Simulation of Dynamic Systems with MATLAB and

Simulink Harold Klee pdf download

https://textbookfull.com/product/simulation-of-dynamic-systems-
with-matlab-and-simulink-harold-klee/

Download more ebook from https://textbookfull.com

We believe these products will be a great fit for you. Click
the link to download now, or visit textbookfull.com
to discover even more!

Classical Feedback Control with Nonlinear Multi-Loop

Systems: With MATLAB® and Simulink® 3rd Edition Boris
J. Lurie

https://textbookfull.com/product/classical-feedback-control-with-
nonlinear-multi-loop-systems-with-matlab-and-simulink-3rd-
edition-boris-j-lurie/

Advanced Simulation of Alternative Energy: Simulation

with Simulink® and SimPowerSystems™ 1st Edition Viktor
M. Perelmuter

https://textbookfull.com/product/advanced-simulation-of-
alternative-energy-simulation-with-simulink-and-
simpowersystems-1st-edition-viktor-m-perelmuter/

Modeling and Simulation of Dynamic Systems 1st Edition

Robert L. Woods

https://textbookfull.com/product/modeling-and-simulation-of-
dynamic-systems-1st-edition-robert-l-woods/

High performance control of AC drives with MATLAB

Simulink Second Edition Jaros■aw Guzi■ski Haithem Abu
Rub Atif Iqbal

https://textbookfull.com/product/high-performance-control-of-ac-
drives-with-matlab-simulink-second-edition-jaroslaw-guzinski-
haithem-abu-rub-atif-iqbal/
Introduction to Modeling and Simulation with MATLAB and
Python 1st Edition Steven I. Gordon

https://textbookfull.com/product/introduction-to-modeling-and-
simulation-with-matlab-and-python-1st-edition-steven-i-gordon/

Design And Optimization Of Thermal Systems, With

MATLAB® Applications Yogesh Jaluria

https://textbookfull.com/product/design-and-optimization-of-
thermal-systems-with-matlab-applications-yogesh-jaluria/

An Introduction to Reservoir Simulation Using MATLAB

GNU Octave User Guide for the MATLAB Reservoir
Simulation Toolbox MRST Lie

https://textbookfull.com/product/an-introduction-to-reservoir-
simulation-using-matlab-gnu-octave-user-guide-for-the-matlab-
reservoir-simulation-toolbox-mrst-lie/

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/

Continuous signals and systems with MATLAB 3rd Edition

Taan S. Elali

https://textbookfull.com/product/continuous-signals-and-systems-
with-matlab-3rd-edition-taan-s-elali/
Simulation of Dynamic
Systems with
MATLAB® and Simulink®
Third Edition
Simulation of Dynamic
Systems with
MATLAB® and Simulink®
Third Edition

Harold Klee and Randal Allen

MATLAB® and Simulink® are a trademark of The MathWorks, Inc. and is used with permission. The MathWorks does
not warrant the accuracy of the text or exercises in this book. This book’s use or discussion of MATLAB® and Simu-
link® software or related products does not constitute endorsement or sponsorship by The MathWorks of a particular
pedagogical approach or particular use of the MATLAB® and Simulink® software.

CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742

© 2018 by Taylor & Francis Group, LLC

CRC Press is an imprint of Taylor & Francis Group, an Informa business

No claim to original U.S. Government works

Printed on acid-free paper

International Standard Book Number-13: 978-1-4987-8777-2 (Hardback)

This book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been
made to publish reliable data and information, but the author and publisher cannot assume responsibility for the
validity of all materials or the consequences of their use. The authors and publishers have attempted to trace the
copyright holders of all material reproduced in this publication and apologize to copyright holders if permission to
publish in this form has not been obtained. If any copyright material has not been acknowledged please write and let
us know so we may rectify in any future reprint.

Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or
utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including pho-
tocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission
from the publishers.

For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://
www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA
01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users.
For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been
arranged.

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for
identification and explanation without intent to infringe.

Library of Congress Cataloging-in-Publication Data

Names: Klee, Harold, author. | Allen, Randal, 1964- author.

Title: Simulation of dynamic systems with MATLAB and Simulink / Harold Klee and Randal Allen.
Description: Third edition. | Boca Raton : Taylor & Francis, CRC Press, 2018.
| Includes bibliographical references and index.
Identifiers: LCCN 2017035511| ISBN 9781498787772 (hardback) | ISBN 9781315154176 (ebook)
Subjects: LCSH: Computer simulation. | SIMULINK. | MATLAB.
Classification: LCC QA76.9.C65 K585 2018 | DDC 003/.3--dc23
LC record available at https://lccn.loc.gov/2017035511

Visit the Taylor & Francis Web site at

http://www.taylorandfrancis.com
and the CRC Press Web site at
http://www.crcpress.com
 To Andrew, Cassie, and in loving memory of their
mother and my devoted wife, Laura.

Harold Klee
To my wife, Christine, whose inner beauty radiates with the warmth of a
sunny day at the beach. Thank you for your everlasting love and support.

Randal Allen
Contents
Foreword ........................................................................................................................................ xiii
Preface.............................................................................................................................................. xv
About the Authors............................................................................................................................xix

Chapter 1 Mathematical Modeling................................................................................................ 1

1.1 Introduction........................................................................................................1
1.1.1 Importance of Models........................................................................... 1
1.2 Derivation of A Mathematical Model................................................................4
1.3 Difference Equations........................................................................................ 10
1.4 First Look at Discrete-Time Systems............................................................... 19
1.4.1 Inherently Discrete-Time Systems...................................................... 19
1.5 Case Study: Population Dynamics (Single Species)......................................... 22

Chapter 2 Continuous-Time Systems........................................................................................... 29

2.1 Introduction...................................................................................................... 29
2.2 First-Order Systems.......................................................................................... 29
2.2.1 Step Response of First-Order Systems................................................ 30
2.3 Second-Order Systems..................................................................................... 36
2.3.1 Conversion of Two First-Order Equations to a Second-Order
Model.................................................................................................41
2.4 Simulation Diagrams........................................................................................ 45
2.4.1 Systems of Equations.......................................................................... 51
2.5 Higher-Order Systems...................................................................................... 54
2.6 State Variables.................................................................................................. 57
2.6.1 Conversion from Linear State Variable Form to Single
Input–Single Output Form.................................................................. 62
2.6.2 General Solution of the State Equations............................................. 63
2.7 Nonlinear Systems............................................................................................66
2.7.1 Friction................................................................................................ 68
2.7.2 Dead Zone and Saturation................................................................... 71
2.7.3 Backlash.............................................................................................. 72
2.7.4 Hysteresis............................................................................................ 72
2.7.5 Quantization........................................................................................ 76
2.7.6 Sustained Oscillations and Limit Cycles............................................ 77
2.8 Case Study: Submarine Depth Control System................................................ 85

Chapter 3 Elementary Numerical Integration.............................................................................. 91

3.1 Introduction...................................................................................................... 91
3.2 Discrete-Time System Approximation of a Continuous First-Order
System........................................................................................................ 92
3.3 Euler Integration............................................................................................... 98
3.3.1 Explicit Euler Integration....................................................................99
3.3.2 Implicit Euler Integration.................................................................. 100
3.4 Trapezoidal Integration.................................................................................. 104

vii
viii Contents

3.5 Discrete Approximation of Nonlinear First-Order Systems.......................... 112

3.6 Discrete State Equations................................................................................. 116
3.7 Improvements to Euler Integration................................................................. 127
3.7.1 Improved Euler Integration............................................................... 127
3.7.2 Modified Euler Integration................................................................ 131
3.7.3 Discrete-Time System Matrices........................................................ 132
3.8 Case Study: Vertical Ascent of a Diver.......................................................... 146

Chapter 4 Linear Systems Analysis........................................................................................... 155

4.1 Introduction.................................................................................................... 155
4.2 Laplace Transform.......................................................................................... 155
4.2.1 Properties of the Laplace Transform................................................. 156
4.2.2 Inverse Laplace Transform................................................................ 163
4.2.3 Laplace Transform of the System Response..................................... 164
4.2.4 Partial Fraction Expansion................................................................ 166
4.3 Transfer Function............................................................................................ 173
4.3.1 Impulse Function............................................................................... 173
4.3.2 Relationship between Unit Step Function and Unit Impulse
Function............................................................................................. 173
4.3.3 Impulse Response.............................................................................. 175
4.3.4 Relationship between Impulse Response and Transfer Function....... 179
4.3.5 Systems with Multiple Inputs and Outputs....................................... 182
4.3.6 Transformation from State Variable Model to Transfer Function...... 190
4.4 Stability of Linear Time Invariant Continuous-Time Systems...................... 194
4.4.1 Characteristic Polynomial................................................................. 195
4.4.2 Feedback Control System..................................................................200
4.5 Frequency Response of LTI Continuous-Time Systems................................206
4.5.1 Stability of Linear Feedback Control Systems Based
on Frequency Response..................................................................... 216
4.6 z-Transform..................................................................................................... 222
4.6.1 Discrete-Time Impulse Function....................................................... 226
4.6.2 Inverse z-Transform........................................................................... 232
4.6.3 Partial Fraction Expansion................................................................ 233
4.7 z-Domain Transfer Function........................................................................... 242
4.7.1 Nonzero Initial Conditions................................................................ 243
4.7.2 Approximating Continuous-Time System Transfer Functions.......... 245
4.7.3 Simulation Diagrams and State Variables......................................... 250
4.7.4 Solution of Linear Discrete-Time State Equations............................ 256
4.7.5 Weighting Sequence (Impulse Response Function).......................... 261
4.8 Stability of LTI Discrete-Time Systems......................................................... 267
4.8.1 Complex Poles of H(z)....................................................................... 271
4.9 Frequency Response of Discrete-Time Systems............................................280
4.9.1 Steady-State Sinusoidal Response.....................................................280
4.9.2 Properties of the Discrete-Time Frequency Response Function........ 282
4.9.3 Sampling Theorem............................................................................ 287
4.9.4 Digital Filters.................................................................................... 293
4.10 Control System Toolbox.................................................................................300
4.10.1 Transfer Function Models................................................................. 301
4.10.2 State-Space Models...........................................................................302
4.10.3 State-Space/Transfer Function Conversion....................................... 303
Contents ix

4.10.4 System Interconnections................................................................... 305

4.10.5 System Response...............................................................................307
4.10.6 Continuous-/Discrete-Time System Conversion...............................309
4.10.7 Frequency Response.......................................................................... 311
4.10.8 Root Locus........................................................................................ 313
4.11 Case Study: Longitudinal Control of an Aircraft........................................... 319
4.11.1 Digital Simulation of Aircraft Longitudinal Dynamics.................... 333
4.11.2 Simulation of State Variable Model.................................................. 335
4.12 Case Study: Notch Filter for Electrocardiograph Waveform......................... 338
4.12.1 Multinotch Filters.............................................................................. 339

Chapter 5 Simulink.................................................................................................................. 349

