . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
2.1.3 Generalized Upper Triangular Equivalent Form . . . . . . . . . 39
2.2 Discrete-Time Descriptor Systems on a Finite Horizon . . . . . . . . . 39
2.3 Controllability of Discrete-Time Descriptor Systems . . . . . . . . . . . 42
2.3.1 C-Controllability (Complete Controllability) . . . . . . . . . . . 42
2.3.2 R-Controllability (Controllability in the Initial
Reachable Set) . . . . . . . . . . . . . . . . . . . . . . . . . ........ 43
2.3.3 Y-Controllability (Causal Controllability) . . . . . ........ 43
2.4 Observability of Discrete-Time Descriptor Systems . . . ........ 45

vii
viii Contents

2.5 Discrete-Time Descriptor Systems on an Infinite Horizon . . . . . . . 46

2.5.1 Time Domain Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 47
2.5.2 Frequency Domain Analysis . . . . . . . . . . . . . . . . . . . . . . . 51
2.5.3 Transfer Functions and Realizations . . . . . . . . . . . . . . . . . 52
2.5.4 Impulse and Frequency Characteristics . . . . . . . . . . . . . . . 53
2.5.5 Controllability and Observability Gramians . . . . . . . . . . . . 54
2.6 Stability of Discrete-Time Descriptor Systems . . . . . . . . . . . . . . . 56
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
3 Anisotropy-Based Analysis of LDTI Descriptor Systems . . ........ 61
3.1 Preliminaries of Anisotropy-Based Control Theory . . . . ........ 61
3.1.1 Anisotropy of the Random Vector and Mean
Anisotropy of the Signal . . . . . . . . . . . . . . . . . . . . . . . . . 61
3.2 System Norms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
3.2.1 Lpm2 - and H2 -Norms . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
pm
3.2.2 L1 - and H1 -Norms . . . . . . . . . . . . . . . . . . . . . . . . . . 67
3.2.3 a-Anisotropic Norm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
3.3 Anisotropy-Based Performance Analysis . . . . . . . . . . . . . . . . . . . 74
3.3.1 Riccati Equations Technique . . . . . . . . . . . . . . . . . . . . . . 75
3.3.2 Convex Optimization Approach . . . . . . . . . . . . . . . . . . . . 83
3.3.3 Novel Anisotropy-Based Bounded Real Lemma:
Strict Conditions . . . . . . . . . . . . . . . . . . . . . . . ........ 89
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ........ 96
4 Optimal Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
4.1 State Feedback Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
4.2 Output Feedback Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
5 Suboptimal Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
5.1 GDARI Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
5.2 GDARE Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
5.2.1 State Feedback Control . . . . . . . . . . . . . . . . . . . . . . . . . . 120
5.2.2 Full Information Control . . . . . . . . . . . . . . . . . . . . . . . . . 123
5.3 Convex Optimization Technique . . . . . . . . . . . . . . . . . . . . . . . . . 126
5.4 Transient Response Shaping for Closed-Loop Systems . . . . . . . . . 129
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
6 Anisotropy-Based Analysis for LDTI Descriptor Systems
with Nonzero-Mean Input Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
6.1 Mean Anisotropy of the Gaussian Sequence with Nonzero
Mean . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
6.2 Anisotropic Norm of Descriptor Systems
with Nonzero-Mean Input Signals . . . . . . . . . . . . . . . . . . . . . . . . 140
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
Contents ix

7 Robust Anisotropy-Based Control . . . . . . . . . . . . . . . . . . . . . . . . . . . 149

7.1 Problem Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
7.2 Anisotropy-Based Analysis for Uncertain Descriptor
Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
7.3 State-Space Anisotropy-Based Robust Control Design
for Uncertain Descriptor Systems . . . . . . . . . . . . . . . . . . . . . . . . 156
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161

Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
Notations and Abbreviations

Symbols and Notations

Z Set of all integers

R Set of all real numbers
Rn Set of all real vectors of dimension n
Rmn Set of all real matrices of dimension m  n
C Set of all complex numbers
degðf ðxÞÞ Degree of the polynomial, degðf ðxÞÞ ¼ n if
f ðxÞ ¼ an xn þ an1 xn1 þ . . . þ a0 , an 6¼ 0
det ðAÞ Determinant of matrix A
rankðAÞ Rank of matrix A; if A is square, then rankðAÞ is equal to the number
of nonzero eigenvalues of matrix P A P
TrðAÞ Trace of matrix A: TrðAÞ ¼ j ajj ¼ j kj ðAÞ
AT Transpose of matrix A: AT ¼ ðaij ÞT ¼ ðaji Þ
In Identity matrix of order n  n
A-1 Inverse matrix of matrix A: AA1 ¼ A1 A ¼ I
KerðWÞ Kernel of linear mapping W: KerðWÞ ¼ fx : x 2 X; WðxÞ ¼ 0g
JmðWÞ Image of linear mapping W: =mðWÞ ¼ fy : y ¼ WðxÞ; 8x 2 V g
spanðWÞ Linear span of linear mapping W
diagðAi Þ Diagonal ðmnÞ  ðmnÞ-matrix with elements Ai of dimension m  m
on the main diagonal, i ¼ 1; n
 Direct sum
F  ðzÞ T
Transposed complex conjugate of FðzÞ: F  ðzÞ ¼ F ðzÞ
b
GðxÞ b
Boundary value of transfer function GðzÞ: GðxÞ ¼ limr!10 Gðreix Þ
Rek Real part of a complex number k
k G k2 H2 -norm of transfer function G
kG k1 H1 -norm of transfer function G
kY k P Power norm of the signal Y
kjGjka a-Anisotropic norm of transfer function G

xi
xii Notations and Abbreviations

kM k Frobenius norm of matrix: ðTr ðM  MÞÞ1=2

EðWÞ Mathematical expectation of a random vector W
kðAÞ Range of matrix A: kðAÞ ¼ fki : detðki I  AÞ ¼ 0; i ¼ 1; ng
qðAÞ Spectral radius of matrix A: qðAÞ ¼ maxi jkðAÞj
Aþ Pseudo-inverse (Moore-Penrose inverse) matrix of matrix A, that
satisfies the equations AA þ A ¼ A, A þ AA þ ¼ A þ , ðAA þ ÞT ¼ AA þ ,
ðA þ AÞT ¼ A þ A
fuðkÞgk2Z Numeric sequence in the form
ð. . .; uðkÞ; . . .; uð1Þ; uð0Þ; uð1Þ; . . .; uðkÞ; . . .Þ
DðPjjM Þ Relative entropy (Kullback-Leibler information divergence) from
probability measure P to probability measure M
AðwÞ Anisotropy of a random vector w 2 Rm
AðW Þ Mean anisotropy of a random sequence W ¼ fwðkÞgk2Z
AðGÞ Mean anisotropy of a random sequence W, generated from the
Gaussian white noise sequence V by the shaping filter G
Z Hermitian conjugate of the matrix Z ¼ ½zij  2 Cmn : Z  ¼ ½zji  2 Cnm
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
rðAÞ Maximal singular value of the matrix A: rðAÞ ¼ qðA AÞ
symðAÞ Symmetrization of matrix A: sym ðAÞ ¼ A þ AT
ðZÞ½fðkÞ Laplace z-transform of a sequence ff ðkÞgk2Z
r Laplace operator r ¼ @x @ @
i þ @y @
j þ @z k

Abbreviations

BRL Bounded real lemma

DAE Differential-algebraic equations
FI Full information
GDARE Generalized discrete-time algebraic Riccati equation
GDARI Generalized discrete-time algebraic Riccati inequality
LDTI Linear discrete-time time-invariant (systems)
LFT Linear fractional transformation
LMI Linear matrix inequality
MOL Method of lines
ODE Ordinary differential equations
PDE Partial differential equations
PDF Probability density function
RMS Root mean-square (gain)
SF State feedback
SVD Singular value decomposition
Introduction

A process has to have qualitative and quantitative description as prerequisite for

solving control problems. The most common way of description is mathematical
model of control object or the workflow.
Mathematical models of control systems are designed based on well-known laws
of nature: physical, chemical, biological laws, etc.
These laws are usually described by differential and algebraic equations based on
laws like Newton’s second law or Kirchhoff’s law. Plant models are normally done
by differential or difference equations. However, in many cases, such description is
not sufficient.
Attempts to represent systems only by differential or difference equations might
result in a loss of relevant information and arrive to a description of the system in
abstract variables, so called phase variables. It can create problems for practical
realizations of controllers and diminish the quality of real object’s control.
During design of mathematical model with real physical values, a designer has to
account for the fact that system description could contain not only differential
equations, but also algebraic constraints and relations.
As a rule, a system of algebraic-differential equations, describing control system,
cannot be solved for the first derivative. This situation explains a new class of
systems, called algebraic-differential systems. Algebraic-differential systems [1] are
also called singular systems [2–4], generalized state-space systems (or generalized
systems) [5], implicit systems [6, 7] or descriptor systems [8, 9]. The origin of
descriptor systems theory dates back to the works of P. Dirac on generalized
Hamiltonian systems [10–12]. In modern science the main idea discussed in these
works, is called the differentiation index of semi-explicit descriptor systems.
A geometric method of studying the so-called constrained systems covered in
the works of Dirac. This method has found its application in mechanics [13–21].
Mechanical systems, represented as descriptor systems, have become a subject
of extensive research [22–24].
Further development of the theory on parametrized sets of bilinear form can be
found in works of K. Weierstrass and L. Kronecker [25, 26]. F. R. Gantmacher used

xiii
xiv Introduction

matrix pencils [27] to analyze matrices of linear normal systems with possible
degeneracies of the main matrix coefficient.
A large number of works on descriptor systems also applies to the theory of
electrical circuits. Presence of differential and algebraic equations in such systems
involves a combination of differential equations describing the behavior of reactive
elements, and algebraic relations, based on Kirchhoff’s laws and characteristics of
elements [28–31].
Mathematical theory of differential-algebraic systems began to develop in the
1970s independently in various fields of technology. We can mention the works of
Gear [32], Takens [33], as well as monographs by Campbell [2, 3] and Petzold [34],
released in the early 1980s. In these works, the main attention was focused on
numerical aspects of descriptor systems modeling. Currently, much attention is paid
to differential-algebraic systems in partial derivatives [35–37] and stochastic
descriptor systems [38].
Descriptor systems have found their application in modeling the motion of
aircrafts [39], chemical processes [40], circuit technique [8, 9], economic systems
[41], description of interconnected systems of high order [42], technical systems
[43], energy systems [44] and robotics [45].
Descriptor systems have some specific differences from systems described
exclusively by differential or difference equations, which we will call normal
systems.
Descriptor systems are characterized by the following properties [46, 47].
• The transfer function of a descriptor system may not necessarily be strictly
proper.
• For arbitrary bounded initial conditions generalized functions:
– can be in the solutions of algebraic-differential equations (impulsive
behavior),
– can depend on future for algebraic-difference equations (noncausal
behavior).
• A solution of a linear algebraic-differential equation typically contains three
components:
– limited dynamic components, corresponding to the differential equations;
– non-dynamic components, corresponding to the algebraic equations;
– unlimited dynamic components from the set of generalized functions, the
presence of which depends on the smoothness of the input signal and on
initial conditions.
Study of descriptor systems is promising from the fundamental research point of
view as well as for practical applications. Significant differences of descriptor
systems from normal systems demanded development and generalization of
mathematical tools.
Introduction xv

