MATLAB® Essentials

A First Course for Engineers and Scientists

MATLAB® Essentials
A First Course for Engineers
and Scientists

William Bober
MATLAB ® is 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® software
or related products does not constitute endorsement or sponsorship by The MathWorks of a particular
pedagogical approach or particular use of the MATLAB ® 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: 978-1-138-56328-5 (Hardback)

International Standard Book Number-13: 978-1-138-03237-8 (Paperback)

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, transmit-
ted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented,
including photocopying, 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.
Visit the Taylor & Francis Web site at
http://www.taylorandfrancis.com

and the CRC Press Web site at

http://www.crcpress.com
Contents

Preface ......................................................................................................................ix
Acknowledgments .............................................................................................. xiii
Author .....................................................................................................................xv

1. Computer Programming with MATLAB® for Engineers.......................1

1.1 Introduction .........................................................................................1
1.2 Computer Usage in Engineering ......................................................1
1.3 Mathematical Model...........................................................................2
1.4 Computer Programming ...................................................................3
1.5 Components of a Typical Desktop/Laptop Computer System..... 3
1.6 Overview of Programming Languages........................................... 5
1.7 Why MATLAB? ...................................................................................5
1.8 Programming Methodologies ...........................................................6
1.9 MATLAB Programming Language .................................................6
1.10 Building Blocks in Writing a Computer Program .........................7
1.11 Example Programs ............................................................................. 7

2. MATLAB® Fundamentals .............................................................................9

2.1 Introduction .........................................................................................9
2.2 MATLAB’s Desktop .......................................................................... 10
2.3 Constructing a Script (Program) in MATLAB.............................. 13
2.4 Variable Names and Types .............................................................. 18
2.5 Assignment Operator ....................................................................... 19
2.5.1 Arithmetic Operators .......................................................... 20
2.6 Some MATLAB Features, Commands, Special Items,
and Built-in Functions...................................................................... 21
2.6.1 Trigonometric and Other Useful Functions .................... 21
2.6.1.1 Special Values ....................................................... 23
2.6.1.2 Trigonometric Functions .................................... 24
2.6.1.3 Inverse Trigonometric Functions....................... 24
2.6.1.4 Exponential, Square Root, and Error
Functions ...............................................................25
2.6.1.5 Complex Numbers ............................................... 26
2.6.2 Other Special Values ........................................................... 26
2.6.2.1 Other Useful MATLAB Functions .................... 26
2.6.2.2 Colon Operator (:) ................................................. 28
2.6.2.3 Preallocation of a Matrix .................................... 29

v
vi Contents

2.7 MATLAB Output ..............................................................................30

2.7.1 The disp Command ....................................................... 31
2.7.2 The fprintf Command ................................................ 31
2.7.3 Printing to a File .............................................................. 32
2.8 Simple Plot Commands ...................................................................34
2.8.1 Linear Plot.........................................................................34
2.9 Loops .................................................................................................. 36
2.9.1 The for Loop ................................................................... 36
2.9.2 The While Loop ..............................................................43
2.10 Input.................................................................................................... 50
2.10.1 The Load Command ....................................................... 50
2.10.2 The dlmread Command ................................................ 52
2.10.3 fscanf Command .......................................................... 52
2.10.4 The input Command.....................................................54
2.11 More on MATLAB Graphics ........................................................... 55
2.11.1 The figure Command .................................................. 55
2.11.2 Multiple Plots ................................................................... 57
2.11.3 The hold on Command ................................................ 59
2.11.4 Plotyy Command .......................................................... 62
2.11.5 The subplot Command ................................................63
2.11.6 Bar Charts .........................................................................63
2.11.7 Pie Charts..........................................................................65
2.11.8 Greek Letters and Mathematical Symbols ................... 68
2.11.9 Interactively Annotating Plots ....................................... 69
2.11.10 Saving Plots ...................................................................... 69
References .......................................................................................................80

3. Conditional Operators, Built-in Functions with Vector

Arguments, MATLAB®’s Interp1 Function, and Some Scalar and
Vector Operations ......................................................................................... 81
3.1 Introduction ....................................................................................... 81
3.2 Conditional Operators and Alternate Paths ................................. 81
3.2.1 The if Command Provides Two Alternate Paths ......81
3.2.2 The if-elseif-else Command Provides More
than Two Alternate Paths ...............................................83
3.2.3 The break Command ....................................................85
3.2.4 The switch Command .................................................. 89
3.2.5 MATLAB’s menu Function ............................................. 90
3.3 Working with Built-in Functions with Vector Arguments ......... 92
3.4 MATLAB’s interp1 Function ....................................................... 93
3.5 Some Scalar and Vector Operations ............................................... 96
3.5.1 Addition of a Scalar and a Vector.................................. 96
3.5.2 Multiplication of a Scalar Times a Vector..................... 96
3.5.3 Addition and Subtraction of Two Vectors of the
Same Length ..................................................................... 96
Contents vii