5.1 Introduction.................................................................................................... 349
5.2 Building a Simulink Model............................................................................ 349
5.2.1 The Simulink Library....................................................................... 349
5.2.2 Running a Simulink Model............................................................... 353
5.3 Simulation of Linear Systems........................................................................ 357
5.3.1 Transfer Fcn Block............................................................................ 357
5.3.2 State-Space Block.............................................................................. 363
5.4 Algebraic Loops............................................................................................. 371
5.4.1 Eliminating Algebraic Loops............................................................ 373
5.4.2 Algebraic Equations.......................................................................... 375
5.5 More Simulink Blocks.................................................................................... 380
5.5.1 Discontinuities................................................................................... 385
5.5.2 Friction.............................................................................................. 386
5.5.3 Dead Zone and Saturation................................................................. 387
5.5.4 Backlash............................................................................................ 389
5.5.5 Hysteresis.......................................................................................... 389
5.5.6 Quantization...................................................................................... 391
5.6 Subsystems..................................................................................................... 394
5.6.1 PHYSBE............................................................................................ 395
5.6.2 Car-Following Subsystem................................................................. 396
5.6.3 Subsystem Using Fcn Blocks............................................................ 398
5.7 Discrete-Time Systems...................................................................................402
5.7.1 Simulation of an Inherently Discrete-Time System..........................403
5.7.2 Discrete-Time Integrator...................................................................406
5.7.3 Centralized Integration.....................................................................409
5.7.4 Digital Filters.................................................................................... 412
5.7.5 Discrete-Time Transfer Function...................................................... 414
5.8 MATLAB and Simulink Interface................................................................. 422
5.9 Hybrid Systems: Continuous- and Discrete-Time Components..................... 431
5.10 Monte Carlo Simulation................................................................................. 435
5.10.1 Monte Carlo Simulation Requiring Solution of
a Mathematical Model....................................................................... 439
5.11 Case Study: Pilot Ejection..............................................................................448
5.12 Case Study: Kalman Filtering........................................................................ 453
5.12.1 Continuous-Time Kalman Filter....................................................... 453
5.12.2 Steady-State Kalman Filter............................................................... 454
5.12.3 Discrete-Time Kalman Filter............................................................ 454
5.12.4 Simulink Simulations........................................................................ 455
x Contents

5.12.5 Summary...........................................................................................468
5.13 Case Study: Cascaded Tanks with Flow Logic Control.................................469

Chapter 6 Intermediate Numerical Integration.......................................................................... 475

6.1 Introduction.................................................................................................... 475
6.2 Runge–Kutta (RK) (One-Step Methods)........................................................ 475
6.2.1 Taylor Series Method........................................................................ 476
6.2.2 Second-Order Runge–Kutta Method................................................ 477
6.2.3 Truncation Errors.............................................................................. 479
6.2.4 High-Order Runge–Kutta Methods...................................................484
6.2.5 Linear Systems: Approximate Solutions Using RK Integration....... 486
6.2.6 Continuous-Time Models with Polynomial Solutions....................... 488
6.2.7 Higher-Order Systems....................................................................... 490
6.3 Adaptive Techniques......................................................................................500
6.3.1 Repeated RK with Interval Halving..................................................500
6.3.2 Constant Step Size (T = 1 min)......................................................... 505
6.3.3 Adaptive Step Size (Initial T = 1 min)............................................. 505
6.3.4 RK–Fehlberg..................................................................................... 505
6.4 Multistep Methods.......................................................................................... 512
6.4.1 Explicit Methods............................................................................... 513
6.4.2 Implicit Methods............................................................................... 515
6.4.3 Predictor–Corrector Methods........................................................... 518
6.5 Stiff Systems................................................................................................... 523
6.5.1 Stiffness Property in First-Order System.......................................... 524
6.5.2 Stiff Second-Order System............................................................... 526
6.5.3 Approximating Stiff Systems with Lower-Order Nonstiff
System Models.................................................................................. 529
6.6 Lumped Parameter Approximation of Distributed Parameter Systems.........546
6.6.1 Nonlinear Distributed Parameter System......................................... 550
6.7 Systems with Discontinuities......................................................................... 555
6.7.1 Physical Properties and Constant Forces Acting on
the Pendulum Bob............................................................................. 563
6.8 Case Study: Spread of an Epidemic............................................................... 573

Chapter 7 Simulation Tools........................................................................................................ 581

7.1 Introduction.................................................................................................... 581
7.2 Steady-State Solver......................................................................................... 582
7.2.1 Trim Function.................................................................................... 584
7.2.2 Equilibrium Point for a Nonautonomous System.............................. 586
7.3 Optimization of Simulink Models.................................................................. 596
7.3.1 Gradient Vector.................................................................................605
7.3.2 Optimizing Multiparameter Objective Functions
Requiring Simulink Models..............................................................607
7.3.3 Parameter Identification.................................................................... 610
7.3.4 Example of a Simple Gradient Search.............................................. 611
7.3.5 Optimization of Simulink Discrete-Time System Models................ 620
7.4 Linearization.................................................................................................. 630
7.4.1 Deviation Variables........................................................................... 631
7.4.2 Linearization of Nonlinear Systems in State Variable Form............ 639
Contents xi

7.4.3 Linmod Function............................................................................... 643

7.4.4 Multiple Linearized Models for a Single System..............................648
7.5 Adding Blocks to The Simulink Library Browser......................................... 659
7.5.1 Introduction....................................................................................... 659
7.5.2 Summary........................................................................................... 665
7.6 Simulation Acceleration................................................................................. 665
7.6.1 Introduction....................................................................................... 665
7.6.2 Profiler............................................................................................... 667
7.6.3 Summary...........................................................................................668
7.7 Black Swans....................................................................................................668
7.7.1 Introduction.......................................................................................668
7.7.2 Modeling Rare Events.......................................................................668
7.7.3 Measurement of Portfolio Risk......................................................... 669
7.7.4 Exposing Black Swans...................................................................... 673
7.7.4.1 Percent Point Functions (PPFs).......................................... 673
7.7.4.2 Stochastic Optimization.................................................... 673
7.7.5 Summary........................................................................................... 676
7.7.6 Acknowledgements........................................................................... 676
7.7.7 References......................................................................................... 676
7.7.8 Appendix—Mathematical Properties of the Log-Stable
Distribution........................................................................................ 676
7.8 The SIPmath Standard................................................................................... 677
7.8.1 Introduction....................................................................................... 677
7.8.2 Standard Specification....................................................................... 677
7.8.3 SIP Details......................................................................................... 678
7.8.4 SLURP Details.................................................................................. 678
7.8.5 SIPs/SLURPs and MATLAB............................................................ 679
7.8.6 Summary...........................................................................................680
7.8.7 Appendix........................................................................................... 681
7.8.8 References......................................................................................... 682

Chapter 8 Advanced Numerical Integration.............................................................................. 683

8.1 Introduction.................................................................................................... 683
8.2 Dynamic Errors (Characteristic Roots, Transfer Function)........................... 683
8.2.1 Discrete-Time Systems and the Equivalent
Continuous-Time Systems.................................................................684
8.2.2 Characteristic Root Errors................................................................ 687
8.2.3 Transfer Function Errors................................................................... 697
8.2.4 Asymptotic Formulas for Multistep Integration Methods................ 704
8.2.5 Simulation of Linear System with Transfer Function H(s)............... 708
8.3 Stability of Numerical Integrators.................................................................. 714
8.3.1 Adams–Bashforth Numerical Integrators......................................... 714
8.3.2 Implicit Integrators............................................................................ 722
8.3.3 Runga–Kutta (RK) Integration......................................................... 726
8.4 Multirate Integration...................................................................................... 738
8.4.1 Procedure for Updating Slow and Fast States:
Master/Slave = RK-4/RK-4.............................................................. 742
8.4.2 Selection of Step Size Based on Stability......................................... 743
8.4.3 Selection of Step Size Based on Dynamic Accuracy........................ 745
8.4.4 Analytical Solution for State Variables............................................. 748
xii Contents

8.4.5 Multirate Integration of Aircraft Pitch Control System.................... 750

8.4.6 Nonlinear Dual Speed Second-Order System................................... 753
8.4.7 Multirate Simulation of Two-Tank System....................................... 760
8.4.8 Simulation Trade-Offs with Multirate Integration............................ 763
8.5 Real-Time Simulation..................................................................................... 766
8.5.1 Numerical Integration Methods Compatible with
Real-Time Operation......................................................................... 769
8.5.2 RK-1 (Explicit Euler)......................................................................... 770
8.5.3 RK-2 (Improved Euler)...................................................................... 771
8.5.4 RK-2 (Modified Euler)...................................................................... 771
8.5.5 RK-3 (Real-Time Incompatible)........................................................ 771
8.5.6 RK-3 (Real-Time Compatible)........................................................... 772
8.5.7 RK-4 (Real-Time Incompatible)........................................................ 772
8.5.8 Multistep Integration Methods.......................................................... 772
8.5.9 Stability of Real-Time Predictor–Corrector Method........................ 774
8.5.10 Extrapolation of Real-Time Inputs.................................................... 776
8.5.11 Alternate Approach to Real-Time Compatibility: Input Delay......... 783
8.6 Additional Methods of Approximating Continuous-Time System Models..... 790
8.6.1 Sampling and Signal Reconstruction................................................ 790
8.6.2 First-Order Hold Signal Reconstruction........................................... 796
8.6.3 Matched Pole-Zero Method.............................................................. 796
8.6.4 Bilinear Transform with Prewarping................................................ 799
8.7 Case Study: Lego MindstormsTM NXT.......................................................... 803
8.7.1 Introduction....................................................................................... 803
8.7.2 Requirements and Installation..........................................................805
8.7.3 Noisy Model......................................................................................806
8.7.4 Filtered Model................................................................................... 810
8.7.5 Summary........................................................................................... 815

References...................................................................................................................................... 817
Index............................................................................................................................................... 821
Foreword
MATLAB is used for so many applications, it defies attempts at categorization. This book demon-
strates some of that interesting diversity.
As you read and use this book, you will find two kinds of knowledge. You may hope to find
insight to the use of MATLAB and Simulink. That hope will be richly fulfilled, I think. But you
should be mindful of another kind of knowledge; how others have solved problems. The rich col-
lection of examples and methods go far beyond the software toolset. These span different technical
disciplines and industries.
The authors show how modeling, simulation, and analysis gets done across a wide range of appli-
cations and industries, including financial markets. Work within and among various professional
societies further broaden this perspective. Their university work in teaching budding scientists and
engineers has honed the ability to make complexity approachable.
This book gives readers a chance to look outside their own discipline or industry, to collect ideas
from afar.
I hope your imagination will be fired while your modeling and simulation skills are being honed.

Steve Roemerman
Chairman & CEO of Lone Star Analysis
Dallas, TX

Simulation has come a long way since the days analog computers filled entire rooms. Yet, it is more
important than ever that simulations be constructed with care, knowledge, and a little wisdom, lest
the results be gibberish or, worse, reasonable but misleading. Used properly, simulations can give
us extraordinary insights into the processes and states of a physical system. Constructed with care,
simulations can save time and money in today’s competitive marketplace.
One major application of simulation is the simulator, which provides interaction between a model
and a person through some interface. The earliest simulator, Ed Link’s Pilot Maker aircraft trainer,
did not use any of the simulation techniques described in this book. Modern simulators, however,
such as the National Advanced Driving Simulator (NADS), cannot be fully understood without them.
The mission of the NADS is a lofty one: to save lives on U.S. highways through safety research
using realistic human-in-the-loop simulation. This is an example of the importance simulation has
attained in our generation. The pervasiveness of simulation tools in our society will only increase
over time; it will be more important than ever that future scientists and engineers be familiar with
their theory and application.
The content for Simulation of Dynamic Systems with MATLAB® and Simulink® is arranged
to give the student a gradual and natural progression through the important topics in simulation.
Advanced concepts are added only after complete examples have been constructed using funda-
mental methods. The use of MATLAB and Simulink provides experience with tools that are widely
adopted in industry and allow easy construction of simulation models.
May your experience with simulation be enjoyable and fruitful and extend throughout your careers.