A significant number of fundamental concepts and results for nominal systems

have been successfully generalized for descriptor systems:
• solvability of algebraic-differential equations, study of controllability and
observability [4];
• canonical, equivalent forms and representations of descriptor systems [48–50];
• minimal realizations [52–54];
• equivalence of systems [4, 47, 55, 56];
• regularity and regularization [57–62];
• stability and stabilization [63–66];
• modal control [4, 67–70];
• linear-quadratic optimal control [46, 71, 72];
• design of observers and filtering [73–77];
• Lyapunov’s theorems and equations [78–81];
• model reduction [82, 83];
• H2 and H1 control [66, 84–87].
Let us consider for a moment the tasks of LQG/H2 and H1 control. In this
case, the control system is designed, assuming that some external disturbance is
influencing the system.
Theory of design for linear-quadratic Gaussian controllers appeared at the end of
50s in 20th century and it is associated with the name of R. Kalman.
This theory provided a powerful tool for multidimensional control systems
design with quadratic quality criterion [88].
The algorithm of control was designed with the assumption that systems are
under disturbances in the form of Gaussian white noise.
This assumption reduced the design problem to the problem of minimizing the
quality criterion, that is quadratic for control and state.
This problem can be reduced to the problem of H2 -optimization, where H2 -
norm of the system’s transfer function supports the quality criterion.
The most significant disadvantage of this approach is loss of stability of the
system under a small perturbation in model description. This deficiency was
investigated in [89].
For the closed-loop system, the design problem for stabilizing controllers that
minimize H1 -norm of the transfer function was stated and solved in [90], this
problem got its further development in works [91–95].
Such a problem belongs to optimal control problems, where H1 -norm of the
transfer function is the quality criterion for the closed-loop system. Another
important factor of the resulting control law in real life application is the degree of
conservatism, it stands for the energy cost actuators of the control object, imple-
menting the law. It is known that H2 -controllers are not robust against the intensity
of input disturbance [89], while H1 -controllers are too conservative.
Among approaches allowing reduction of conservatism of controllers, is the
approach in which the system is subjected to random disturbances with imprecisely
known probability characteristics.
xvi Introduction

Additional information about the input disturbance allows to expend less energy
for control, and, at the same time, remove the strong assumption that the input
disturbance is white noise sequence.
This concept is related to an application of information-theoretical quality cri-
teria, and it is called stochastic H1 -optimization.
One of such information criteria is a stochastic norm of the closed-loop system.
Stochastic norm is induced by power norm of random signals that belong to specific
class of probability distributions. Anisotropic norm is a special case of a stochastic
norm. This norm is used when input disturbance is a Gaussian random sequence
with zero mean and bounded mean anisotropy [96, 97]. The latter is a measure of
correlation for random vector components in a sequence or, in other words, a
measure of random sequence deviation from Gaussian white noise sequence also
known as “spectral color”.
Minimization of anisotropic norm of the transfer function for a closed-loop
system in anisotropy-based controllers was first stated in [97] and solved in [98].
Paper [99] shows that a-anisotropic norm as a function of its parameter a  0, has
H2 - and H1 -norms as its limiting cases. This implies that design problem of
anisotropy-based controllers includes classical problems of H2 - and H1 -optimi-
zation as limiting cases.
[100] shows that H2 - and H1 -controllers are limiting cases of
anisotropy-based controller. This monograph introduces reader to the solutions
of the anisotropy-based analysis and design for linear stationary descriptor systems.
The content of the book: Chapter 1 deals extensively with practical applications
of descriptor systems. Chapter 2 is dedicated to theory basics of discrete-time
descriptor systems. Chapter 3 is dedicated to anisotropy-based performance anal-
ysis using Riccati and convex optimization techniques. Chapter 4 deals with
optimal anisotropy-based control for descriptor systems. Chapter 5 deals with
suboptimal anisotropy-based control for descriptor systems. Chapter 6 develops
anisotropy-based performance analysis with nonzero mean input sequences.
Chapter 7 presents anisotropy-based analysis and robust control problems for
uncertain descriptor systems.

References

1. Byrne, G.D., Ponzi, P.R.: Differential-algebraic systems, their applications and solutions.
Comput. Chem. Eng. 12, 377–382 (1988)
2. Campbell, S.L.: Singular Systems of Differential Equations. Pitman (1980)
3. Campbell, S.L.: Singular Systems of Differential Equations II. Pitman (1982)
4. Dai, L.: Singular Control Systems, Lecture Notes in Control and Information Sciences.
Springer, New York (1989)
5. Beellena, T., Vandooren, P.: A numerical-method for deadbeat control of generalized
state-space systems. Syst. Control Lett., 10, 225–233 (1988)
6. Aplevich, J.D.: Implicit Linear Systems. Springer, Berlin (1991)
Introduction xvii

7. Banaszuk, A., Kociecki, M., Lewis, F.L.: Kalman decomposition for implicit linear systems.
IEEE Trans. Autom. Control. 37, 1509–1514 (1992)
8. Newcomb, R.W.: The semistate description of nonlinear time-variable circuits. IEEE Trans.
Circuits Syst. 2(8), 62–71 (1981)
9. Newcomb, R.W., Dziurla, B.: Some circuits and systems applications of semistate theory.
Circuit, Syst. Sig. Process. 8, 235–260 (1989)
10. Dirac, P.A.M.: Generalized Hamiltonian dynamics. Can. J. Math. 2, 129–148, (1950)
11. Dirac, P.A.M.: Generalized Hamiltonian dynamics. Proc. Royal Soc. London. 246, 326–332
(1958)
12. Dirac, P.A.M.: Lectures on Quantum Mechanics. Yeshiva University (1964); Dover (2001)
13. Barone, F., Grassini, R., Mendella, G.: A unified approach to constrained mechanical
systems as implicit differential equations. Ann. Inst. Henri Poincar, Phys. Theor., 70, 515–
546 (1999)
14. Gràcia, X., Pons, J.M.: Constrained systems: a unified geometric approach. Intl. J. Theor.
Phys. 30, 511–516 (1991)
15. Gràcia, X., Pons, J.M.: A generalized geometric framework for constrained systems. Diff.
Geom. Appl. 2, 223–247 (1992)
16. Marmo, G., Mendella, G., Tulczyjew, W.M.: Symmetries and constants of the motion for
dynamics in implicit form. Ann. Inst. Henri Poincar, Phys. Theor. 57, 147–166 (1992)
17. Marmo, G., Mendella, G., Tulczyjew, W.M.: Constrained Hamiltonian systems as implicit
differential equations. J. Phys. A: Math. Gen. 30, 277–293 (1997)
18. Mendella, G., Marmo, G., Tulczyjew, W.M.: Integrability of implicit differential equations.
J. Phys. A: Math. Gen. 28, 149–163 (1995)
19. Seiler, W.M.: Involution and constrained dynamics I: the Dirac approach. J. Phys. A: Math.
Gen. 28, 4431–4451 (1995)
20. Seiler, W.M.: Numerical analysis of constrained Hamiltonian systems and the formal theory
of differential equations. Math. Comp. Simul. 45, 561–576 (1998)
21. Seiler, W.M.: Numerical integration of constrained Hamiltonian systems using Dirac
brackets. Math. Comp. 68, 661–681 (1999)
22. Gèradin, M., Carona, A.: Flexible Multibody Dynamics: A Finite Element Approach. Wiley
(2001)
23. Hairer, E., Wanner, G.: Solving Ordinary Differential Equations II: Stiff and
Differential-Algebraic Problems. Springer (1996)
24. Leimkuhler, B., Reich, S.: Simulating Hamiltonian Dynamics. Cambridge University Press
(2004)
25. Kronecker, L.: Algebraische reduction der schaaren bilinearer formen. Sitzungsberichte
Akad. Wiss. Berlin. 1225–1237 (1890); Leopold Kronecker’s Werke, Chelsea. 139–155
(1968)
26. Weierstrass, K.: Zur Theorie der bilinearen und quadratischen Formen. Monatsberichte
Akad. Wiss. Berlin. 310–338 (1868); Mathematische Werke, II, Georg Olms Verlag and
Johnson Reprint Co. 19–44 (1967)
27. Gantmacher, F.R.: Matrix Theory. Chelsea Publishing (1959)
28. Dziurla B., Newcomb R.: The Drazin inverse and semi-state equations. Proc. Intl.
Symp. Math. Theory of Networks and Systems. 283–289 (1979)
29. Riaza, R., Torres-Ramirez, J.: Nonlinear circuit modeling via nodal methods. Internat.
J. Circuit Theory Appl. 33, 281–305 (2005)
30. Riaza, R.: Differential-algebraic Systems: Analytical Aspects and Circuit Applications.
World Scientific (2008)
31. Steinbrecher, A., Stykel, T.: Model order reduction of nonlinear circuit equations. Internat.
J. Circuit Theory Appl. 41, 1226–1247 (2013)
32. Gear, C.W.: The simultaneous numerical solution of differential-algebraic equations. IEEE
Trans. Circuit Theory. 18, 89–95 (1971)
xviii Introduction

33. Takens, F.: Constrained equations; a study of implicit differential equations and their
discontinuous solutions. Lect. Notes Maths, Springer. 525, 143–234 (1976)
34. Petzold, L.R.: Differential/Algebraic equations are not ODE’s. SIAM J. Sci. Slat, Camp. 3,
367–384 (1982)
35. Bodestedt, M., Tischendorf, C.: PDAE models of integrated circuits and index analysis.
Math. Comput. Model. Dyn. Syst. 13, 1–17 (2007)
36. Marszalek, W.: Analysis of Partial Differential Algebraic Equations. Ph.D. Thesis, North
Carolina State University (1997)
37. Reis, T.,Tischendorf, C.: Frequency domain methods and decoupling of linear infinite
dimensional differential algebraic systems. J. Evol. Equ. 5, 357–385 (2005)
38. Schein, O.: Stochastic Differential-Algebraic Equations in Circuit Simulation. Shaker Verlag
(1999)
39. Stevens, B.L., Lewis, F.L.: Aircraft Modeling. Dynamics and Control. Wiley, New York
(1991)
40. Kumar, A., Daoutidis, P.: Feedback control of nonlinear differential-algebraic equation
systems. AIChE J. 41, 619–636 (1995)
41. Luenberger, D.G., Arbel, A.: Singular dynamic Leontief systems. Econometrica. 45, 991–
995 (1977)
42. Luenberger, D.G.: Dynamic equations in descriptor form. IEEE Trans. Automat. Control. 22,
312–321 (1977)
43. Hemami, H., Wyman, B. F.: Modeling and control of constrained dynamic systems with
application to biped locomotion in the frontal plane. IEEE Trans. Automat. Control. 24, 526–
535 (1979)
44. Scott, B.: Power system dynamic response calculations. Proc. IEEE. 67, 219–247 (1979)
45. Mills, J.K., Goldenberg, A.A.: Force and position control of manipulators during constrained
motion tasks. IEEE Trans. Robot. Automat. 5, 30–46 (1989)
46. Bender, D.J., Laub, A.J.: The linear-quadratic optimal regulator for descriptor system. IEEE
Trans. Automat. Control. 32, 672–687 (1987)
47. Verghese, G.C., Levy, B.C., Kailath, T.: A generalized state-space for singular systems.
IEEE Trans. Automat. Control. 26(4), 811–831 (1981)
48. Campbell, S.L.: One canonical form for higher index linear time-varying singular systems.
Circ. Sys. Signal Process. 2, 311–326 (1983)
49. Glusingluerssen, H.: Feedback canonical form for singular systems. Int.J. Control, 52, 347–
376 (1990)
50. Glusingluerssen, H., Hinrichsen, D.: A Jordan control canonical form for singular systems.
Int.J.Control. 48, 1769–1785 (1988)
51. Lebretand, G., Loiseau, J.J.: Proportional and proportional-derivative canonical-forms for
descriptor systems with outputs. Automatica. 30, 847–864 (1994)
52. Conte, G., Perdon, A.: Generalized state-space realizations of non-proper rational
transfer-functions. Syst. Control Lett. 1, 270–276 (1982)
53. Fang, C.-H., Chang, F.-R.: A strongly observable and controllable realization of descriptor
systems. Control Theory Adv. Tech., 6, 133–141 (1990)
54. Grimm, J.: Realization and canonicity for implicit systems. SIAM J. Control Optim. 26,
1331–1347 (1988)
55. Hayton, G., Fretwell, P., Pugh, A. Fundamental equivalence of generalized state-space
systems. IEEE Trans. Automat. Control. 31, 431–439 (1986)
56. Stykel, T.: Analysis and numerical solution of generalized Lyapunov equations. PhD. Thesis,
Institut fur Mathematik, Techische Universitat Berlin, Berlin (2002)
57. Bunse-Genster, A., Byers, R., Mehrmann, V., Nichols, N.K.: Feedback design for
regularizing descriptor systems. Linear Algebra Appl. 299, 119–151 (1999)
58. Bunse-Gerstner, A., Mehrmann, V., Nichols, N.K.: Regularization of descriptor systems by
derivative and proportional state feedback. SIAM J. Matrix Anal. Appl. 13, 46–67 (1992)
Introduction xix

59. Bunse-Gerstner, A., Mehrmann, V., Nichols, N.K.: Regularization of descriptor systems by
output feedback. IEEE Trans. Automat. Control. 39, 1742–1748 (1994)
60. Byers, R., Kunkel, P., Mehrmann, V.: Regularization of linear descriptor systems with
variable coefficients. SIAM J. Control Optim. 35, 117–133 (1997)
61. Chu, D.L., Chan, H.C., Ho, D.W.C.: Regularization of singular systems by derivative and
proportional output feedback. SIAM J. Matrix Anal. Appl. 19, 21–38 (1998)
62. Chu, D.L., Ho, D.W.C.: Necessary and sufficient conditions for the output feedback
regularization of descriptor systems. IEEE Trans. Automat. Control. 44, 405–412 (1999)
63. Geerts T.: Stability concepts for general continuous-time implicit systems—definitions,
hautus tests and Lyapunov criteria. Int. J. Systems Sci. 26, 481–498 (1995)
64. Lee, L., Chen, J.L.: Strictly positive real lemma and absolute stability for discrete-time
descriptor systems. IEEE Trans. Circuits Syst. I. 50, 788–794 (2003)
65. Pandolfi, L.: Controllability and stabilization for linear systems of algebraic and differential
equations. J. Optim. Theory Appl. 30, 601–620 (1980)
66. Xu, S., Lam, J.: Robust Control and Filtering of Singular Systems, Lecture Notes in Control
and Information Sciences. Springer, Berlin (2006)
67. Ailon, A.: On the design of output-feedback for finite and infinite pole assignment in singular
systems with application to the control problem of constrained robots. Circuit Syst Sig.
Process. 13, 525–544 (1994)
68. Goodwin, M.S.: Exact pole assignment with regularity by output feedback in descriptor
systems. II. Int. J. Control. 62, 413–441 (1995)
69. Goodwin, M.S., Fletcher, L.R.: Exact pole assignment with regularity by output feedback in
descriptor systems. I. Int. J. Control. 62, 379–411 (1995)
70. Zhang, S.Y.: Pole placement for singular systems. Syst & Control Lett. 12, 339–342 (1989)
71. Kunkel, P., Mehrmann, V.: The linear quadratic optimal control problem for linear descriptor
systems with variable coefficients. Maths. Contr. Sig. Sys. 10, 247–264 (1997)
72. Pandolfi, L.: On the regulator problem for linear degenerate control systems. J. Optim.
Theory Appl. 33, 241–254 (1981)
73. Dai, L.: Observers for discrete singular systems. IEEE Trans. Automat. Control. 33,
187–191, (1988)
74. Dai, L.: Filtering and LQG problems for discrete-time stochastic singular systems. IEEE
Trans. Automat. Control. 34, 1105–1108 (1989)
75. Minamide, N., Fujisaki, Y., Shimizu, A.: A parametrization of all observers for descriptor
systems. Int. J. Control. 66, 767–777 (1997)
76. Paraskevopoulos, P.N., Koumboulis, F.N.: Unifying approach to observers for regular and
singular systems. IEEE Proc.-Control Theory Appl. 138, 561–572 (1991)
77. Paraskevopoulos, P.N., Koumboulis, F.N.: Observers for singular systems. IEEE Trans.
Automat. Control. 37, 1211–1215 (1992)
78. Bender, D.J.: Lyapunov-like equations and reachability observability gramians for descriptor
systems. IEEE Trans. Automat. Control. 32, 343–348 (1987)
79. Stykel, T.: Stability and inertia theorems for generalized Lyapunov equations. Linear
Algebra Appl. 355(1–3), 297–314 (2002)
80. Takaba, K., Morihira, N., Katayama, T.: A generalized Lyapunov theorem for descriptor
system. Syst & Control Lett. 24, 49–51 (1995)
81. Zhang, G., Zhang, Q.L., Chen, T., Lin, Y.: On Lyapunov theorems for descriptor systems.
Discrete Contin. Dyn. Syst.-Ser. B. 10, 709–725 (2003)
82. Stykel, T.: Gramian-based model reduction for descriptor systems. Math. Control Signals
Syst. 16, 297–319 (2004)
83. Zhang, L., Lam, J., Zhang, Q.L.: Optimal model reduction of discrete-time descriptor
systems. Int. J. Systems Sci. 32, 575–583 (2001)
84. Takaba, K., Katayama, T.: H2 output feedback control for descriptor systems. Syst &
Control Lett. 34, 841–850 (1988)
xx Introduction