Chris Schwarz, PhD

Iowa City, Iowa

xiii
Preface
In the first article of SIMULATION magazine in the Fall of 1963, the editor John McLeod pro-
claimed simulation to mean “the act of representing some aspects of the real world by numbers or
symbols which may be easily manipulated to facilitate their study.” Two years later, it was modified
to “the development and use of models for the study of the dynamics of existing or hypothesized
systems.” More than 40 years later, the simulation community has yet to converge upon a univer-
sally accepted definition. Either of the two cited definitions or others that followed convey a basic
notion, namely, that simulation is intended to reinforce or supplement one’s understanding of a sys-
tem. The definitions vary in their description of tools and methods to accomplish this.
The field of simulation is experiencing explosive growth in importance because of its ability to
improve the way systems and people perform, in a safe and controllable environment, at a reduced
cost. Understanding the behavior of complex systems with the latest technological innovations in
fields such as transportation, communication, medicine, aerospace, meteorology, etc., is a daunting
task. It requires an assimilation of the underlying natural laws and scientific principles that gov-
ern the individual subsystems and components. A multifaceted approach is required, one in which
simulation can play a prominent role, both in validation of a system’s design and in training of per-
sonnel to become proficient in its operation.
Simulation is a subject that cuts across traditional academic disciplines. Airplane crews spend
hours flying simulated missions in aircraft simulators to become proficient in the use of onboard sub-
systems during normal flight and possible emergency conditions. Astronauts spend years training in
shuttle and orbiter simulators to prepare for future missions in space. Power plant and petrochemical
process operators are exposed to simulation to obtain peak system performance. Economists resort
to simulation models to predict economic conditions of municipalities and countries for policymak-
ers. Simulations of natural disasters aid in preparation and planning to mitigate the possibility of
catastrophic events.
While the mathematical models created by aircraft designers, nuclear engineers, and economists
are application specific, many of the equations are analogous in form despite the markedly differ-
ent phenomena described by each model. Simulation offers practitioners from each of these fields
the tools to explore solutions of the models as an alternative to experimenting with the real system.
This book is meant to serve as an introduction to the fundamental concepts of continuous sys-
tem simulation, a branch of simulation applied to dynamic systems whose signals change over a
continuum of points in time or space. Our concern is with mathematical models of continuous-
time systems (electric circuits, thermal processes, population dynamics, vehicle suspension, human
physiology, etc.) and the discrete-time system models created to simulate them. The continuous sys-
tem mathematical models consist of a combination of algebraic and ordinary differential equations.
The discrete-time system models are a mix of algebraic and difference equations.
Systems that transition between states at randomly occurring times are called discrete-event
systems. Discrete-event simulation is a complementary branch of simulation, separate from con-
tinuous system simulation, with a mathematical foundation rooted in probability theory. Examples
of discrete-event systems are facilities such as a bank, a tollbooth, a supermarket, or a hospital
emergency room, where customers arrive and are then serviced in some way. A manufacturing plant
involving multiple production stages of uncertain duration to generate a finished product is another
candidate for discrete-event simulation.
Discrete-event simulation is an important tool for optimizing the performance of systems that
change internally at unpredictable times due to the influence of random events. Industrial engineer-
ing programs typically include a basic course at the undergraduate level in discrete-event simula-
tion. Not surprisingly, a number of excellent textbooks in the area have emerged for use by the
academic community and professionals.

xv
xvi Preface

In academia, continuous simulation has evolved differently than discrete-event simulation.

Topics in continuous simulation such as dynamic system response, mathematical modeling, differ-
ential equations, difference equations, and numerical integration are dispersed over several courses
from engineering, mathematics, and the natural sciences. In the past, the majority of courses in
modeling and simulation of continuous systems were restricted to a specific field like mechanical,
electrical, and chemical engineering or scientific areas like biology, ecology, and physics.
A transformation in simulation education is underway. More universities are beginning to offer
undergraduate and beginning graduate courses in the area of continuous system simulation designed
for an interdisciplinary audience. Several institutions now offer master’s and PhD programs in sim-
ulation that include a number of courses in both continuous and discrete-event simulation. A critical
mass of students are now enrolled in continuous simulation-related courses and there is a need for
an introductory unifying text.
The essential ingredient needed to make simulation both interesting and challenging is the inclu-
sion of real-world examples. Without models of real-world systems, a first class in simulation is little
more than a sterile exposition of numerical integration applied to differential equations.
Modeling and simulation are inextricably related. While the thrust of this text is continuous
simulation, mathematical models are the starting point in the evolution of simulation models.
Analytical solutions of differential equation models are presented, when appropriate, as an alterna-
tive to simulation and a simple way of demonstrating the accuracy of a simulated solution. For the
most part, derivations of the mathematical models are omitted and references to appropriate texts
are included for those interested in learning more about the origin of the model’s equations.
New and revised topics in the third edition are discussed in the later paragraphs dedicated to the
content of each chapter. However, certain changes appearing in the third edition apply to the entire
book, Chapters 1 through 8. These changes consist of the following:

1. All MATLAB script and function .m files have been renamed and the references to them
in the text have been changed to reflect the new file names. This eliminates the confusion
present in the second edition which retained the MATLAB file names from the first edition
based on the old system for naming chapter sections, figures, tables and exercise problems.
Simulink model .mdl file names remain unchanged since they do not contain chapter or
section references in their names.
2. With very few exceptions, nearly every graph generated in MATLAB has been redone
to improve its appearance in printed form. Specifically, all line plots and markers are
produced with a heavier weight, annotation and titles of most graphs have been changed
to better communicate the significance of each graph. Whilst the graphs are in black and
white in the text, every graph generated in MATLAB appears on screen in vivid colors to
enhance their appearance. Updated MATLAB and Simulink files are accessible from CRC
Press.
3. Simulink diagrams have been updated to be compatible with version R2016a of MATLAB/
Simulink. Diagrams with numerous Simulink blocks have been expanded to reveal the
details of each block and their interconnections.
4. Certain non-graph figures have been eliminated as a result of being unnecessary, while
others have been modified to be more informative.

Simulation is best learned by doing. Accordingly, the material is presented in a way that per-
mits the reader to begin exploring simulation, starting with a mathematical model in Chapter 1.
The notation used to represent discrete-time variables has been simplified in the new edition mak-
ing it easier to comprehend the difference equations developed to approximate the dynamics of
continuous-time systems. The latter part of Section 1.1 and all of Sections 1.2 through 1.5 have been
rewritten to better explain the underlying concepts.
Preface xvii

Chapters 2 and 4 remain basically unchanged. They present a condensed treatment of linear,
continuous-time, and discrete-time dynamic systems, normally covered in an introductory linear
systems course. The instructor can skip some or all of the material in these chapters if the students’
background includes a course in signals and systems or linear control theory.
Numerical integration is at the very core of continuous system simulation. Instead of treating the
subject in one exhaustive chapter, coverage is distributed over three chapters. Elementary numeri-
cal integration in Chapter 3 is an informal introduction to the subject, which includes discussion
of several elementary methods for approximating the solutions of first order differential equations.
Presentation of the topics in Chapter 3 has been completely revised. Much of the material in Chapter
3 from the second edition appears in a reorganized format while some material has been deleted
and new material added.
Simulink, from The MathWorks, is the featured simulation program because of its tight
­integration with MATLAB, the de facto standard for scientific and engineering analysis, and data
visualization software. Chapter 5 takes the reader through the basic steps of creating and running
Simulink models. Monte Carlo simulation for estimation of system parameters and probability of
events occurring in dynamic systems is covered. A new case study is introduced in Section 5.13
involving logically-controlled flows between two interconnected tanks.
Chapter 6 delves into intermediate-level topics of numerical integration, including a formal pre-
sentation of One-Step (Runga–Kutta) and multistep methods, adaptive techniques, truncation errors
and a brief mention of stability.
Chapter 7 highlights some advanced features of Simulink useful in more in-depth simulation
studies. Section 7.7 was added to demonstrate rare event modeling and portfolio risk measurement,
thereby exposing potential Black Swans as they may pertain to the financial markets. Section 7.8
was added to introduce SIPmath as a means for efficiently representing uncertainty as probability
distributions, enabling legacy and future simulation models to communicate with each other.
Chapter 8 is for those interested in more advanced topics on continuous simulation. Coverage
includes a discussion of dynamic errors, stability, real-time compatible numerical integration and
multirate integration algorithms for simulation of stiff systems.
The basic minimum requirement for anyone using this text is a first course in Ordinary
Differential Equations. An outline for a one-semester, preferably senior-level course in continuous
system simulation is subject to the individual requirements of the instructor as well as the prior
education of the students. As a starting point, some basic recommendations by the authors for a
one-semester course are:

Chapter 1 Chapter 2 Chapter 3 Chapter 4 Chapter 5 Chapter 6

Sections Sections Sections Sections Sections Sections
For students knowledgeable 1–4 1–8 1–8 Review of 1–3, 5, 6, 8, 1–5, 8
in linear systems theory 1–8 9, 11 or 13
For students not well-versed 1–4 1–8 1–8 1–5 1–3, 5, 6, 8, 1–4, 8
in linear systems theory 11 or 13

All remaining sections are appropriate for a second course in a two-semester sequence, either at
the senior, or more appropriately graduate level. The material in Chapters 7 and 8 is well suited as a
reference for practicing engineers and researchers involved in more advanced simulation endeavors.
The first and second editions of this text has been field-tested for nearly a decade. Despite numer-
ous revisions based on the scrutiny and suggestions of students and colleagues, some errors man-
age to go undetected. Further suggestions for improvement and revelations of inaccuracies can be
brought to the attention of the authors at aerospace321@outlook.com and klee.harold@gmail.com.
Numerous individuals deserve our thanks and appreciation for making the third edition pos-
sible. Thanks to Nora Konopka at Taylor & Francis/CRC Press for committing to the third edition
xviii Preface

and Kyra Lindholm, also with Taylor & Francis/CRC Press, for facilitating the transition from the
second to the third edition.

MATLAB® and Simulink® are registered trademarks of The MathWorks, Inc. For product
­information, please contact:

The MathWorks, Inc.

3 Apple Hill Drive
Natick, MA 01760-2098 USA Tel: 508 647 7000
Fax: 508-647-7001
E-mail: info@mathworks.com
Web: www.mathworks.com
About the Authors
Dr. Harold Klee received his PhD in systems science from Polytechnic Institute of Brooklyn in
1972, his MS in systems engineering from Case Institute of Technology in 1968, and his BSME
from The Cooper Union in 1965.
Dr. Klee has been a faculty member in the College of Engineering at the University of Central
Florida (UCF) since 1972. During his tenure at UCF, he has been a five-time recipient of the
­college’s Outstanding Teacher Award. He has been instrumental in the development of simula-
tion courses in both the undergraduate and graduate curricula. He is a charter member of the Core
Faculty, which is responsible for developing the interdisciplinary MS and PhD programs in simula-
tion at UCF. Dr. Klee served as graduate coordinator in the Department of Computer Engineering
from 2003 to 2006. Two of his PhD students received the prestigious Link Foundation Fellowship
in Advanced Simulation and Training. Both are currently enjoying successful careers in academia.
Dr. Klee has served as the director of the UCF Driving Simulation Lab for more than 15 years.
Under the auspices of the UCF Center for Advanced Transportation Systems Simulation, the lab
operates a high-fidelity motion-based driving simulator for conducting traffic engineering-related
research. He also served as editor-in-chief for the Modeling and Simulation magazine for three
years, a publication for members of the Society for Modeling and Simulation International.