85. Takaba, K., Katayama, T.: Discrete-time H1 algebraic Riccati equation and parametrization
of all H1 filters. Int. J. Control. 64, 1129–1149 (1996)
86. Takaba, K., Morihira, N., Katayama, T.: H1 control for descriptor systems—a J-spectral
factorization approach. Proc. 33rd IEEE Conf. Decision and Control, Lake Buena Vista, FL,
USA. 2251–2256 (1994)
87. Wang, H.-S., Yung, C.-F., Chang, F.-R.: H1 Control for Nonlinear Descriptor Systems.
Springer, London (2006)
88. Kwakernaak, H., Sivan, R.: Linear Optimal Control Systems. Wiley, New York (1972)
89. Doyle, J.C.: Guaranteed margins for LQG regulators. IEEE Trans. on Autom. Control. 23,
756–757 (1978)
90. Zames, G.: Feedback minimax sensitivity and optimal robustness. IEEE Trans. on Autom.
Control., 28, 585–601 (1983)
91. Doyle, J.C., Glover, K., Khargonekar, P.P., Francis, B.A. State-space solutions to standard
H2 and H1 -control problems. IEEE Trans. on Autom. Control. 34, 831–848 (1989)
92. Francis, B.A.: A Course in H1 -Control Theory. Lecture notes in control and information
sciences. 88, Springer, New York (1987)
93. Gu, D.-W., Tsai, M.C., O’Young, S.D., Postlethwaite, I.: State-space formulae for
discrete-time H1 -optimization. Int. J. Control. 49, 1683–1723 (1989)
94. Iglesia,s P.A., Glover, K.: State-space approach to discrete-time H1 -control. Int. J. Control.
54, 1031–1073 (1991)
95. Yaesh, I., Shaked, U.: A transfer function approach to the problems of discrete-time systems:
H1 -optimal linear control and filtering. IEEE Trans. on Autom. Control. 36, 1264–1271
(1991)
96. Semyonov, A., Vladimirov, I., Kurdyukov, A.: Stochastic approach to H1 -optimization.
Proc. 33rd IEEE Conf. Decision and Control, Florida. 3, 2249–2250 (1994)
97. Vladimirov, I.G., Kurdyukov, A.P., Semyonov, A.V.: A stochastic problem of H1 -
optimization, Doklady Math. 52:1, 155–157 (1995)
98. Vladimirov, I.G., Kurdyukov, A.P., Semyonov, A.V.: State-space solution to
anisotropy-based stochastic H1 -optimization problem. Proc. 13th IFAC World Congress,
San-Francisco, USA. 427–432 (1996)
99. Vladimirov, I.G., Kurdyukov, A.P., Semyonov, A.V.: Asymptotics of the anisotropic norm
of linear time-independent systems. Autom. Remote Control. 60:3, 359–366 (1999)
100. Kurdyukov, A., Maximov, E.: Solution of the stochastic H1 -optimization problem for
discrete-time linear systems under parametric uncertainty. Autom Remote Control. 67, 1283–
1310 (2003)
Chapter 1
Practical Application of Descriptor
Systems

Mathematical Modeling Using Descriptor Systems

Systems whose variables describe some physical processes are called descriptor
systems. Let’s consider the following simple example of such systems.
The object of mass m shown in Fig. 1.1 has position x(t) and speed v(t). It is
driven by force F(t). Equations describing the system have the form:

ẋ(t) = v(t), (1.1)

mv̇(t) = F(t). (1.2)

Introducing notations x(t) = ξ1 (t) and v(t) = ξ2 (t), these equations can be written
in a state-space representation as
       
1 0 ξ̇1 (t) 01 ξ̇1 (t) 0
= + F(t). (1.3)
0m ξ̇2 (t) 0 0 ξ̇2 (t) 1

System (1.3) is descriptor because its variables describe physical processes. Sys-
tem (1.3) is nonsingular and can easily be rewritten in the form of the normal system.
However, there are situations when additional restrictions affect the physical sys-
tem. Then a mathematical model of the plant or process includes not only differential
but algebraic equations as well.
In general, a descriptor system is given by the relation:

F(ẋ(t), x(t), t) = 0, (1.4)

the output equation is given as

© Springer International Publishing AG, part of Springer Nature 2018 1

A. A. Belov et al., Control of Discrete-Time Descriptor Systems, Studies
in Systems, Decision and Control 157, https://doi.org/10.1007/978-3-319-78479-3_1
2 1 Practical Application of Descriptor Systems

Fig. 1.1 Body moved by

force F(t).

h(x(t), t) = 0. (1.5)

Before Gear’s works appeared [1], descriptor systems in the form (1.4) had usually
been rewritten by means of analytical transformations in ODE form:

ẏ(t) = g(y(t), t). (1.6)

This made it possible to reduce the dimension of the original system; on the other
hand, that representation required solving complex algebraic equations that in the
general case had no analytical solutions for high-order systems.
Another possible way to get rid of algebraic equations was their differentiation in
order to obtain an ordinary differential equation with the same number of variables
as for the original system. The approach described above is time-consuming because
of the need to use the implicit function theorem.
However, due to a possible change of basis, obtained state variables would have no
physical meaning (i.e., they were abstract phase variables). Moreover, as a result of
the numerical integration of the resulting system of ordinary differential equations,
the solution could exceed the area limits defined by algebraic equations. Descriptor
systems given in so-called special forms are the most studied at present. Take a closer
look. Suppose that the relation in system (1.4) could be extracted with respect to the
derivative ẋ(t). Then descriptor system (1.4) and (1.5) is

E(x(t))ẋ(t) = 
F(x(t), u(t), t), (1.7)
y(t) = H (x(t), u(t), t).
 
∂E(x(t))
Assuming that rank ∂x
= const, we can rewrite (1.7) in the form
1 Practical Application of Descriptor Systems 3

ẋ1 (t) = f1 (x1 (t), x2 (t)),

0 = f2 (x1 (t), x2 (t)).

This form is called semi-explicit.

A linear stationary system is also one of the special descriptions of descriptor
systems and has the form:

E ẋ(t) = Ax(t) + Bu(t),

y(t) = Cx(t) + Du(t).

As mentioned above, algebraic relations between variables do not allow us to

solve the original equations with respect to the derivative. For example, in the case
of linear stationary systems matrix E is singular; that is, rank (E) < n. This fact does
not allow inverting matrix E and transforming the descriptor system to the system
of ordinary differential equations.
Here we consider some practical applications related to the construction of math-
ematical models of processes and control plants in descriptor form.

1.1 Chemistry and Biology

Descriptor systems found their wide application in the fields of chemistry and biology.
This is due to laws such as the transfer of mass and energy. We start a review of
mathematical models with an isothermal reactor.

1.1.1 Isothermal Reaction in an Isothermal Batch Reactor

System

A kinetic model describing the chemical reaction in an isothermal batch reactor

system is obtained in [2]. The reaction occurs in an anhydrous, homogeneous, liquid
phase catalyzed by a completely dissociated species.
The desired reaction is given by

HA + 2BM → AB + MBMH

where AB is the desired product.

The reaction is initiated by adding the catalyst QM to the batch reactor containing
two miscible reactants with reactant BM in excess.
The catalyst QM is initially assumed to be 100% dissociated to Q+ and M − ions.
The following mechanism is proposed to describe the reaction.
Slow Kinetic Reactions
4 1 Practical Application of Descriptor Systems

M − + BM ⇐k−1 k1 ⇒ MBM − ,
A− + BM k2 ⇒ ABM − ,
M + AM ⇐k−3 k3 ⇒ ABM − .
−

Rapid Acid-Base Reactions

MBMH ⇐ K1 ⇒ MBM − + H + ,
HA ⇐ K2 ⇒ A− + H + ,
HABM ⇐ K3 ⇒ ABM − + H + .

In order to derive a model to account for these reactions, it is necessary to distinguish

between the overall concentration of species and concentration of its neutral form.
Overall concentrations are defined for three components based on neutral and ionic
species.
[MBMH ] = [(MBMH )N ] + [MBM − ],

[HA] = [(HA)N ] + [A− ],

[HABM ] = [(HABM )N ] + [ABM − ],

where [ ] denotes concentration of the species in gmol/kg. By assuming the rapid

acid-base reactions are at equilibrium, the equilibrium constants K1 , K2 , K3 can be
defined as follows.
[MBM − ][H + ]
K1 = ,
[(MBMH )N ]

[A− ][H + ]
K2 = ,
[(HA)N ]

[ABM − ][H + ]
K3 = .
[(HABM )N ]

Anionic species may then be represented by

K1 [MBMH ]
[MBM − ] = ,
(K1 + [H + ])

K2 [HA]
[A− ] = ,
K2 + [H + ]

K3 [HABM ]
[ABM − ] = .
K3 + [H + ]

Material balance equations for the three reactants in slow kinetic reactions yield
1.1 Chemistry and Biology 5

d [M − ]
= −k1 [M − ][BM ] + k−1 [MBM − ] − k3 [M − ][AB] + k−3 [ABM − ],
dt
d [BM ]
= −k1 [M − ][BM ] + k−1 [MBM − ] − k2 [A− ][BM ],
dt
d [AB]
= −k3 [M − ][AB] + k−3 [ABM − ].
dt
As we can see from stoichiometry, rate expressions can also be written for the
total species
d [MBMH ]
= k1 [M − ][BM ] + k−1 [MBM − ],
dt
d [HA]
= k2 [A− ][BM ],
dt
d [HABM ]
= k2 [A− ][BM ] + k3 [M − ][AB] − k−3 [ABM − ].
dt

An electroneutrality constraint gives the hydrogen ion concentration [H + ] as

[H + ] + [Q+ ] = [M − ] + [MBM − ] + [A− ] + [ABM − ].

We also assume
k3 = k1 and k−3 = 0, 5k−1

in terms of similarities of the reacting species. Denoting the amount of the ith reactant
by xi , we get a mathematical model in the form:

ẋ1 = −k2 x2 x8 ,
ẋ2 = −k1 x2 x6 + k−1 x10 − k2 x2 x8 ,
ẋ3 = k2 x2 x8 + k1 x4 x6 − 0.5k−1 x9 ,
ẋ4 = −k1 x4 x6 + 0.5k−1 x9 ,
ẋ5 = k1 x2 x6 + k−1 x10 ,
ẋ6 = −k1 x2 x6 − k1 x4 x6 + k−1 x10 + 0.5k−1 x9 ,
0 = −x7 + x6 + x8 + x9 + x10 − Q+ ,
0 = −x8 (K2 + x7 ) + K2 x1 ,
0 = −x9 (K3 + x7 ) + K3 x3 ,
0 = −x10 (K1 + x7 ) + K1 x5 .

This mathematical model is a nonlinear descriptor system, and consists of six differ-
ential and four algebraic equations.
6 1 Practical Application of Descriptor Systems

Parameter Q+ , that stands for the amount of catalyst in the reactor, is known, as
paid before the reaction.