Dr. Randal Allen is an aerospace and defense consultant working under contract to provide
6DOF aerodynamic simulation modeling, analysis, and design of navigation, guidance, and con-
trol ­systems. His previous experience includes launch systems integration and flight operations for
West Coast Titan-IV missions, propulsion modeling for the Iridium satellite constellation, and field
applications engineering for MATRIXx. He also chairs the Central Florida Section of the American
Institute of Aeronautics and Astronautics (AIAA).
Dr. Allen is certified as a modeling and simulation professional (CMSP) by the Modeling and
Simulation Professional Certification Commission (M&SPCC) under the auspices of the National
Training and Simulation Association (NTSA). He is also certified to deliver FranklinCovey’s Focus
and Execution track, which provides training on achieving your highest priorities.
Dr. Allen’s academic background includes a PhD in mechanical engineering from the University
of Central Florida, an engineer’s degree in aeronautical and astronautical engineering from Stanford
University, an MS in applied mathematics, and a BS in engineering physics from the University of
Illinois (Urbana–Champaign). He also serves as an adjunct professor at the University of Central
Florida in Orlando, Florida.

xix
 1 Mathematical Modeling

1.1 INTRODUCTION
1.1.1 Importance of Models
Models are an essential component of simulation. Before a new prototype design for an automobile
braking system or a multimillion dollar aircraft is tested in the field, it is commonplace to “test
drive” the separate components and the overall system in a simulated environment based on some
form of model. A meteorologist predicts the expected path of a tropical storm using weather models
that incorporate the relevant climatic variables and their effect on the storm’s trajectory. An econo-
mist issues a quantitative forecast of the U.S. economy predicated based on key economic variables
and their interrelationships with the help of computer models. Before a nuclear power plant operator
is “turned loose” at the controls, extensive training is conducted in a model-based simulator where
the individual becomes familiar with the plant’s dynamics under routine and emergency conditions.
Health care professionals have access to a human patient simulator to receive training in the rec-
ognition and diagnosis of disease. Public safety organizations can plan for emergency evacuations
of civilians from low-lying areas using traffic models to simulate vehicle movements along major
access roads.
The word “model” is a generic term referring to a conceptual or physical entity that resembles,
mimics, describes, predicts, or conveys information about the behavior of some process or system.
The benefit of having a model is to be able to explore the intrinsic behavior of a system in an eco-
nomical and safe manner. The physical system being modeled may be inaccessible or even nonexis-
tent as in the case of a new design for an aircraft or automotive component.
Physical models are often scaled-down versions of a larger system of interconnected compo-
nents as in the case of a model airplane. Aerodynamic properties of airframe and car body designs
for high-performance airplanes and automobiles are evaluated using physical models in wind tun-
nels. In the past, model boards with roads, terrain, miniaturized models of buildings, and land-
scape, along with tiny cameras secured to the frame of ground vehicles or aircraft, were prevalent
for simulator visualization. Current technology relies almost exclusively on computer-generated
imagery.
In principle, the behavior of dynamic systems can be explained by mathematical equations and
formulae, which embody either scientific principles or empirical observations, or both, related to
the system. When the system parameters and variables change continuously over time or space, the
models consist of coupled algebraic and differential equations. In some cases, lookup tables con-
taining empirical data are employed to compute the parameters. Equations may be supplemented by
mathematical inequalities, which constrain the variation of one or more dependent variables. The
aggregation of equations and numerical data employed to describe the dynamic behavior of a system
in quantitative terms is collectively referred to as a mathematical model of the system.
Partial differential equation models appear when a dependent variable is a function of two or
more independent variables. For example, electrical parameters such as resistance and capacitance
are distributed along the length of conductors carrying electrical signals (currents and voltages).
These signals are attenuated over long distances of cabling. The voltage at some location x mea-
sured from an arbitrary reference is written v(x, t) instead of simply v(t), and the circuit is modeled
accordingly.
A mathematical model for the temperature in a room would necessitate equations to predict
T(x, y, z, t) if a temperature probe placed at various points inside the room reveals significant vari-
ations in temperature with respect to x, y, z in addition to temporal variations. Partial differential

1
2 Simulation of Dynamic Systems with MATLAB® and Simulink®

equations describing the cable voltage v(x, t) and room temperature T(x, y, z, t) are referred to as
“distributed parameter” models.
The mathematical models of dynamic systems where the single independent variable is “time”
comprise ordinary differential equations. The same applies to systems with a single spatial indepen-
dent variable; however, these are not commonly referred to as dynamic systems since variations of
the dependent variables are spatial as opposed to temporal in nature. Ordinary differential equation
models of dynamic systems are called “lumped parameter” models because the spatial variation of
the system parameters is negligible or else it is being approximated by lumped sections with con-
stant parameter values. In the room temperature example, if the entire contents of the room can be
represented by a single or lumped thermal capacitance, then a single temperature T(t) is sufficient
to describe the room. We focus exclusively on dynamic systems with lumped parameter models,
hereafter referred to simply as mathematical models.
A system with a lumped parameter model is illustrated in Figure 1.1. The key elements are the
system inputs u1(t), u2(t), …, ur(t), which make up the system input vector u(t), the system outputs
y1(t), y2(t), …, yp(t), which form the output vector y(t), and the parameters p1, p2, …, pm constituting
the parameter vector p. The parameters are shown as constants; however, they may also vary with
time.
Our interest is in mathematical models of systems consisting of coupled algebraic and differ-
ential equations relating the outputs and inputs with coefficients expressed in terms of the system
parameters. For steady-state analyses, transient responses are irrelevant, and the mathematical mod-
els consist of purely algebraic equations relating the system variables.
An example of a mathematical model for a system with two inputs, three outputs, and several
parameters is

d2 d d
p1 y (t ) + p2 p3 y1 (t ) + p4 y1 (t ) + p5 y2 (t ) + p6 y2 (t ) = p7u1 (t )
2 1
(1.1)
dt dt dt

d p d
p8 y2 (t ) + 9 y2 (t ) + p11 y1 (t ) y2 (t ) = p12 u1 (t ) + p13u1 (t ) + p14u2 (t ) (1.2)
dt p10 dt

p16 y1p17 (t )
p15 y3 (t ) = (1.3)
y2 (t )

The order of a model is equal to the sum of the highest derivatives of each of the dependent
variables, in this case y1(t), y2(t), y3(t), and the order is therefore 2 + 1 + 0 = 3. Equation 1.1 is a
linear differential equation. Equation 1.2 is a nonlinear differential equation because of the term
involving the product of y1(t) and y2(t). The mathematical model is nonlinear due to the presence of
the nonlinear differential equation and the nonlinear algebraic equation (Equation 1.3). It is to be
borne in mind that it is the nature of the equations that determines whether a math model is linear

p1 p2 pm
u1(t) y1(t)
······· y2(t)
u2(t)
Input Output
u(t) = SYSTEM y(t) =

ur(t) yp(t)

FIGURE 1.1 A system with a lumped parameter model.

Mathematical Modeling 3

or nonlinear. An adjective such as linear or nonlinear applies to the mathematical model as opposed
to the actual system.
It is important to distinguish between the system being modeled and the model itself. The former
is unique, even though it may exist only at the design stage, while the mathematical model may
assume different forms. For example, a team of modelers may be convinced that the lead term in
Equation 1.1 is likely to be insignificant under normal operating conditions. Consequently, two
distinct models of the system exist, one third order and the other second order. The third-order
model includes the second derivative term to accurately reflect system behavior under unusual or
nontypical conditions (e.g., an aircraft exceeding its flight envelope or a ground vehicle performing
an extreme maneuver). The simpler second-order model ignores what are commonly referred to as
higher-order effects. Indeed, there may be a multitude of mathematical models to represent the same
system under different sets of restricted operating conditions. Regardless of the detail inherent in a
mathematical model, it nevertheless represents an incomplete and inexact depiction of the system.
A model’s intended use will normally dictate its level of complexity. For example, models for
predicting vehicle handling and responsiveness are different from those intended to predict ride
comfort. In the first case, accurate equations describing lateral and longitudinal tire forces are para-
mount in importance, whereas passenger comfort relies more on vertical tire forces and suspension
system characteristics.
Mathematical modeling is an inexact science, relying on a combination of intuition, experience,
empiricism, and the application of scientific laws of nature. Trade-offs between model complexity
and usefulness are routine. Highly accurate microclimatic weather models that use current atmo-
spheric conditions to predict the following day’s weather are of limited value if they require 48 h
on a massively parallel or supercomputer system to produce results. At the extreme opposite, overly
simplified models can be grossly inaccurate if significant effects are overlooked.
The difference between a mathematical model and a simulation model is open to interpretation.
Some in the simulation community view the two as one and the same. Their belief is that a math-
ematical model embodies the attributes of the actual system and simulation refers to solutions of the
model equations, albeit generally approximate in nature. Exact analytical solutions of mathematical
model equations are nonexistent in all but the simplest cases.
Others maintain a distinction between the two and express the view that simulation model(s)
originate from the mathematical model. According to this line of thinking, simulating the dynam-
ics of a system requires a simulation model that is different in nature from a mathematical model.
A reliable simulation model must be capable of producing numerical solutions in reasonably
close agreement with the actual (unknown) solutions to the math model. Simulation models are
commonly obtained from discrete-time approximations of continuous-time mathematical mod-
els. Much of this book is devoted to the process of obtaining simulation models in this way. More
than one simulation model can be developed from a single mathematical model of a system.
Stochastic models are important when dealing with systems whose inputs and parameters are
best modeled using statistical methods. Discrete event models are used to describe processes that
transit from one state to another at randomly spaced points in time. Probability theory plays a sig-
nificant role in the formulation of discrete event models for describing the movement of products
and service times at different stages in manufacturing processes, queuing systems, and the like. In
fact, the two pillars of simulation are continuous system simulation, the subject of this book, and
discrete event simulation.
There is a great deal more to be said about modeling. Entire books are devoted to properly iden-
tifying model structure and parameter values for deterministic and stochastic systems. Others con-
centrate more on derivation of mathematical models from diverse fields and methods of obtaining
solutions under different circumstances. The reader is encouraged to check the references section at
the end of this book for additional sources of material related to modeling.
Modeling is essential to the field of simulation. Indeed, it is the starting point of any simula-
tion study. The emphasis, however, in this book is on the presentation of simulation fundamentals.
4 Simulation of Dynamic Systems with MATLAB® and Simulink®

Accordingly, derivation of mathematical models is not a prominent component. For the most part,
the math models are taken from documented sources listed in the references section, some of which
include step-by-step derivations of the model equations. The derivation is secondary to a complete
understanding of the model, that is, its variables, parameters, and knowledge of conditions that may
impose restrictions on its suitability for a specific application.
Simulation of complex systems requires a team effort. The modeler is a subject expert responsi-
ble for providing the math model and interpreting the simulation results. The simulationist produces
the simulation model and performs the simulation study. For example, an aerodynamicist applies
principles of boundary layer theory to obtain a mathematical model for the performance of a new
airfoil design. Starting with the math model, simulation skills are required to produce a simulation
model capable of verifying the efficacy of the design based on numerical results. Individuals with
expert knowledge in a particular field are oftentimes well versed in the practice of simulation and
may be responsible for formulation of alternative mathematical models of the system in addition to
developing and running simulations.
A simple physical system is introduced in the next section, and the steps involved in deriving an
idealized math model are presented. In addition to benefiting from seeing the process from start
to finish, the ingredients for creating a simulation model are introduced. Hence, by the end of this
chapter, the reader will be able to perform rudimentary simulation.

1.2 DERIVATION OF A MATHEMATICAL MODEL

We begin our discussion of mathematical modeling with a simple derivation of the mathematical
model representing the dynamic behavior of an open tank containing a liquid that flows in the top
and is discharged from the bottom. Referring to Figure 1.2, the primary input is the liquid flow rate
Fl(t), an independent variable measured in appropriate units such as cubic feet per minute (volumet-
ric flow rate) or pounds per hour (mass flow rate). Responding to changes in the input are dependent
variables H(t) and F0(t) the fluid level, and flow rate from the tank.
Once the derivation is completed, we can use the model to predict the outflow and fluid level
response to a specific input flow rate F1(t), t ≥ 0. Note that we have restricted the set of possible
inputs to F1(t) and in the process relegated the remaining independent variables, that is, other vari-
ables which affect F0(t) and H(t), to second-order importance. Our assumption is that the eventual

F0(t)
F1(t) TANK
H(t)

FIGURE 1.2 Tank as a dynamic system with input and outputs.

F1(t) F1(t + Δt)

ΔH

H(t) H(t + Δt)

F0(t) F0(t + Δt)

Time: t Time: t + Δt

FIGURE 1.3 A liquid tank at two points in time.