1.1.2 Chemical Reaction of Urethane

Consider a chemical process that takes place in the urethane reactor system. The
reactions are given as follows.
A + B → C,

A + C  D,

3A → E,

where A is phenyl isocyanate, B is butanol, C is urethane, D is allophanate, and E is

isocyanurate.
A mathematical model describing the reaction of isocyanate (n1 ), butanol (n2 ),
urethane (n3 ), allophanate (n4 ), and isocyanurate (n5 ) consists of three differential
and three algebraic equations.

ṅ3 = V (r1 − r2 + r3 ),
ṅ4 = V (r2 − r3 ),
ṅ5 = V r4 ,
0 = n1 + n3 + 2n4 + 3n5 − na1 − n1ea (t),
0 = n2 + n3 + n4 − na2 − n2eb (t),
0 = n6 − na6 − n6ea (t) − n6eb (t),

where n6 is the solvent of dimethylsulfoxide, n3 (0) = 0, n4 (0) = 0, n5 (0) = 0, and

the following parameters are
6 Ea1 (1/T (t)−1/Tref 1 (t))
V = i=1
Mi ni
ρi
, k1 = kref 1 exp(− R
),

Ea2 (1/T (t)−1/Tref 2 (t))

r1 = k1 nV1 n22 , k2 = kref 2 exp(− R
),

r2 = k2 nV1 n22 , k3 = k2 /kc ,

Ea4 (1/T (t)−1/Tref 4 (t))

r3 = k3 nV4 , k4 = kref 4 exp(− R
),
1.1 Chemistry and Biology 7

n2 dh2 (1/T (t)−1/Tg2 (t))

r4 = k4 V12 , kc = kc2 exp(− R
).

Two control feeds have the form of nonincreasing functions feeda (t) and feedb (t),
and determine n1ea = na1ea feeda (t), n2eb = na2eb feedb (t), n6ea = na6ea feeda (t),
and n6eb = na6eb feedb (t). Mole ratios of the active ingredients and the initial volume
satisfy the constraints
0.1 ≤ M V1 ≤ 10,

0 ≤ M V2 ≤ 1000,

0 ≤ M V3 ≤ 10,

0 ≤ ga ≤ 0, 8,

0 ≤ gaea ≤ 0, 9,

0 ≤ gaeb ≤ 1,

0 ≤ Va ≤ 0.00075,

and they are connected with the other parameters by the following algebraic rela-
tionships.

M V1 (na1 + na1ea ) = na2 + na2eb ,

M V2 na1 = na1ea ,
M V3 na1 = na2eb ,
ga (na1 M1 + na2 M2 + na6 M6 ) = na1 M1 + na2 M2 ,
gaea (na1ea M1 + na6ea M6 ) = na1ea M1 ,
gaeb (na2eb M2 + na6eb M6 ) = na2eb M2 ,
Va = na1 M1 /ρ1 + na2 M2 /ρ2 + na6 M6 /ρ6 ,

that are nonlinear constraints. Numerical values of the parameters can be found in [3].

1.1.3 Evaporator

Consider a single-component system of phase equilibrium where there are gaseous

and liquid phases. The plant is shown in Fig. 1.2. Vapor (denoted by subscript V ) and
liquid (subscript L ) are in a heated container. There are two components with masses
MV , ML , and temperatures TV and TL inside the tank. The system has a feed with
a flow F. In this model, two volume balance equations are under consideration for
the gaseous and liquid phases. Equations describing the dynamics of the process are
discussed below.
8 1 Practical Application of Descriptor Systems

Conservation Laws
Mass balance is described by

ṀV = E − V,
ṀL = F − E − L.

Energy balance is given in the form:

U̇V = EhLV − V hV + QE ,
U̇L = FhF − EhLV − LhL + Q − QE .

Transfer equations of mass and energy are given below.

E = (kLV + kV L )A(P ∗ − P), (1.8)

QE = (uLV + uV L )A(TL − TV ), (1.9)

where the subscript LV in transfer coefficients of mass and energy ki and ui means
a transfer from liquid to vapor, and V L from vapor to liquid. The coefficients for V L
and LV are usually different.
Balance of volumes equation is

VV = VT − VL . (1.10)

Control laws are

L = f1 (ML , P) or L = f2 (ML ).

The relation between the variables is expressed as

hV = hV (TV , P),

hL = hL (TL , P),

hLV = hLV (TL , P),

hF = hF (TF , P),

kLV = kLV (TL , TV , P),

kV L = kV L (TL , TV , P),

uLV = uLV (TL , TV , P),

uV L = uV L (TL , TV , P),

ρL = ρL (TL , P).
1.1 Chemistry and Biology 9

Notations and variables shown in the example are the following.

MV Mass of vapor
UV Internal energy of vapor
F Feed speed
L Flow rate of liquid
TV Temperature of vapor
Q Power of a heat source
P System pressure
A Surface area of the phase boundary
VV Vapor volume
hV Vapor enthalpy
hF Feed enthalpy
ρL Liquid density
VL Liquid volume
ML Liquid mass
UL Internal energy of liquid
V Vapor flow rate
E Speed of phase separation
TL Liquid temperature
P∗ Vapor pressure
R Universal gas constant
VT Tank volume
hL Liquid enthalpy
mw Molecular mass
hLV Vapor enthalpy at the interface
ki Exchange ratio weight
ui Heat transfer coefficient

1.1.4 Multispecies Food Chain

This problem concerns modeling of the multispecies ratio of predator-prey in a closed

area [4]. We assume that the model consists of s individuals where the individuals
with numbers s/2 + 1, . . . , s (predators) have an infinitely fast reaction. Then we
have

∂ci
= fi (x, y, t, c) + di (cxx
i
+ cyy
i
), (i = 1, 2, . . . , s/2), (1.11)
∂t
0 = fi (x, y, t, c) + di (cxx
i
+ cyy
i
), (i = s/2 + 1, . . . , s), (1.12)