Mathematical Modeling 5

model will be suitable for its intended application. It must be borne in mind that if extremely accu-
rate predictions of the level H(t) are required, it may be necessary to include second-order effects
such as evaporation and hence introduce additional inputs related to ambient conditions, namely,
temperature, humidity, air pressure, wind speed, and so forth.
The derivation is based on conditions of the tank at two discrete points in time, as if snapshots of
the tank were available at times “t” and “t + Δt,” as shown in Figure 1.3.
The following notation is used with representative units given for clarity:

F1(t): Input flow at time t, ft3/min

H(t): Liquid level at time t, ft
F0(t): Output flow at time t, ft3/min
A: Cross-sectional area of tank, ft2

At time t + Δt, from the physical law of conservation of volume,

V (t + ∆t ) = V (t ) + ∆V (1.4)

where
V(t) is the volume of liquid in the tank at time t
ΔV is the change in volume from time t to t + Δt

The volume of liquid in the tank at times t and t + Δt is given by

V (t ) = AH (t ) (1.5)

V (t + ∆t ) = AH (t + ∆t ) (1.6)

Equations 1.5 and 1.6 assume constant cross-sectional area of the tank, that is, A is independent
of H.
The change in volume from t to t + Δt is equal to the volume of liquid flowing in during the interval
t to t + Δt minus the volume of liquid flowing out during the same period of time. The liquid volumes
are the areas under the input and output volume flow rates from t to t + Δt as shown in Figure 1.4.
Expressing these areas in terms of integrals,

t +∆t t +∆t

∆V =
∫ F (t )dt − ∫ F (t )dt
t
1

t
0 (1.7)

The integrals in Equation 1.7 can be approximated by assuming F1(t) and F0(t) are constant over the
interval t to t + Δt (see Figure 1.4). Hence,

F1(t)
H(t)
F0(t)

ΔH

t t + Δt t t + Δt t t + Δt

FIGURE 1.4 Volumes of liquid flowing in and out of tank from t to t + Δt.
6 Simulation of Dynamic Systems with MATLAB® and Simulink®

t +∆t

∫ F (t )dt ≈ F (t )∆t
t
1 1 (1.8)

t +∆t

∫ F (t )dt ≈ F (t )∆t
t
0 0 (1.9)

Equations 1.8 and 1.9 are reasonable approximations provided Δt is small. Substituting Equations
1.8 and 1.9 into Equation 1.7 yields

∆V ≈ F1 (t )∆t − F0 (t )∆t (1.10)

Substituting Equations 1.5, 1.6, and 1.10 into Equation 1.4 gives

AH (t + ∆t ) ≈ AH (t ) + [ F1 (t ) − F0 (t )]∆t (1.11)

⇒ A  H (t + ∆t ) − H (t ) ≈  F1 (t ) − F0 (t ) ∆t (1.12)

 ∆H 
⇒ A  ≈ F1 (t ) − F0 (t ) (1.13)
 ∆ 

where ΔH is the change in liquid level over the interval (t, t + Δt). Note that ΔH/Δt is the average
rate of change in the level H over the interval (t, t + Δt). It is the slope of the secant line from pt A
to pt B in Figure 1.5.
In the limit as Δt approaches zero, pt B approaches pt A, and the average rate of change in H over
the interval (t, t + Δt) becomes the instantaneous rate of change in H at time t, that is,

∆H dH
lim = (1.14)
∆t → 0 ∆t dt

where dH/dt is the first derivative of H(t). From the graph, it can be seen that dH/dt is equal to the
slope of the tangent line of the function H(t) at t (pt A).

H
B
B′
Pt Coordinates
Tangent
ΔH A [t, H(t)]
A Δt B [t + Δt, H(t + Δt)]

B′ [t + Δt′, H(t + Δt′)]

t t + Δt′ t + Δt

FIGURE 1.5 Average rate of change ΔH/Δt as Δt gets smaller.

Mathematical Modeling 7

F0

1 1
F0 = H
R
R
H

FIGURE 1.6   A tank with out-flow proportional to fluid level.

Taking the limit as Δt approaches zero in Equation 1.13 and using the definition of the derivative
in Equation 1.14 give

 ∆H 
lim   = lim [ F1 (t ) − F0 (t )] (1.15)
∆t → 0  ∆t 
  ∆t →0

 dH 
⇒ A  = F1 (t ) − F0 (t ) (1.16)
 dt 

Since there are two dependent variables, a second equation or constraint relating F0 and H is
required in order to solve for either one given the input function F1(t). It is convenient at this point
to assume that F0 is proportional to H, that is, F0 = constant × H (see Figure 1.6). The constant of
proportionality is expressed as 1/R where R is called the fluid resistance of the tank. At a later point,
we will revisit this assumption.

1
F0 = H (1.17)
R

Equations 1.16 and 1.17 constitute the mathematical model of the liquid tank, namely,

dH 1
A + F0 = F1 and F0 = H
dt R

where F1, F0, H, and dH/dt are short for F1(t), F0(t), H(t), and (d/dt)H(t).
In this example, the model is a coupled set of equations. One is a linear differential equation and
the other is an algebraic equation, also linear. The differential equation is first order since only the
first derivative appears in the equation and the tank dynamics are said to be first order.
The outflow F0 can be eliminated from the model equations by substituting Equation 1.17 into
Equation 1.16 resulting in

dH 1
A + H = F1 (1.18)
dt R

Before a particular solution to Equation 1.18 for some F1(t), t ≥ 0 can be obtained, the initial tank
level H(0) must be known.
There are several reasons why an analytical approach to solving Equation 1.18 may not be the
preferred method. Even when the analytical solution is readily obtainable, for example, when the
system model is linear, as in the present example, the solution may be required for a number of dif-
ferent inputs or forcing functions. Recall from studying differential equations what happens when
the right-hand side of the equation changes. A new particular solution is required that can be time-
consuming, especially if the process is repeated for a number of nontrivial forcing functions.
Second, the input F1(t) may not even be available in analytical form. Suppose the input func-
tion F1(t) is unknown except as a sequence of measured values at regularly spaced points in time.
8 Simulation of Dynamic Systems with MATLAB® and Simulink®

An exact solution to the differential equation model is out of the question since the input is not
expressible as an analytic function of time.

EXERCISES
1.1 A system consists of two tanks in series in which the outflow from the first tank is the inflow
to the second tank as shown in the figure below.

a. Find the algebraic and differential equations comprising the mathematical model of the two
tank system. Assume both tanks are linear, i.e. the outflows are proportional to the liquid
levels and R1, R2 are the fluid resistances of the tanks.
b. Eliminate the flows F0(t) and F2(t) from the model to obtain a model in the form of two dif-
ferential equations involving the system input F1(t) and the tank levels H1(t) and H2(t).
c. Obtain the model differential equations when F0(t) and F2(t) are present instead of H1(t) and
H2(t).
d. The initial fluid levels in the tanks are H1(0) and H2(0) Suppose the flow in to the first tank
is constant, F1 (t ) = F1, t ≥ 0. Obtain expressions for H1(∞) and H2(∞), the eventual fluid
levels in Tanks 1 and 2. Do H1(∞) and H2(∞) depend on the initial fluid levels? Explain.
e. Find the ratio of tank resistances R1/R2 if H1(∞) = 2H2(∞).
f. Suppose the flow between the two tanks is reduced to zero by closing the valve in the line.
Show that this is equivalent to R1 = ∞ and determine the values of H1(∞) and H2(∞)
assuming the inflow to the first tank is still constant.
1.2 The two tanks in Exercise 1.1 are said to be non-interacting because the flow rate from the
first tank only depends on the fluid level in the first tank and is independent of the fluid level
in the second tank. Suppose the discharged fluid from the first tank enters the second tank at
the bottom instead of the top as shown in the figure below.

The flow between the tanks is now a function of the fluid levels in both tanks. The driving
force for the inter-tank flow is the difference in fluid levels and for the time being we can
assume the two quantities are proportional. That is,

H (t ) − H 2 (t )
F0 (t ) ∝  H1 (t ) − H 2 (t ) ⇒ F0 (t ) = 1
R12
Other documents randomly have
different content
 after a tough struggle up the precipitous and clayey bank
of the stream found a firm footing at the top. He was
obliged to walk till daylight before he found a shelter.
There was enough music, fortunately, in the violin–case for
immediate use; and although the company had lost all their luggage,
only one concert was given up. The second night after the accident
Ole Bull performed in Cincinnati as announced, but was obliged to
appear in his traveling dress.