where

s
fi (x, y, t, c) = c (bi +
i
aij cj ). (1.13)
j=1
Another random document with
no related content on Scribd:
is a Pretender, he may have his agents in the army as well as he has
everywhere else.’ Officers (according to the duke) might be led away
from their duty, and he held it to be unjust to the king to deprive him
of the right to dismiss officers suspected of Jacobitism, or known to
be disloyal, on evidence which a court martial might not think
sufficient for cashiering them. The Bill was lost, and to the king was
left the power of doing wrong.
In a portion of the Duke of Newcastle’s speech he asserted that
the right claimed for the king was indispensable, on the ground that
not only were private soldiers being recruited in London for ‘foreign
service,’ but that officers might be tampered with, and that there was
no real security that a general-in-chief might not be seduced into the
enemy’s camp. This spread some alarm. The debates, indeed, were
supposed to be delivered in private, but what was called ‘the
impudence of some fellows’ gave all that was essential to the public.
For defence of the nation, however, every precaution had been
taken. Early in the spring, a fleet of twenty sail of the line was sent to
the Downs. Eight regiments were brought from Ireland to England. It
is certain that these precautions preserved the public tranquility of
the kingdom. Young Prince Charles Edward was
THE YOUNG
serving ‘with particular marks of distinction’ in the CHEVALIER.
army of Don Carlos; and the boy gave no obscure
hints that he would, whenever it was in his power, favour the
pretensions of his family. An exclamation of Sergeant Cotton, at a
review in Hyde Park, that he would shoot the king; and the fact that
the sergeant’s musket was loaded with ball, and that he had a
couple of bullets in his pocket which had no right to be there,
seemed to imply that Cotton was ready to favour the Stuart family’s
pretensions.
The metropolis, moreover, was disturbed this year by the
appearance of strangers in the streets, with more or less of a military
air about many of them. These were, however, for the most part,
Jacobites who were void of offence, and who had hastily come over
from France. The Government there had given them a taste of what
it was to live under such a system in Church and State as the Stuarts
would establish in England, if they could get permanent footing
there. A royal edict was published throughout France, peremptorily
commanding all English, Irish, and Scotch, of the ages between
eighteen and fifty, who were without employment, to enter the
French army, within a fortnight. Disobedience to this edict was to be
punished:—civilians, by condemnation to the galleys;—men who had
formerly served, to be shot as deserters! Those who were not
fortunate enough to get away from such a paternal Government
found friends in the ministers of that George II. whom they still styled
‘Duke of Brunswick’ and ‘Elector of Hanover.’ Lord Waldegrave, the
British Ambassador in France, sharply censured the edict,
remonstrated against the injustice of treating the persons named in
the edict worse than the natives of any other country, and pointed to
the ingratitude of the French Government for various good service
rendered to it by England on recent occasions. There was not a
place in London where men met, but there Lord Waldegrave’s health
was drunk. Whatever the politics of the drinkers were, all parties
were glad to find a cause for drinking which carried unanimity with it.
There was another Jacobite incident of the year, LORD
not without interest. Queen Anne’s old naval captain, DUFFUS.
the gallant Kenneth, Lord Duffus, when attainted for
his share in the affair of 1715, was in safety in Sweden, but he gave
formal notice of his intention to repair to England and surrender
himself. On his way, the British Minister at Hamburg had him
arrested, and he held Lord Duffus prisoner till after the limited time
had elapsed for the surrender of attainted persons. Lord Duffus was
brought captive to London, was shut up in the Tower, and, destitute
of means, was maintained at the expense of Government. By the Act
of Grace, of 1717, he obtained his liberty, and he subsequently
entered the naval service of Russia. At his death, he left an only son,
Eric Sutherland (whose mother was a Swedish lady) who, in this
year, 1734, at the age of twenty-four, claimed the reversal of his
father’s attainder (as Lord Duffus was forcibly prevented from
obeying the statute), and his own right to succeed to the baronial
title. The claim excited much interest while it was being pursued; and
there was some disappointment in Jacobite circles when the Lords
came to a decision that the claimant had no right to the honour, title,
and dignity of Baron Duffus. Eric was, at this time, a loyal officer in
the British army; he died in 1768. He left a son, James, born in 1747,
who was restored to the title, by Act of Parliament, in 1826, when he
was in his eightieth year. He enjoyed it only a few months. His
successor, Benjamin, died in 1875, when the title became extinct.
The 30th of January 1735 was kept in memory by THE CALVES’
other means than ‘services’ before the Senate, and HEAD CLUB.
others in the parish churches. By a tradition which
was founded in a lie, and which rooted itself and grew in the public
mind by additional lying, there was a popular belief that a Calves’
Head Club, from the time of Cromwell, had a special meeting and
dinner on every anniversary of the death of King Charles, to
dishonour his memory. The calf’s head served at table was in
derisive memory of the decollated head of that sovereign; and the
ocean of liquor drunk was in joyous celebration of those who brought
about the monarch’s death. The story was a pure invention, but the
invention led to a sort of realisation of the story. Here and there, anti-
Jacobites observed the 30th of January as a festival.
THE CALVES’
Hearne mentions a dinner given on that day by a HEAD RIOT.
number of young men at All Soul’s College, Oxford.
They had ordered a calf’s head to be served up, but the cook
refused to supply it. He unwittingly, however, gave the guests an
opportunity of declaring their approval of the sentence executed on
Charles, by sending them a dish of woodcocks, and these the
audacious Oxford Whigs solemnly decapitated. In the present year,
1735, occurred the famous Calves’ Head riot at and in front of a
tavern in Suffolk Street. According to the record, some noblemen
and gentlemen had the traditional dinner on the above day, when
they exhibited to the mob, which had assembled in the street, a calf’s
head in a napkin dipped in claret to represent blood, and the
exhibitors, each with a claret-stained napkin—in his hand and a
glass of strong liquor in the other, drank anti-Stuart toasts, and finally
flung the head into a bonfire which they had commanded to be
kindled in front of the house. The Jacobite mob broke into the house
and would have made ‘martyrs’ of the revellers but for the timely
arrival of the guards. Now, with regard to this incident, there are two
opposite and contemporary witnesses, whose testimony
nevertheless is not irreconcilable. The first is ‘a lady of strong
political tendencies and too busy in matters of taste to be ignorant of
party movements.’ She is so described by a correspondent of the
‘Times,’ who, under the signature ‘Antiquus,’ sent to that paper a few
years ago the following copy of a letter, written by the lady, and
forming one of a collection of old letters in the possession of
‘Antiquus, of Lincoln’s Inn’:—
‘I suppose you have heard of the Suffolk-street
THE ‘30TH OF
Expedition on the Thirtieth of January, and who JANUARY.’
the blades were; they went and bespoke a dinner
of calves’ heads at the Golden Eagle, and afterwards ordered
a bonfire at the door, then came all to the window with
handkerchiefs dipt in blood, and shook them out, and dress’d
up a calf’s head in a nightcap and had it thrown into the
bonfire. The mob gather’d about the door and were
exceedingly inraged, so that they broke ye door open and
broke all the windows, and threw fire into the house. The
gentlemen were forc’d to take sanctuary in the garret, and
had not the Guards been sent for the house would have been
pull’d down and the actors, no doubt, pull’d to pieces.
‘Feb. 5, 1734-5.’
‘The list of the British worthies I formerly sent you an
account of are as follows:—Lord Middlesex, Lord Harcourt,
Lord Boyne, and Lord Middleton—Irish; Lord John Murray, Sir
James Grey, Mr. Smith, Mr. Stroud, and, some say, Mr.
Shirley. Lord A. Hamilton dined with them, but, I am told, went
away before the riot began.
‘Feb. 16, 1734-5.’
Unfortunately, the name of the writer of the above OBJECTIONAB
letter is not given. On the other hand, a letter written LE TOASTS.
by one of the guests, a week earlier than the above,
has often been published. Therein, Lord Middlesex informs Spence,
then at Oxford, that he and seven others met at the Golden Eagle to
dine, without any thought as to what the date of the month was. The
eight included men of various political and religious principles. Lord
Middlesex says nothing as to the dishes served up, but he states
that all the guests had drunk hard and some were very drunk indeed,
when, happening to go to the window, they saw a bonfire in the
street, and straightway ordered fresh faggots, by which they had a
bonfire of their own. Then, they remembered the day, and fearful of
the consequences of this demonstration, the soberer part of the
guests proposed, from the open windows, loyal toasts to be drunk by
all. To a Jacobite mob this was an aggravation of insult, for to drink
the king, the Protestant succession, and the administration, was to
express affection for what they cordially hated. The mob besieged
the house, and then made an ugly rush to get at the offenders,
which, however, was checked by the arrival of the soldiers. Lord
Middlesex says that the leader of the mob was ‘an Irishman and a
priest belonging to Imberti, the Venetian Envoy.’
In the pulpits of the chapels of some of the foreign ambassadors,
—most Christian, most Catholic, or most Apostolic,—the preachers,
naturally enough, expounded Christianity in a politico-religious point
of view. The Protestant-succession papers speak of them as a
daring vanguard dashing forward to secure improved and fixed
positions. Of course, the preachers, when supporting the Papacy,
were advocating the Pretender by whom, were the Stuarts restored,
the Papacy would be supported. This led to an outburst of anti-papal
sermons from half the London pulpits. Secker, the ex-dissenter, ex-
medical student, and now Bishop of Bristol, was at the head of this
body. They preached sermons against Popery in a long and fiery
series, in some cases to the extent of two or three dozen. Where, on
one side doctrines were sincerely held which made the other side
sincerely shudder, as at awful blasphemy, charity got sadly mauled
and knocked about. It occurred to James Foster, the
FOSTER, IN THE
celebrated Baptist who had passed through Arianism OLD JEWRY.
and Socinianism, before he became a Trinitarian, that
good citizens of both churches and factions might be made even
better by their understanding the excellence of charity. His pulpit in
the Old Jewry became accordingly a point to which men of opposite
opinions resorted,—just indeed as they did to the Popish
ambassadorial chapel, where they could hear gratis the great tenor
Farinelli sing mellifluously. In reference to Foster, the general
‘Evening Post,’ of March 25th, says that on the previous Sunday
evening, ‘upwards of a hundred Gentlemen’s coaches came to the
Rev. Mr. Foster’s lecture in the Old Jewry. It must give,’ adds the
newswriter, ‘a great Satisfaction to that ingenious and polite
Preacher, to see such an Audience at his Lectures, as well as to be
a Reputation to his Hearers, in their discovering a disposition to be
pleased with his useful and instructive Discourses, they turning upon
the Truth, Excellency, and Usefulness of the grand Parts of Moral
Science; not tending to support private or party egotism of Religion,
or Rule of Conduct, but a Conduct founded on the most sacred
Rights of Mankind, a universal Liberty, and a diffusive and extensive
Benevolence.’
Another account states that ‘at his chapel there was a confluence
of persons of every rank, station, and quality; wits, freethinkers, and
numbers of the regular clergy who, while they gratified their curiosity,
had their prepossessions shaken and their prejudices loosened.’
There was one Jacobite who died this year, whose THE QUEEN
prejudices were never in the least degree softened, AND THE
namely, Hearne, the antiquary. Richardson the ARTIST.
painter, when party spirit between Whig and Tory,
Hanoverian and Jacobite raged bitterly, was as severe in a remark to
Queen Caroline, as Hearne was in what he wrote upon her. The
queen once visited Richardson’s studio to view his series of portraits
of the kings of England. Her Majesty pointed to the portrait of a
stern-looking individual between those of Charles I. and II. She very
well knew the likeness was that of a man who had helped to
dethrone the Stuarts on whose throne her husband was seated, and
she therefore might have entertained a certain respect for him; but
she asked the artist if he called that personage a king? ‘No, madam,’
answered the undaunted Richardson, ‘he is no King, but it is good
for Kings to have him among them as a memento!’
The queen’s favourite painter, Anniconi, was more of a courtier
than blunt Richardson. To that artist who, for a season, drew the
‘Quality’ to Great Marlborough Street, she gave an order to paint a
picture, which was designed as a gift to the young Duke of
Cumberland’s tutor, Mr. Poyntz. It was an allegorical composition, in
which the queen herself was to be seen delivering her royal son to
the Goddess of Wisdom,—who bore the features of Mrs. Poyntz.
The year 1736 may be said to have opened CHESTERFIEL
merrily, with Chesterfield’s paper in ‘Fog’s Journal,’ on D’S WIT.
‘An Army in Wax Work.’ In the course of this lively
essay, the writer argues that since the English army had not been of
the slightest active use during many years, in time of war,—a waxen
army (to be ordered of Mrs. Salmon, the wax-work woman) would be
cheap and sufficient in time of peace. He then alludes to the
Government cry against all who opposed it. ‘Let nobody put the
“Jacobite” upon me, and say that I am paving the way for the
Pretender, by disbanding the army. That argument is worn
threadbare; besides, let those take the “Jacobite” to themselves who
would exchange the affections of the people for the fallacious
security of an unpopular standing army.’
While there were, at this time, Nonjurors worthy of SCENE IN
the esteem of honourable men of all parties, there WESTMINSTE
were others who were contemptible for their R HALL.
spitefulness, and for the silliness with which they
displayed it. Here is an example. Parliament had passed the Gin Act,
the Mortmain Act, the Westminster Bridge Act, the Smugglers’ Act,
and the Act for borrowing 600,000l. on the Sinking Fund. A
difference of opinion might exist as to the merits of one or two of
these Acts, but there was no justification for the method taken by
one person to show his hostility. On July 14th, in Westminster Hall,
while the Courts were sitting therein, a bundle, dropped in front of
the Court of Chancery, suddenly exploded, and blew into the air a
number of handbills, which announced that, on this, the last day of
term, copies of the above-named Acts would be publicly burned in
the hall during the afternoon! One of the bills was handed in to the
judges in the Court of King’s Bench, where it was presented as a
false and scandalous libel. Three days later a proclamation was
issued for the discovery of the persons concerned in this outrage,
and a reward of 200l. offered for the respective arrests of either the
author, printer, or disperser of the handbills. This led to the arrest,
trial, and conviction of the Rev. Mr. Nixon, a brainless Nonjuring
clergyman, who was proved to be the author of the bills, and the
blower-up of the bundle of crackers. On the 7th of December he was
condemned to pay 200 marks, to be imprisoned for five years, and to
be paraded before the different Courts, in the Hall, with a parchment
round his head—a sort of foolscap—bearing a summary of his
audacious offence. A portion of this sentence was fulfilled soon after,
and, finally, this foolish Nonjuror was required to find security for his
good behaviour during the remainder of his life.
This daring, yet stupid, act was supposed to be part of an
organised Jacobite plot. In the month of April, when Frederick,
Prince of Wales, was married to the Princess of Saxe-Gotha, Sir
Robert Walpole had information which set him on his guard. After the
explosion in Westminster Hall, he wrote a letter to his brother
Horace, in which the following passage is to be found:—‘Since my
coming to town I have been endeavouring to trace out the authors
and managers of that vile transaction, and there is no reason to
doubt that the whole was projected and executed by a set of low
Jacobites, who talked of setting fire to the gallery built for the
marriage of the Princess Royal, by a preparation which they call
phosphorus, which takes fire from the air. Of this I have had an
account from the same fellow that brought me these, and many such
sorts of intelligencies.’
And again, in September, when it was decreed that JACOBITES
unlicensed dealing in gin should cease, riots AND GIN-
occurred, and more than mere rioting was intended, DRINKERS.
in the metropolis, about Michaelmas. On this
occasion Sir Robert wrote to his brother:—‘I began to receive
accounts from all quarters of the town that the Jacobites were busy
and industrious, in endeavouring to stir up the common people and
make an advantage of the universal clamour that prevailed among
the populace at the expiration of their darling vice.’ The Jacobite idea
was, according to the information received by Walpole, to make the
populace drunk gratis by unlimited supplies of gin from the
distilleries, and then turning them loose in London to do such work
as such inspiration was likely to suggest to them; but an efficient
display of the constitutional forces was sufficient to preserve the
peace of the metropolis.
The alleged abuse of the liberty of the press and of THE STAGE
that of the stage was denounced, as all opposition to FETTERED.
the Government was, as the work of Jacobites for the
subversion ‘of our present happy establishment.’ The Government
undoubtedly hoped, by suppressing the liberty of satire on the stage,
to be enabled to go a step further, and to crush the liberty of
comment in the press. Sir Robert made his own opportunity to
ensure the success of his preliminary step. Mr. Giffard, of the theatre
in Goodman’s Fields, waited on Sir Robert in 1737 with the MS. of a
piece named ‘The Golden Rump,’ which had been sent to him, for
performance, by the anonymous author. Its spirit was so licentiously
manifested against the Ministry, and was so revolutionary in its
speech, suggestions, and principles, that the prudent manager felt
bound to place it at the discretion of the minister. Sir Robert put it in
his pocket, went down to the House with it, and ultimately
succeeded, by its means, in carrying the Licensing Act, by which the
stage has been ever since fettered. The anonymous piece brought
by Giffard was never acted, never printed, probably never seen by
anyone except manager and minister; and the question remains,—
Was it not written to order, to afford a plausible pretext for protecting
the administration from all its antagonists? Chesterfield, in his
speech in the Lords against the proposed Act, denounced it as a
long stride towards the destruction of liberty itself. He declared that it
would be made subservient to the politics and schemes of the Court
only. In the same speech occurred the famous passage: ‘This Bill,
my Lords, is not only an encroachment upon Liberty, but it is likewise
an encroachment upon Property. Wit, my Lords, is a sort of property.
It is the property of those who have it, and too often the only property
they have to depend on. It is indeed but a precarious dependence.
Thank God! we, my Lords, have a dependence of another kind.’
In 1738, when the Opposition proposed a FEAR OF THE
reduction of the army, the Government manifested an PRETENDER.
almost craven spirit. They believed that if the number
of armed men were diminished, the king would not be secure from
assault in St. James’s, nor the country safe from foreign invasion.
In the Commons, Sir Robert Walpole spoke as WALPOLE, ON
follows, on the Jacobites, their views, and their JACOBITES.
dealings at that period:—‘There is one thing I am still
afraid of, and it is indeed I think the only thing at present we have to
fear. Whether it be proper to mention it on this occasion, I do not
know; I do not know if I ought to mention it in such an Assembly as
this. I am sure there is no necessity for mentioning it, because I am
convinced that every gentleman that hears me is as much afraid of it
as I am. The fear I mean is that of the Pretender. Everyone knows
there is still a Pretender to his Majesty’s crown and dignity. There is
still a person who pretends to be lawful and rightful sovereign of
these kingdoms; and what makes the misfortune much the more
considerable, there is still a great number of persons in these
kingdoms so deluded by his abettors, as to think in the same way.
These are the only persons who can properly be called disaffected,
and they are still so numerous that though this government had not a
foreign enemy under the sun, the danger we are in from the
Pretender and the disaffected part of our own subjects, is a danger
which every true Briton ought to fear; a danger which every man who
has a due regard for our present happy establishment, will certainly
endeavour to provide against as much as he can.
‘I am sorry to see, Sir, that this is a sort of fear which many
amongst us endeavour to turn into ridicule, and for that purpose they
tell us that though there are many of our subjects discontented and
uneasy, there are very few disaffected; but I must beg leave to be of
a different opinion, for I believe that most of the discontents and
uneasinesses that appear among the people proceed originally from
disaffection. No man of common prudence will profess himself
openly a Jacobite. By so doing he not only may injure his private
fortune, but he must render himself less able to do any effectual
service to the cause he has embraced; therefore there are but few
such men in the kingdom. Your right Jacobite, Sir, disguises his true
sentiments. He roars out for Revolution principles. He pretends to be
a great friend of Liberty, and a great adviser of our ancient
Constitution; and under this pretence there are numbers who every
day endeavour to sow discontent among the people, by persuading
them that the constitution is in danger, and that they are
unnecessarily loaded with many and heavy taxes. These men know
that discontent and disaffection are, like wit and madness, separated
by thin partitions, and therefore hope that if they can once render the
people thoroughly discontented, it will be easy for them to render
them disaffected. These are the men we have the most reason to be
afraid of. They are, I am afraid, more numerous than most
gentlemen imagine; and I wish I could not say they have been lately
joined, and very much assisted, by some gentlemen who, I am
convinced, have always been, and still are, very sincere and true
friends to our happy establishment.’
Walpole went on to say that he hoped Jacobitism
CURIOUS
would die out. He was sure the Jacobites were daily DISCUSSION.
decreasing; but if such a mad step were taken as that
of reducing the army—‘I should expect to hear of the Pretender’s
standards being set up in several parts of the island, perhaps in
every part of the three kingdoms.’
Wyndham ridiculed the idea that the army must not SAFETY OF
be reduced, because ‘a certain gentleman was afraid THE ROYAL
of the Pretender.’ Lord Polwarth (afterwards Earl of FAMILY.
Marchment) went further. He could scarcely see the
use of an army at all, and did not believe that there were Jacobites to
be afraid of. ‘I am sure his Majesty, and all the rest of the Royal
Family, might remain in St. James’s Palace, or in any other part of
the kingdom, in the utmost safety, though neither of them had any
such thing as that now called a soldier to attend them. Of this now
we have a glaring proof every day before our eyes. His royal
highness the Prince of Wales has now no guards to attend him. He
passes every day to and fro in the streets of London, and travels
everywhere about London without so much as one soldier to guard
him. Nay, he has not so much as one sentry upon his house in St.
James’s Square, and yet his Royal Highness lives, I believe, in as
great security, at his house in St. James’s Square, without one sentry
to guard him, as his Majesty can be supposed to do in St. James’s
Palace with all the guards about him.’
The debate in the Lords was of much the same quality as that in
the Commons. Farewell to liberty if there be a standing army. On the
other side:—Freedom will perish if the king cannot back his will by
force of bayonets. The Government, of course, succeeded.
The debates encouraged the Jacobites to hope. ‘AGAMEMNON.
They were evidently feared, and opportunity might yet ’
serve them. The wise men at Westminster had
declared it. Meanwhile, the stage recommended them to consider
the difficulties of Government, and to make the best of the one under
which they lived. Thomson put his tragedy ‘Agamemnon’ under the
protection of the Princess of Wales, trusting she would ‘condescend
to accept of it.’ In the tragedy itself, in which there is much blank
verse that is only honest prose in that aspiring form, there are few
political allusions; but the following passage was undoubtedly meant
as incense for Cæsar, and instruction for his people—Whigs and
Jacobites.
Agamemnon . . . —Know, Ægisthus,
That ruling a free people well in peace,
Without or yielding, or usurping, power;—
Maintaining firm the honour of the laws,
Yet sometimes soft’ning their too rigid doom,
As mercy may require, steering the state
Thro’ factious storms, or the more dangerous calms
Of Peace, by long continuance grown corrupt;
Besides the fair career which Fortune opens
To the mild glories of protected arts,
To bounty, to beneficence, to deeds
That give the Gods themselves their brightest beams;—
Yes, know that these are, in true glory, equal
If not superior to deluding conquest;
Nor less demand they conduct, courage, care,
And persevering toil.
 Ægisthus answered with a slight rebuke to the Jacobites who
denounced the merits of all government that had not their James III.
at its head:—
Say, thankless toil,
Harsh and unpleasing, that, instead of praise
And due reward, meets oft’ner scorn, reproach,
Fierce opposition to the clearest measures,
Injustice, banishment, or death itself,
Such is the nature of malignant man.
Quin, as Agamemnon, rolled his measured lines THE KING, IN
out with double emphasis, his anti-Stuart feelings PUBLIC.
adding to the force. The ‘fierce oppositions’ of Ægisthus were not to
be found in factious shape, at least, in the next session of
Parliament. The debates at the opening of the session had but the
slightest touch of Jacobitism in them; and that was in a speech by
Lord Gower,—whom Horace Walpole classed with the Prince of
Wales himself as a thorough Jacobite! Lord Gower spoke ill of the
‘so-called’ King’s Speech as being no royal speech at all, but one
which conveyed the dictates of the Ministry to the country. ‘The
King,’ said Lord Gower, ‘has no more share in the councils of the
country than I have.’ A faint allusion in the Commons to his Majesty
and family being less popularly esteemed than formerly, Mr. Lyttelton
remarked: ‘I’ve repeatedly seen proofs to the contrary. In the streets
of London I’ve seen the people clinging to the wheels of his coach,
so as almost to impede it;’—and the inference was that they would
not have so affectionately clung to the chariot-wheels of the
Pretender. Other proofs, during the session, were adduced of the
satisfactory condition of things. Recruiting for his Majesty’s army was
successfully going on in Scotland, and the last cargo of old firelocks,
resulting from the disarming of the Highlanders, was just then being
landed at the Tower. Nevertheless, there were Jacobites who were
hoping for the best, and keeping their powder dry.
Thomson made another effort in the year 1739 to POLITICAL
introduce politics on the stage. His ‘Edward and DRAMA.
Eleanora’ (after being publicly rehearsed) was
advertised for representation, on March 29th, at Covent Garden; but,
before the doors were open, the licenser withdrew his permission,
and prohibited the performance absolutely. Thomson’s almost servile
worship of the reigning family was manifested in the dedication of the
tragedy to his patroness, the Princess of Wales. ‘In the character of
Eleanora,’ he says, ‘I have endeavoured to represent, however
faintly, a princess distinguished for all the virtues that render
greatness amiable. I have aimed particularly to do justice to her
inviolable affection and generous tenderness for a prince who was
the darling of a great and free people.’ As Eleanora loved Edward,
so, it was hinted, did Augusta love Frederick!
Dr. Johnson could not see why this play was ‘obstructed.’ Genest
could no more see the reason than Dr. Johnson. Yet, the licenser
may be easily justified in withdrawing a license which should never
have been granted. The play touched nearly on the dissensions
between George II. and his son Frederick, who were then living in
open hostility. Such passages as the following would certainly have
been hailed with hilarious sarcasm by the Jacobites, who dwelt with
satisfaction on the unseemly antagonisms in the royal family:—
Has not the royal heir a juster claim
To share the Father’s inmost heart and Counsels,
Than aliens to his interest, those who make
A property, a market, of his honour?
The prince is urged to save the king from his ministers; England is
represented as in peril from without as well as from within. Frederick,
under the name of Edward, is described as one who ‘loves the
people he must one day rule,’—Whigs and Jacobites equally, for:—
Yet bears his bosom no remaining grudge
Of those distracted times.
When Henry III. is declared to be dead, his son
HENRY
thus speaks of him in terms applicable, by the poet’s PELHAM AND
intention, to George II.:— THE
JACOBITES.
The gentlest of mankind, the most abus’d!
Of gracious nature, a fit soil for virtues,
 g , ,
Till there his creatures sow’d their flatt’ring lies,
And made him—No! not all their cursed arts
Could ever make him insolent or cruel.
O my deluded father! Little joy
Had’st thou in life;—led from thy real good,
And genuine glory, from thy people’s love,—
That noblest aim of Kings,—by smiling traitors!
These domestic and political allusions pervade the play. Its
production would probably have led to riot, and the Lord
Chamberlain, or his deputy, did well in prohibiting the play and thus
keeping the peace.
In January, 1740, Mr. Sandys moved for leave to bring in a Bill for
the better securing the freedom of Parliament, by limiting the number
of Government officers to sit in the House of Commons. Among the
opponents was Mr. Henry Pelham, who was convinced that the Bill
would help the Jacobites to carry out their designs. ‘We know,’ he
said, ‘how numerous the disaffected still are in this kingdom; and
they, we may suppose, are not insensible to the prejudice that has
been done to their faction, by the places and offices which are at the
disposal of the crown. These places and offices are of great use to
the crown and, I think, to the nation, in preventing gentlemen from
joining with a faction, or winning them away from it; and the
Jacobites are sensible they have lost many by this means, some,
perhaps, after they had got a seat in this House.’
Mr. Pulteney, alluding to the assertion that if most JACOBITE
placemen were excluded from the House, there would PROSPECTS.
soon be a majority of Jacobites in it, said this was
supposing that there was a majority of Jacobites among the people,
a supposition which he denied, and which he stigmatised as very
uncomplimentary to the king and his family. ‘But,’ he added, ‘if there
should once come to be a majority of placemen and officers in this
House, that majority would soon create a majority of Jacobites in the
nation.’ The consequences, he was sure, would be an insurrection,
the army joining with the insurgents. This motion, in the debate on
which the Jacobites figured as both a dangerous and a mercenary
people, was lost by 222 to 206. Sixteen placemen saved Sir Robert,
who had spoken with much plausibility and cunning against the leave
asked for. The Bishop of Salisbury in the House of Lords, in the
discussion in March on the Pension Bill, could only express a hope
that faction would not foster insurrection. The opposition papers
maintained that no such thing as faction existed, and that Jacobitism
was a name now utterly unknown to the mass of the people. The
opposition to Sir Robert was increasing in strength, and this was
taken to be a proof that the Jacobites were increasing in number; but
everything was done to sustain the minister. ‘’Tis observable,’ says
the ‘Craftsman,’ ‘that St. Stephen’s Chapel was never attended with
more devotion than at present, the very lame and the blind hardly
being excused; and both Parties seeming to indicate by their conduct
that nobody knows what a day or an hour may bring forth.’
The opposition, rather than the Jacobite party, DEATH OF
experienced an immense loss this year, by the death WYNDHAM.
of Sir William Wyndham. This able man ceased to be
a Jacobite after he gave in his allegiance to the accomplished fact of
the established supremacy of the House of Hanover. Wyndham
became simply a ‘patriot,’ never ceasing his fierce, but polished,
hostility to Walpole, yet lending himself to no measure likely to
disturb the ‘happy establishment.’ Two years before his death he
took for second wife the widowed Marchioness of Blandford, whose
relatives opposed a match with an ex-Jacobite. ‘She has done quite
right,’ said the old Dowager Duchess of Marlborough. ‘I’d have had
him myself, if he’d only asked me sometime ago!’
The camp pitched at Hounslow this year reminded quidnuncs of
the one formed by James II., to overawe London. Londoners
themselves expressed a hope that the army would sweep ‘the
infamous road’ of its mounted highwaymen and its brutal footpads.
But by the presence of soldiers there was only an addition to the
number of robbers and of victims.