In the summer of 1869 he gave his services to the great Peace

Jubilee in Boston, conducted by Mr. P. S. Gilmore, and this so
delayed his departure for Europe that he could make only a flying
visit to Norway. He returned again in the autumn. The winter
following, he gave concerts and traveled constantly. He was
everywhere warmly received by the public, both East and West.
In April, 1870, he sailed for Norway. The New York Tribune made the
following mention of his departure:—
“Herr Ole Bull, from the N. Y. Philharmonic Society,” was
the inscription upon a beautiful silken flag presented to
the great violinist, yesterday, on the deck of the United
States revenue cutter, which conveyed him from the
barge–office at the Battery to the steamship Russia. The
flag was the Norwegian colors, with the Star–Spangled
Banner inserted in the upper staff section. The committee
of presentation were Messrs. Hill, Schaad, and Doremus,
the latter being the spokesman. Ole Bull was accompanied
on board the steamer by quite a large number of friends,
among whom were Miss Adelaide Phillips, Miss Alide Topp,
Mrs. Belknap and sister, Dr. and Mrs. Doremus, General
Banks, Senator Conkling, Mr. F. S. Appleton, and others.
Dr. Doremus’s presentation speech, happily conceived,
was responded to in the warm–hearted and impulsive
manner peculiar to the artist, whose impulsiveness has
ever characterized the products of his genius, and whose
 warm–heartedness is known to hundreds who have
blessed him for his generosity. Senator Conkling and
General Banks also made appropriate speeches. As the
cutter left the steamer, the company waved together their
regrets and their farewells; and the form of the fine old
gentleman, bare–headed and swinging his hat, was seen
as long as forms could be distinguished in the distance.
This beautiful flag was, according to Ole Bull’s promise on its
acceptance, always carried in the 17th of May processions in Bergen,
and floated on the 4th of July.
A beautiful silver vase presented by the N. Y. Philharmonic Society
that season, a piece of silver plate given by the Young Men’s
Christian Association of New York city, for whom he had played, and
a very rich and beautiful gold crown given him in San Francisco,
were among the mementos which he carried home to Norway.
A still closer tie was soon to bind him to the United States, the
country which seemed already his by adoption. In Madison, Wis., in
the winter of 1868, Ole Bull first made the acquaintance which
resulted in his second marriage. He took a kindly interest in the
musical studies of his friend there, and later in New York. To others
this delightful relation of teacher, adviser, and friend seemed the only
one permissible; but he wrote: “Other than human powers have
decided my fate.... The sunbeams I shut out, but the sun itself I
could not annihilate.”
The marriage was delayed in deference to the wishes of others for
some months, but without resulting in a modification of their fears
concerning the disparity of years and other conditions. It was later
decided to have a private marriage. This was consummated in
Norway, and publicly announced and confirmed on the return to the
United States, three months later, in the autumn of 1870.
During the years 1867–69 Ole Bull had worked on his improvement
of the piano–forte. This attempt to build a piano outside a
manufactory would have been a very doubtful experiment even if
the principle of construction had not been itself an experiment. He
would not permit the use of any of the old means for strengthening
and sustaining the sounding–board, which necessarily in time
destroy its power of giving out a good quality of tone. His principal
effort was to sustain the board at the ends, leaving the sides free,
not permitting the board to be pierced for the insertion of screws to
unite the upper and lower frame–work, as is generally done. The
wooden strips employed to strengthen the great surface of the
board should, he thought, be made to help the tone as well, on the
principle of the bar in the violin; the whole to be so adjusted that
the wood might grow better with use and age.
The first instrument was made under very great disadvantages. The
workmen had it in hand while Ole Bull was absent on his concert
tour. A telegram would announce to him the breaking of the frame,
which he would try to remedy by suggestions sent by telegram or
letter. This instrument cost some $15,000. Not satisfied with this, he
commenced another, and met the same old difficulties in the frame,
or new ones quite as serious. John Ericsson learned of his trouble,
had him explain his idea, made a frame of the right weight and
strength, only insisting upon one condition,—the acceptance of it by
Ole Bull as a present. This second piano proved satisfactorily that
the theory was sound and practicable. Ole Bull had hoped to perfect
his invention and introduce the instruments to the public. This would
have given him the greatest satisfaction; for even the best pianos
give only for a short time a musical quality of tone, and necessarily
deteriorate from use, not only because of the wear in the action,
which can be replaced, but also on account of the destruction and
breaking down of the woody cells and tissues of the sounding–board
itself, under the improperly distributed strain brought to bear upon
it.
The opportunity never came to Ole Bull of doing more than to satisfy
himself that his theory could be realized. Two pianos only were
made; but these proved the possibility of doing away with the
objectionable features in the present construction of the instrument,
which, in time, must destroy the vitality of the sounding–board and
its power to produce a pure tone. It is well known that none but new
pianos are used by professional players; and an instrument which
has been used for any length of time positively vitiates a musical ear.
No friendly service ever touched Ole Bull more deeply than the
generous helpfulness of John Ericsson, whom he admired and loved.
It is with a sense of grateful recognition that the following letter is
now given in concluding this mention of the piano:—
New York, December 13, 1880.
My dear Mrs. Bull,—In adverting to the fact,—when I had
the pleasure of addressing you last week,—that everything
connected with Ole Bull’s memory is dear to me, I omitted
to refer to his admirable conception of securing the strings
of pianos to a separate frame, composed of metal, so
formed that it may be applied to any wooden stand more
or less ornamented. It was my privilege, often, to listen to
my lamented friend’s disquisitions relating to the violin,
showing his clear mechanical conceptions of the laws
which govern the construction of that most perfect of all
musical instruments. The great violinist possessed a
singularly accurate knowledge of the necessary relations
between the capability of resisting the tension of the
strings, and the elasticity requisite to admit of a perfectly
free movement of the sounding–board, and other parts of
the delicate structure, indispensable to produce infinitely
minute vibrations, the control of which by his master hand
created tones which enabled him to charm his hearers as
none of his rivals could.
I regard the independent metallic frame for holding the
strings of pianos as an invention which would do honor to
any professional mechanician; and I contemplate with
much satisfaction the circumstance that my departed
friend intrusted to me the construction of the first
specimen of his important improvement.
 I am, my dear Mrs. Bull,
Yours very sincerely, J. Ericsson.
Mr. Ericsson had previously written of Ole Bull: “So warm a heart and
so generous a disposition as his I have never known.” These words,
it may be truthfully said, express the sentiment and the judgment of
the violinist concerning the great engineer and inventor.
Professor H. Helmholtz, whose works on tone Ole Bull had carefully
studied, says of the violinist in a letter written in May, 1881:—
I saw that he was thoroughly well informed as to the
mechanical problems concerning the violin, which came in
question.... I was much impressed by his personal
character; he was at the same time so enthusiastic and so
intelligent, interested in all the great problems of
humanity.
The pleasantest incident of his concert tour in the season of 1871–
72 was a visit he made to his friend, Mr. William H. Seward, in
Auburn, N. Y. The two delightful days spent at Mr. Seward’s house
relieved the tedium of the whole trip.
The summer of 1872 he was in Norway, and in the fall returned
again to the United States, giving some seventy–five concerts, with
his usual success, during the winter. The last concerts announced
were given up, on account of the illness of his infant daughter.
In the spring he returned to Norway, and occupied for the first time
the new house which had been built at Lysö during his absence. The
winter following was mostly passed in Southern France. While on a
visit to Florence, he met again his old friend Prince Poniatowsky, and
also many others whom he had known nearly forty years before,
when he had been made an honorary member of the Philharmonic
Society of that city. Professor Sbolczis urged Ole Bull to permit his
friends to hear him, offering the aid of his own orchestra; and not
only the importunity of his friends, but the thought of again trying
his power over such an audience as the city of the Arno can offer,
tempted him to play. The hall was crowded, the tickets having been
taken by storm, and the performers suffered considerable
inconvenience therefrom. But to this an Italian readily
accommodates himself under such circumstances.
Brizzi, the leader of the “Orfeo,” now claimed that what Ole Bull had
done for Sbolczis should also be granted to his celebrated band of
performers, and offered to secure for the concert the largest hall in
the city, one that could hold nearly five thousand people. Ole Bull
consented, and the greatest enthusiasm prevailed. In the orchestra
sat old men who had accompanied Paganini—“And here he is again,”
they cried. The Corriere Italiano said of the concert:—
The Teatro Pagliano yesterday presented an imposing
scene. Every part was full to overflowing. The most
distinguished families and the ladies of the highest circles,
including the Grand Duchess Marie of Russia, whose
frequent applause showed how intensely she enjoyed the
music, were present at the concert; also Commodore
Peruzzi and wife, Princess Carolath, Princess Strozzi, and
an immense number of musical amateurs of our artistic
and aristocratic society.... The chief honors of the concert
were given to the Norse Paganini, the original, inspiring,
great violinist, Ole Bull, who kindly contributed his
valuable assistance, and paid his respects to the “Orfeo.”
His artistic nature prompted him to this graceful act of
fraternal courtesy, the value of which is extraordinary,
because he himself is extraordinary,—he, the king of all
violinists of the present time, the old formidable rival of
Paganini. In the fantasy of the “Nightingale” he gave us
an idea of his charming and wonderful skill.... In the
“Carnival of Venice,” the frantic dance of the notes
combined with the most graceful execution could not be
excelled. Both pieces produced a storm of deafening
applause. The audience desired the latter repeated, but,
instead of this, the musician gave them the celebrated
fantasy of Paganini on the no less celebrated song of
Paisiello, “Nel cor non più mi sento,” and in this we heard
 from Ole Bull the most secret beauties of song, the
sweetness of the flute, the transitions of the violin to the
viola, and to the sadness of the violoncello. In the
“Polacca Guerriera,” one of his own compositions, burst
forth the exciting and powerful notes of war. As a
composer he was graceful, wild, full of imagination,
feeling, and originality; as a performer he was mighty,
wonderful, indescribable. At this point a golden wreath
from the “Orfeo” was presented to Ole Bull, while the
audience applauded rapturously. The old and handsome
hero was visibly moved by the enthusiasm which he had
evoked.
He wrote of that evening:—
My violin did not fail me. I was never more thrilled by its
tone myself, and I cannot describe to you the pitch to
which the excitement ran, or the warmth of my reception.
I am so thankful that I have not disappointed my old
friends.
From Florence he went to Rome to see Liszt, and he found there
quite a colony of his countrymen.
The year following was spent in Norway. He had many vexations and
troubles at that time; but the hospitable cheer and ever–affectionate
welcome of his dear Lysekloster neighbors made these easier to
bear. During the summer he visited the North of Norway—the “Land
of the Midnight Sun”—for the first time. That trip was ever afterward
a source of delightful reminiscences, and, every successive winter, a
repetition of it for the coming summer would be suggested. He used
graphically to picture for us the morning which gave him his first
sight of the Lofoden Islands: the changeful, illusive beauty of sea
and sky through the long day, every feature and outline of isle and
coast being sharpened or softened by the play of brilliant light, now
and then dimmed by the fitful shadow of a fleecy cloud; and how,
towards midnight, all this culminated in a glory indescribable, the
warm prismatic colors flooding sky and sea, not followed by twilight,
but kindled anew by the beams of the rising sun suddenly shooting
athwart the sky, the warmth and glow at last giving way to the
tender flush of morning, and then to the white light of day. Then,
too, he would tell how his fellow–travelers, mostly Englishmen, were
ennuied and indifferent the first days of the trip, but as they
approached the North, and felt the influence of the champagne–like
atmosphere, how they gradually yielded themselves to the charm of
a new sensation, becoming social and even gay, some of them
climbing the shrouds like boys, and confiding to him that they had
never beheld a scene comparable for beauty, or felt such
exhilaration.
At one of the coast villages, which was so small that a concert had
not been thought of, they found the whole population turned out,
and, with the recruits from the surrounding country, making an
imposing array on the quay, which was decorated with evergreens
and flags. Surprised at such a welcome, the captain and passengers
good–naturedly agreed to wait till the concert could be given. As the
steamer, gay with flags from stem to stern, passed out of the fjord,
the mountains echoed with the repeated shouts and peals of
cannon.
His countrymen received him everywhere with the warm–hearted
hospitality they know so well how to dispense. But the dream of a
repetition of that trip was never realized. Three years later he went
as far north as Throndhjem, but mists and rain prevented a
continuation of the voyage.
The celebration of his birthday in 1876 deserves some mention here.
The following account is taken, with a few corrections, from Adolf
Ebeling’s “Bilder aus Kairo” (Stuttgart, 1878):—
On a beautiful September day in 1875, Ole Bull, by
invitation, visited Drotingholm, the charming summer
palace near Stockholm, and a favorite residence of the
queen. A fine park and inviting gardens separate it from
the busy world. Still it is only necessary to climb the
heights near by, and a glorious panorama is spread out
below. The bay of the Malar is seen filled with islands, and
the sea dotted with large and small sails; and, on the
other side, Stockholm itself, with its towers and palaces,
its forest of masts in the harbor, and the dark–green
mountains in the background.
The queen, now convalescent from a recent illness, was
giving no audiences, but had expressed a desire to see
the artist, and invited him to breakfast with the royal
family. While at table the queen happened to mention Ole
Bull’s “Saeterbesög,” her favorite composition. He was on
the point of telling her that it was on the programme for
one of his Stockholm concerts, but a glance from the king
checked him, as she was not yet permitted the excitement
of hearing music. At the same moment his majesty
laughingly suggested: “You are about to undertake a new
professional trip, Ole Bull. Perhaps you will visit Egypt.
What do you say to playing the ‘Saeterbesög’ on the top
of the Pyramid of Cheops? Nothing of the kind has ever
been done, and it seems to me that the idea might tempt
a virtuoso.” Ole Bull accepted the plan heartily, and the
king further suggested the 5th of February, the artist’s
birthday, as an appropriate time for the pyramid–concert.
On taking leave, the royal pair most warmly wished him all
success on his trip, especially the Egyptian portion of it.
After playing in Copenhagen, Berlin, Stettin, Hamburg,
Lübeck, and Bremen, awakening the same enthusiasm as
of old, we find him on the last day of January, 1876, in
Brindisi, taking passage for Egypt, accompanied by an
impressario and his pianist, the young Emil Bach. The trip
was a pleasant one. At early dawn on the 4th of February
Ole Bull stepped on shore at Alexandria, determined to
carry out the “royal idea,” as he called it. The same
evening he reached Cairo, where the Swedish consul, to
whom a telegram had been sent, met him at the station
and took him to his hospitable home. Early the next
morning, the 5th of February, several carriages were ready
before the villa of the consul, who had sent word in haste
to a few friends to invite them to share the trip; and about
ten o’clock the party found themselves at the Pyramids,
the goal of Ole Bull’s journey. Some of the company were
to ascend the Pyramid, and some to remain below; the
latter were in the minority, and were mostly ladies with a
few elderly gentlemen. The oldest of all was Ole Bull
himself, but he had already mounted the first ten blocks of
the huge structure alone and unaided. The powerful son
of the Norse mountains, to whom, in his boyhood, no crag
or peak was too high to be scaled, declared that, in spite
of the sixty–six years which he was celebrating, he should
be ashamed to have foreign arms help him to the top. But
the carrying up of his violin was a matter of great concern
and anxiety to him. Wrapped in a silk handkerchief, it was
intrusted to one of the most stalwart of the Bedouins, and
the bow, protected in the same way, was given to another
muscular fellow. After a quarter of an hour’s climb, Ole
Bull stood first of all upon the small world–famed plateau,
and greeted the Norse flag which the consul had had
raised there. Gradually the rest of the guests came also;
but from all sides clambered and crowded the Bedouins,
for the report had quickly circulated that a king of the
North had sent a player down to the Pyramids. Had they
known of it in Cairo, the tourists would certainly have
flocked thither in crowds.
Ole Bull had now taken his violin and given two powerful
strokes to assure himself that it was in good condition
after its dangerous journey. He then drew himself up to
his full height, and let his penetrating glance wander along
the horizon for a few moments, to scan the wonderful
scene below. At his right lay the valley of the Nile with its
bright green fields stretching into vanishing distances, the
waves of that broad, majestic stream gleaming like molten
silver; to the left lay great, boundless, golden deserts and
the Libyan Mountains; before him, at his feet, he had the
wide–spreading city of the khedive, with its minarets,
domes, and palm–gardens, all bathed in the brilliant
sunlight. Now he suddenly began to play a hymn of
praise, as it seemed; it was like a cry of joy to the Fates
who had vouchsafed him to stand there and to behold,
with his own eyes, the magnificent picture, the goal of so
many desires. Then he turned towards his home in the
North, and began his own mountain–song, the
“Saeterbesög”.... In the pure, calm air of this height,—the
loftiest of all structures made by human hands,—the tones
were so clear and penetrating, and at the same time so
powerful, that we felt ourselves moved as by magic power
and thrilled to our inmost souls; then, again, they wailed
like soft maiden voices—it was the home longing, the cry
to the Norway mountains; and then, again, we heard the
hero’s song of triumph, proud of his beautiful fatherland.
As Uhland makes the Münster Tower shake when the
young Goethe writes his name upon it,—

“Von seinem Schlage knittern

Die hellen Funken auf,
Den Thurm durchfährt ein Zittern
Vom Grundstein bis zum Knauf,”—

so those tones must have reverberated to the centre of

this royal grave of six thousand years within the Pyramids.
And that this beautiful, poetic moment should lack
nothing, there rose, just as the master gave the last
strokes of his bow, two majestic pelicans from the valley
of the Nile, which swept with the silvery sheen of their
wings towards the north, as if they would take the
message of this happy event to Ole Bull’s home. The
Bedouins, children of nature, who, during the playing, lay
 in a circle, motionless as fallen statues, sprang up when
the master had finished, as if electrified, and shouted
aloud and repeated “Allah! Allah!”
Thus Ole Bull had kept his promise. Returning to Cairo he
telegraphed the king, and the next forenoon received the
royal reply. The telegrams were as follows:—
“To King Oscar, Christiania, Norway,—
“According to my promise at Drotingholm, I played, on
this my sixty–sixth birthday, on the top of Cheops’
Pyramid, in honor of Norway and its beloved king, my
‘Saeterbesög.’”
The king answered:—
“I thank you heartily for your telegram, and the queen
and myself are rejoiced at all your successes.”
The artist’s singular journey to the Pyramids was soon
known in Cairo, and the khedive, during an audience,
complimented him on his courage and youthful strength.
Ole Bull gave a concert at the Opera House, and
harvested laurels, flowers, wreaths, and poems. His
“Tarentella” and “Carnival of Venice” were especially
admired. He played them with unwonted fire,
accompanied as he was by the admirable Italian
orchestra, under the leadership of his old friend Bottesini;
but his Norse “Saeterbesög” was not heard again.
Ole Bull returned to America in time to visit the great Centennial
Exhibition at Philadelphia. Offers from managers came at once,
desiring him to appear in concerts. He arranged with local managers
in the principal cities, and he never assumed again the burden of his
own concert company. His engagements were made on a new plan,
which relieved him of all responsibility outside his own
performances. The time had come when it seemed right that his
own comfort and pleasure should be the first consideration. He
played by contract never more than three times a week, and the
hours for travel were limited. It is but just to mention in this
connection that all the engagements filled from 1876 to 1880 were
satisfactory as to the management and courtesy of the gentlemen
who arranged for his concerts, and he had the pleasure of being
associated with many distinguished artists.
He first appeared in Music Hall, Boston, and after an absence of six
years the audience seemed most kindly determined to convince him
of their warm welcome. As he entered they rose to their feet, and
the applause was long continued. He gave that season, in Boston,
nine concerts to crowded houses, and the great desire of the public
to hear him so often was a surprise and a pleasure to him. One
occasion should be specially mentioned.
The great interest which he felt in the recognition of the Norsemen’s
discovery of America, not only in and for itself, but because it may
have given a hint to Columbus, who visited Iceland, made Ole Bull
desire to bring this subject before the people of Boston; and an
invitation signed by many gentlemen, prominent in social, political,
and literary circles, urging him to give one more concert, afforded
him an opportunity of doing this. The matter was therefore talked
over with friends, and subsequently a committee was appointed for
the erection of a monument to Leif Ericsson, commemorative of the
event.
It is hoped that this undertaking is to be carried out and completed
in the near future. Ole Bull’s friends know how earnest he was in
promoting it; and in Mr. Thomas G. Appleton, the chairman of the
committee, he found a liberal and enthusiastic helper. Many others,
beside his own countrymen, were interested in the plan, as will be
seen from the remarks made by Rev. Edward Everett Hale at the
concert, and reported below.
The concert was given in Music Hall, which was elaborately and
tastefully decorated for the occasion. Mr. Curtis Guild, who
introduced the artist, said:—
 I have been deputed by his Excellency Governor Rice, his
Honor Mayor Cobb, and other members of the committee,
with whom I have the honor to be associated, to present
to you one of whom it with truth may be said that he
needs no introduction to a Boston audience.... From the
commencement of his career in this country, when an
entranced audience listened to his wonderful melody at
the Melodeon, in May, 1844, to the present time, a period
of more than thirty years, the citizens of Boston, more
especially those of musical culture, have recognized Ole
Bull as a great musical artist, and one whose composition
and performance commanded the tribute that only true
genius can exact....
Ole Bull’s reply in acknowledgment of his reception was thus
reported in the Daily Advertiser:
Ladies and Gentlemen,—I see here among the audience stars
of the first magnitude. Why should they address me?
What am I that I can stand before this audience and
address such stars as these? I am but an atom of failure
in the universe; yet you are all united with me in that
failure in that you have indorsed me. You belong to me
and I belong to you. [Applause.] But to explain to you the
relation in which I stand here, and the names you see
about me [referring to the names Thorvald, Thorfinn, Leif,
and Washington], and how they are connected together, I
allow myself to say a few words. Everybody in this
audience knows that in the year 984 Bjarne sailed to meet
his father in Greenland. He was driven south, and after a
long voyage came at last to a beautiful land, which he
went back and reported he had seen. Fourteen years
after, in 1000, his son, Leif Ericsson, took his ships and
proceeded to seek that land; and he came along by
Newfoundland and the coast of Nova Scotia, and further
down to what he called Vineland, because he found vines
there. He remained the fall and winter there, and then
sailed back again to Iceland. Then two years after, his
brother Thorvald went over and met his death near
Martha’s Vineyard. In 1831 they discovered his body in
armor. His body was taken to Boston, and the armor was
analyzed, and they found it was the same metal that the
Norsemen had used in the ninth and tenth centuries. It
had the same ornaments they had used, so there could be
no doubt to whom it belonged. But, unfortunately, the
armor was lost by fire, and that calamity which we both
share together was an atom of fate that clings to us. You
lost the armor, but there was one who could give it to
posterity, and I say, almost to eternity, and that was our
illustrious star, Longfellow. [Applause.] It was given for
him to do it in his “Skeleton in Armor.”
Now what connection has the name of Thorfinn to
Washington? Washington not only belongs to the whole
world of the present generation, for that would be little to
say,—he belongs to all future generations. What you
educate is not for you alone, but it is for the whole world.
The name of Washington stands as the greatest pinnacle
of glory. It signifies liberty, it signifies every thing that
ennobles man. [Applause.] We find in the recent
discoveries concerning his ancestors, that they came over
in a ship, but that his ancestors’ name was Thorfinn. Well,
Thorfinn is a Norwegian name, and it is not very easy to
see how Thorfinn could be changed to Washington. But
we see every day that strangers come here, and after
some little time they change their names to some other
taken from the new surroundings. We see it in Norway,
often, that a man who has taken a new farm takes his
name, not from the ancient farm but the new farm. And
this is the connection of the ideas which prompted me to
come here to–night and have the honor to reply to the
memory of Washington and the memory of the
Revolution, which delivered not only America from
 oppression, but the whole world. Now I beg leave to take
my instrument in order to explain the rest.
After an account of the concert, during which the artist “seemed
stirred, by the sympathy of his hearers, to a sympathy, intensity, and
vividness of style unwonted even to himself,” the Advertiser
continues:—
Near the close of the performance, the Rev. E. E. Hale
rose in his place on the floor, and said he supposed it was
known to every person present that the distinguished
artist had spent almost the whole of his active life in
knotting those ties which connected his country with ours.
It was hoped that in some future time there would be
erected a physical memorial to the early discoverers of
whom he had spoken. It was the wish of those about him
[Mr. Hale], at whose request he spoke, that Boston should
not be behind in any expression of gratitude to him [Ole
Bull] for his work, as well as in expressing interest in our
Norse ancestors. He was sure he spoke the mind, not only
of the audience, but of all New England, when he spoke of
the interest with which he regarded his countrymen,
whom they regarded as almost theirs. He remembered,
although it was nearly forty years ago, when much such
an audience as he saw about him cheered and applauded
Edward Everett, when the early discoveries had just been
made, and, when in one of the last of his public poems he
expressed the wish that the great discoveries of Thorvald
might be commemorated by Thorvald’s great descendant,
the Northern artist, Thorwaldsen. The last words of that
poem as it died upon the ear were,—