[2] ‘Last Journals of Horace Walpole,’ vol. i. p. 41.

 CHAPTER IV.

(1741 to 1744.)
t the time when to be discovered carrying on a
treasonable correspondence with the Chevalier might
cost a man his life, Walpole made such a discovery in
the person of a friend of honest Shippen who, himself,
kept up such correspondence, but was successful in
keeping it concealed. Shippen went to the minister with an urgent
entreaty not to bring down destruction on his friend. Mercy was a
card it suited the minister to play; he granted the prayer of his great
political opponent. But he suggested a stipulation. ‘I do not ask you,’
said Sir Robert, ‘to vote against your principles; but if questions
should arise in the House, personal to myself, do not then forget
what I have done for you to-day.’
A great personal question did arise,—this year. Lord Carteret in
the Lords, and Mr. Sandys, with his long cravat of Queen Anne’s
days, in the Commons, moved, on the same day and in precisely the
same words, that the king should be requested to dismiss Walpole
from his service and counsels for ever. The debate was hot in each
House, and the object of the movers was unsuccessful in both. In the
Commons, this incident occurred. The impetuous Jacobite Shippen
rose to speak. He certainly astonished the House. The motion, he
said, was merely made to put out one minister, and to put in another.
For his part, he did not care who was in or who was out. He would
not vote at all. Shippen walked out of the House, and he was
followed by thirty-four friends who had yielded to his persuasions. He
thus proved to Walpole the gratefulness of his memory.
This was not the only incident of the debate. Mr.
INCIDENTS IN
Edward Harley, uncle of the Earl of Oxford, was one PARLIAMENT.
of the speakers. Walpole had borne hardly against
the earl as an enemy to the Protestant Succession, for being which
the peer had stood in some peril of his life, and had temporarily lost
his liberty. Mr. Harley said, he would refrain from acting as unjustly to
Walpole, (against whom there was no specific charge, only a general
accusation, without any proofs,) as Walpole had acted against his
nephew on mere suspicion;—and Mr. Harley walked out of the
House, without voting.
Walpole said of members of Parliament,—he
PARTY
would not declare who was corrupt, but that Shippen CHARACTERIS
was incorruptible. Coxe, in his Life of the Minister, TICS.
does not describe Shippen as a ‘Hanover Tory,’
running with the hare and holding with the hounds, of whom there
were many, but as an uncompromising Jacobite, one who repeated
among his Whig friends that there would be neither peace nor
content till the Stuarts were restored; and who confessed to his
confidants, that there were occasions on which he never voted in the
House till he had received orders from Rome;—that is, not from
Innocent, Benedict, or Clement, but from King James III. Shippen
used to say of Walpole and himself, ‘Robin and I understand each
other. He is for King George, I am for King James; but those men
with the long cravats, Sandys, Rushout, Gybbon, and others, only
want places, and they do not care under which King they hold them.’
This corresponds with John, Lord Harvey’s, account of parties under
George II. The Whigs were divided into patriots and courtiers, or
Whigs out, and Whigs in; the Tories into Jacobites and Hanover
Tories,—the first ‘thorough,’ the second joining with their opponents
when there was a promise of profit, personal or political. But their
prayer was something like that of the half-starved Highland chieftain:
‘Lord, turn the world upside down, that honest men may make bread
of it!’
At this time, there was much reiteration of the assurance of
Jacobitism being either dead or in despair. As a proof of the contrary,
on May 19th, the London ‘Champion’ referred to movements in the
Chevalier’s court at Rome. He had held several meetings with
Ecclesiastics, and also with laymen, ‘well-wishers to his interests.’
The ‘Champion’ could not explain the meaning of these two