“Thorvald shall live for aye in

Thorwaldsen.”
 He, the speaker, thought it was a misfortune for New
England that the great northern artist died before he could
accomplish this wish. But New Englanders had never
forgotten it, and had never forgotten their Norse
ancestors. It was an enterprise which ought to engage
Massachusetts men,—the preservation of a physical
memorial of Thorvald, Leif, and Thorfinn,—and he
suggested that the committee which had arranged the
meeting should become a committee of New England, in
conjunction with Mr. Appleton, to take this matter in
special charge. Mr. Hale put a motion to this effect, and it
was carried, and the committee constituted.
The holidays of that year were memorable. Thanksgiving found Ole
Bull at the home of Professor Horsford in Cambridge, and he shared
the Christmas cheer of Craigie House with the beloved poet; while
on New Year’s Eve, he watched the old year out and the new in with
a few friends in the library of James T. Fields.
In the early months of that year he made an engagement for thirty
concerts with his friend, Mr. Maurice Strakosch. He first gave four
concerts at Steinway Hall, in New York, and had the honor of being
assisted by Madame Essipoff. The New York Herald, of March 4,
1877, said:—
Age seems to have been contented with scattering snow
upon his head, leaving untouched the fire of his genius.
The lithe and agile figure has lost none of its elasticity,
and the nerves are as steady as in the noon–day of life.
There is still the same nervous delicacy of touch and
precision of execution, which, in the years gone by,
charmed two continents and led the people of many lands
in pleasant bondage after the car of the enchanting
musician. The triumphs of to–day are destined to be no
less great than those obtained in the past. Ole Bull, like a
prince that had wandered from his own land and returned
after a long absence, is restored to the throne he had
 abandoned by a delighted people; the memory of his
great feats is not forgotten.
The New York audiences were remarkable for their size, as were
those in the cities he visited east, south, and west. Miss Emma
Thursby was now associated with him, and this first acquaintance
with that great and charming artist resulted in a warm, personal
friendship. The very last engagement Ole Bull filled was in
conjunction with Miss Thursby, who was, he hoped, to accompany
him to Norway, where he desired to present her himself to his
countrymen; but, overtaken by the last fatal illness in England, he
was so weak on reaching home that waiting friends were advised of
the impossibility of even a visit. The following summer, Miss Thursby
paid a most loving and tender tribute to her departed friend, by
visiting Ole Bull’s birthplace to add her generous gift to the memorial
his countrymen are to erect, and won for herself the admiration and
the affection of the Northmen, as she had that of Ole Bull before.
The press notices are reminders of the large audiences of 1876, and
the pleasure manifested at the efforts of the artists. In some places,
—in Brooklyn and Philadelphia, for instance,—not only the audience–
room was filled to overflowing, but hundreds of extra seats were
placed upon the stage, while large numbers were turned away.
Describing such a scene, the Philadelphia Press said:—
 And when Ole Bull appeared, what a right royal shout of
welcome he received! The wind waves almost could be
felt, and the applause could be heard blocks away....
There is yet the delicacy of touch and the wondrous
power that tells of genius without loss of lustre, and of the
marvelous sympathetic instrumentation of his brightest
years. Nardini’s Sonata was the first selection, and it was
superbly played.... The concert throughout was a most
brilliant success, and may fairly rank as one of the great
events in the musical history of Philadelphia.
To the last, these great audiences welcomed Ole Bull in the
Philadelphia Academy, under the able management of Mr. Pugh.
The entire season was to the violinist an enjoyable one, since he was
in perfect health and condition for his work. He had time to see his
friends and to visit them, playing, as he did, at intervals only. He
renewed many of the intimacies he had formed on his first visits to
this country, which incessant work in previous years had compelled
him, against his wish, to drop; and he found that he was not only
not forgotten, but even better loved than he had thought.
After finishing his series of concerts, and only three days before his
departure for Europe, a request came, signed by many distinguished
names in New York, asking for still another appearance. He played,
on the evening before he sailed, in Booth’s Theatre, to a very full
house, the stage and boxes being occupied by his friends. The
artists were all inspired to do their best, and, after the “Ave Maria” of
Gounod had been sung by Miss Thursby with Ole Bull’s obligato, the
audience seemed more determined than ever to recall the artist and
to demand familiar airs. After the sixth encore he made a little
speech, but it was not until he had responded to the tenth recall, for
which he had given, “We won’t go home till morning,” that the
audience, with the last waving of handkerchiefs and cheers, said
good–by. Later, at a little supper at the hotel, the paper written by
Dr. Crosby (printed in the Appendix to this volume) was produced
and read for the first time—a surprise to Ole Bull. This labor of love
so generously given when it meant hours of toil to an already
overworked physician, with illness at home, was the crowning
kindness of the many showered upon him by loving friends. No man
ever filled a larger place in Ole Bull’s heart than Professor R. Ogden
Doremus, who was present, and had made Dr. Crosby known to the
artist; and how he wished that he might carry the two with him to
his own beloved Norway, when they bade him good–by the next
morning on the deck of the steamer! He felt a concern for Dr.
Crosby’s health which proved only too well founded, for in June
came the sad news of his death, the result of overwork.
The Atlantic voyage was a rest in itself, for sea–sickness was
unknown to Ole Bull, and he was a good person to cheer others who
were unhappy on shipboard. The inevitable concert would be given
the last evening of the voyage, and often on very stormy passages
he had played to make others forget the fear and discomfort of the
hour.
The summer of 1877 was quietly spent in Norway, and the winter
was passed on the Continent in travel. In Brussels, among the
friends who called on the artist was Vieuxtemps, then suffering from
the effects of a paralytic stroke. When he would himself try Ole Bull’s
instrument his poor, numb hand could not obey his will; and at last
he exclaimed, as he handed it back, “’Tis no use, I cannot command
my fingers!” His talk concerning style, composition, and virtuosity
was most interesting, as were also the incidents of travel which he
recalled. He said of Mendelssohn and Schumann, “Ils sont virtuoses
parce qu’ils connaissent à fond leur art; ils sont virtuoses parce qu’ils
sont de grands poètes. Ils le sont parce qu’ils ont le génie.
Virtuosité, génie, sont deux termes à peu près synonymes, deux
notions presque identiques.” Vieuxtemps’s indignation at the
constant abuse of the term may be imagined.
Ole Bull had planned to go to Italy and Sicily for the winter, as he
had long desired to revive and live over again the memories of his
first visits to that sunny clime; but on his reaching Vienna in January,
the time from week to week passed so delightfully, and old friends
were so cordial, that all thought of going further south was given up.
He had not intended to appear in public, and did so on a few
occasions only. The general interest taken in his visit there and
elsewhere by the press and the people surprised him. He busied
himself while in Vienna with repairing violins for friends, with so
much success that his acquaintances would urge him to direct the
work on their instruments, which they wished adjusted according to
his method, and he could seldom refuse such a request. It
sometimes seemed as if he were happier at work on an old decrepit
fiddle, which he saw could be restored, than when playing on his
own superb instrument.
While in that city he celebrated with friends the birthday of Madame
Mathilde Marchesi. Among the guests at the musical party was
Madame Christine Nilsson, and the hostess told with pride of a
telegram she had just received from her favorite pupil, Gerster,
whose brilliant success in America she predicted.
Nilsson recalled how when a little girl she had been admitted by the
stage–door to one of Ole Bull’s concerts in Sweden, and how, while
the artist stood talking to a friend, she had asked to look at his
violin, which he left in her hands when called away for a moment.
On venturing to draw the bow she found to her delight that it
“almost played itself.”
The courtesy of Mr. Joseph Hellmesberger, Kapelmeister, with whom
the artist played in public, was an incident of his stay which he
remembered with much pleasure.
A visit was made to Pesth, where each day was sure to bring a
charming note or thoughtful message from Liszt, whose kindly face
often looked in upon his friend. The following missive was sent on
the morning before Ole Bull left the city:—
Mardi, 19 Février.
Mon Illustre,—Je vous prie amicalement de passer la
soirée d’aujourd’hui avec Madame Ole Bull chez votre
vieux collègue et dévoué ami,
 Franz Liszt.
On se réunit à 9 heures.
(Il n’y aura pas de “violon” ni même de piano.)
At midnight, however, the violin was sent for at Liszt’s request, and
not till after two o’clock in the morning did the company disperse.
The walk to the hotel along the fine river embankment in the brilliant
starlight, with the wonderful tones still sounding in one’s brain,
cannot be forgotten. After a brilliant improvisation on the same
motives which Ole Bull had chosen for the violin, Liszt had closed
with a dreamy, tender nocturne.
The master’s real interest in his friend’s work was shown by his
chiding him with warmth for the state of his musical manuscripts
which he insisted on looking through, and which he earnestly
entreated him to prepare for publication.
The following note brought Liszt himself in answer, and the last
adieux were said:—
Illustre Ami,—En partant, le courage de vous remercier de
vive voix de votre hospitalité princière me fait défaut. Vos
précieux conseils, inséparables compagnons de votre âme,
inséparables souvenirs de lumière de notre réunion après
tant d’années d’épreuves, sont gage sincere d’amitié—
gage et promesse en même temps! Ma chère femme,
toute émue sous l’influence de votre génie si gracieux, me
prie d’exprimer sa reconnaissance, et permettez–moi de
souhaiter que l’Etre Suprême vous rende aussi heureux
que possible; voila ce que désire ardemment
Votre dévoué admirateur et ami,
F. Liszt Ole Bull.
A month at the baths in Wiesbaden, where friends made the stay
most pleasant, and a summer in the Norse home, followed. That
summer home–coming was always a delight to Ole Bull.
The grand old mountains, weird and forbidding in the early spring
storms but glorified by the Northern summer, called him, and he
heard. The beloved Lysekloster valley, whose wooded slopes
commanded the fjord, the sea, the islands, and the great range of
the Hardanger; the road his childish steps had trod, winding its way
down to the sea from the church of the old ancestral home, at every
turn giving a picturesque glimpse of lake or cliff; the path shaded by
birches and maples, and the fields fragrant with wood violets and
lilies–of–the–valley; the cottagers at work, the red jackets of the
women and caps of the men giving a dash of color here and there;
the workers shouting their respectful “welcome back” as he hurries
down to the boat waiting to carry him to his own enchanted isle,—
this was the picture which lured him every spring, and when realized
gave him the happiest moment of the year.
From his eighth year he had loved Lysekloster, and often said that he
would choose that of all places in the world for his home. In 1872
the estate was divided on the death of the owner, and the mansion
itself came into the possession of a friend and schoolmate of Ole
Bull, who, at his suggestion, bought the island opposite and decided
to make a new home for himself there, thus fulfilling his boyish
dream. The island had hardly been explored or its rocky shores
visited by those living on the main land. Its tall pines had grown and
rocked in the winds alone; its sod, except in one little spot, had
never been broken; its lakes mirrored only the stars and clouds.
From the foundation of the cloister on the main land seven hundred
years before, it had been noted only as furnishing some of the
largest and finest trees in the neighborhood for building purposes;
but fortunately it was still well wooded with pine.
A visitor would find Ole Bull while there interested chiefly in the
subject of drainage, the care of trees, and the grading of roads and
paths, which he had himself laid out; strewn with white sea–shells
they could be seen from a height circling the lakes and opening up
the island in every direction.
The little steamer gliding into the fjord at breakfast time seldom
failed to bring one or more guests from town. Old or young, they
were taken on walks of exploration about the island, and even the
oldest were sure to catch the enthusiasm of their host. If fretted by
a guidance which did not spare them a short cut over rough ground,
down ravines and along never–ending turns of paths, so confusing
that it was impossible to return alone, they forgave him later, when
in his music they learned what such a walk had been to him.
The autumn days were the days for study. The guests had then
gone, and sometimes a week of storm would succeed the brightest
sunshine, and dark nights suddenly replace the long twilight. The
walks then were to the farthest points out towards the sea, where
the ocean symphony sounded loudest; the paths must also be
explored to protect them from the miniature waterfalls overleaping
their proper channels, or to save tree or shrub from the flood which
threatened its destruction; but the fiercest storms could not disturb
the tranquil lakes guarded by the pine–clad cliffs which furnished a
quiet retreat on those wild walks. Then came the contrast of the
cozy room brightly lighted, and the tempting delicacy, or, better still,
the old–fashioned dish reserved for such times by the faithful
Martha.[23] How he enjoyed it all! The music–room cheerful with
wood–fires and candles, while the storm without promised seclusion,
tempted him to do the best work, often far into the night. When the
fire and candles had burned low, and the shadows seemed the
intruding spirits of the storm, then the notes would be thrown aside,
and that wonderful instrument, a soul in the hand of its master,
would voice the tempest outside and the peace within. Never did the
picture of him drawn by Longfellow in the “Tales of the Wayside Inn”
seem more strikingly true than in that room and at that hour:—
 Before the blazing fire of wood
Erect the rapt Musician stood;
And ever and anon he bent
His head upon his instrument,
And seemed to listen, till he caught
Confessions of its secret thought,—
The joy, the triumph, the lament,
The exultation and the pain;
Then, by the magic of his art
He soothed the throbbings of its heart
And lulled it into peace again.

The exquisite pictures of the artist which appear later in the poem,
breathing the Northern tradition and spirit, follow naturally here:—
Last the Musician, as he stood
Illumined by that fire of wood;
Fair–haired, blue–eyed, his aspect blithe,
His figure tall and straight and lithe,
And every feature of his face
Revealing his Norwegian race;
A radiance, streaming from within,
Around his eyes and forehead beamed;
The angel with the violin,
Painted by Raphael, he seemed.
He lived in that ideal world
Whose language is not speech, but song;
Around him evermore the throng
Of elves and sprites their dances whirled;
The Strömkarl sang, the cataract hurled
Its headlong waters from the height;
And mingled in the wild delight
The scream of sea–birds in their flight,
The rumor of the forest trees,
The plunge of the implacable seas,
The tumult of the wind at night,
Voices of eld, like trumpets blowing,
Old ballads and wild melodies
Through mist and darkness pouring forth,
Like Elivagar’s river flowing
Out of the glaciers of the North.
The instrument on which he played
Was in Cremona’s workshops made,
By a great master of the past,

Ere yet was lost the art divine;

Fashioned of maple and of pine,
That in Tyrolian forests vast
Had rocked and wrestled with the blast;
Exquisite was it in design,
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

textbookfull.com