3.5.4 Element-by-Element Operations.................................... 96

3.5.5 Operation of Two Vector Functions .............................. 98

4. Self-Written Functions and MATLAB ®’s fminbnd Function ........... 105

4.1 Introduction ....................................................................................... 105
4.2 Self-Written Function........................................................................ 105
4.3 Anonymous Functions ..................................................................... 110
4.4 MATLAB’s fminbnd ........................................................................ 113
References ..................................................................................................... 122

5. Working with Characters and Strings ................................................... 123

5.1 Introduction ....................................................................................... 123
5.2 MATLAB’s textscan Function ..................................................... 127

6. Roots of Algebraic and Transcendental Equations ............................. 131

6.1 Introduction ....................................................................................... 131
6.2 Search Method................................................................................... 132
6.3 Bisection Method .............................................................................. 133
6.4 MATLAB’s fzero Function............................................................. 134
6.5 MATLAB’s roots Function ............................................................ 139
References ..................................................................................................... 152

7. System of Algebraic, Linear Equations .................................................. 153

7.1 Introduction ....................................................................................... 153
7.2 System of Algebraic, Linear Equations .......................................... 153
7.2.1 MATLAB’s inv Function .................................................... 154
7.2.2 Gauss-Elimination Method ................................................ 154
7.3 Treatment of Large Systems of Algebraic, Linear Equations ......156
7.4 A Resistive Circuit Problem............................................................. 159
7.5 Gauss Elimination............................................................................. 161
7.6 Number of Solutions......................................................................... 162
References ..................................................................................................... 167

8. Curve Fitting................................................................................................ 169

8.1 Introduction ....................................................................................... 169
8.2 MATLAB’s Curve-Fitting Functions .............................................. 169
8.3 Curve Fitting with the Exponential Function ............................... 174
8.4 Cubic Splines...................................................................................... 178
8.4.1 MATLAB’s Cubic Spline Curve-Fitting Function............ 179

9. Numerical Integration ............................................................................... 187

9.1 Introduction ....................................................................................... 187
9.2 Numerical Integration and Simpson’s Rule .................................. 187
9.3 Improper Integrals ............................................................................ 190
viii Contents

9.4 MATLAB’s integral Function ..................................................... 190

9.5 MATLAB’s integral2 Function ................................................... 194
Reference ....................................................................................................... 204

10. Numerical Integration of Ordinary Differential Equations ............. 205

10.1 Introduction ..................................................................................... 205
10.2 Initial Value Problem and MATLAB’s Ordinary
Differential Equations Function ....................................... 206
Reference .......................................................................................................226

11. Boundary Value Problems of Ordinary Differential Equations ...... 227

11.1 Introduction ..................................................................................... 227
11.2 Difference Formulas ....................................................................... 227
References .....................................................................................................235

Appendix: Greek Letters and Special Characters in MATLAB® Plots..... 237

Review Answers ................................................................................................ 241
Index ..................................................................................................................... 255
Preface

I have taught computer applications course for engineers in the mechanical

and civil engineering departments at Florida Atlantic University (FAU), Boca
Raton, Florida, for many years. I first started teaching the course using the
Fortran language. Some years later, the department switched to the C/C++
language. More recently, the course has been taught using MATLAB®. The
advantage of using MATLAB over many other programming languages is
that MATLAB contains functions that enable the user to solve various math-
ematical problems, such as interpolation, roots of algebraic equations, the
relative minimum and maximum of a function, a system of linear algebraic
equations, curve-fitting problems, definite integrals, a system of ordinary
differential equations, and many others, some of which require special tool
boxes at an extra cost. There are also programming techniques available in
MATLAB but not available in either Fortran or C/C++. Because not all engi-
neering firms use MATLAB, I decided, in writing this textbook, to first cover
some very basic building blocks applicable to most, if not all, computer pro-
gramming languages used by engineers before getting into programming
that is specific to MATLAB. The syntax of these basic building blocks may be
different in different languages, but the concept is the same. The basic build-
ing blocks in programming covered in Chapters 2 through 4 are as follows:

1. Variable types, scalars, vectors, and matrices

2. Assignments (which, in most cases in this book, are arithmetic
statements)
3. Input/output statements
4. Loop statements
5. Conditional operators
6. Functions (built-in and self-written)

Before MATLAB, it was rare that I would write a computer program without
using a for loop. With that in mind, in this textbook, I introduce for loops
as early as is feasible. The authors of most other MATLAB textbooks intro-
duce for loops at a much later stage in their books.
Although students at FAU take the computer programming course in their
sophomore year, having taken Calculus II, the textbook can also be used at
the freshman level (the first eight chapters do not involve calculus). Although
there are many engineering example applications, the governing equations
are given without derivations. Therefore, students not only see variables x and
y but also see variables of pressure (p), temperature (T), time (t), velocity (V),

ix
x Preface

voltage (v), current (i), and so on. The chapters include review sections,
which may be used by the course instructor to ask the class questions on the
material that has been recently covered.
The primary objectives of the textbook are as follows:

1. To teach the reader the basic concepts in writing a computer

program (script) on the MATLAB platform, although many of the
concepts taught are also applicable to other computer programming
languages.
2. To familiarize the reader with many of MATLAB’s built-in functions,
some of which can be used to solve several mathematical problems,
such as interpolating for properties between table values, finding
the roots of transcendental and polynomial equations, determin-
ing the relative minimum or maximum of a function, and solving a
system of linear algebraic equations and curve fitting. The last two
chapters involve calculus and thus would only be applicable for a
course at the sophomore or higher level. These last two chapters
cover MATLAB’s functions for determining the value of a definite
integral and for solving a system of ordinary differential equations.

I have tried to organize the material so that the student gets to write a mean-
ingful program within several weeks of starting the course. The students
are required to add a comment section to their programs describing what
the program is about. Nearly all exercises and projects require the student to
produce tables or graphs or both.
The text contains many complete sample MATLAB programs and their
results, including tables, graphs, and comments what the program is about.
These examples should provide guidance to the student on completing the
exercises and projects that are listed in each chapter. Projects are at the end
of the chapters and are usually more difficult than the exercises. Many of
the projects are nontrivial. In recent times, I have used several exercise prob-
lems as in-class exams in which students submit their MATLAB programs
and results to me on blackboard. Projects are given as take-home exams to
be submitted to me within 1 or 2 weeks, depending on the difficulty of the
project. The projects require the student to write a computer program in
MATLAB to solve a mathematical or engineering-type problem.
The computer applications course that I teach is run as a lecture-laboratory
course. The advantage of running the course in this manner is that the
instructor is in the computer laboratory to help the student debug his or her
program. This includes the example programs as well as the exercises and
the projects. See the Table of Contents to get a more complete description of
the material covered in this textbook.
Preface xi

All example scripts in this book are available for download on the CRC
Press Website at https://www.crcpress.com/MATLAB-Essentials-A-First-
Course-for-Engineers-and-Scientists/Bober/p/book/9781138032378 .

MATLAB® is a registered trademark 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: [email protected]
Web: www.mathworks.com
Acknowledgments

I thank Jonathan Plant of CRC Press for his confidence and encouragement in
writing this textbook. I thank Dr. Andrew Stevens for allowing me to extract
many electrical engineering concepts and projects from our joint textbook
titled Numerical and Analytical Methods with MATLAB for Electrical Engineers.
I also thank Ed Curtis and Bala Gowri for guiding me through the textbook
submission process. I also thank the following people for their graphic con-
tributions: Danielle Mitchell and Jacqueline Ferrer. Finally, I wish to express
my deep gratitude to my wife for tolerating the many hours I spent on prepa-
ration of this manuscript—time which otherwise would have been devoted
to my family.

xiii
Author

William Bober, PhD, earned a BS in civil engineering at the City College

of New York (CCNY), New York, New York, an MS in engineering science at
Pratt Institute, Brooklyn, New York, and a PhD in engineering science and
aerospace engineering at Purdue University, West Lafayette, Indiana. At
Purdue University, he was on a Ford Foundation Fellowship and was
assigned to teach one engineering course in each semester. After he com-
pleted his doctoral work, he worked as an associate engineering physicist
in the Applied Mechanics Department at Cornell Aeronautical Laboratory,
Buffalo, New  York. After leaving Cornell Labs, he was employed as an
associate professor in the Department of Mechanical Engineering at the
Rochester Institute of Technology (RIT), Rochester, New York, for 12 years.
After leaving RIT, he accepted a position as an associate professor in the
Department of Mechanical Engineering at the Florida Atlantic University
(FAU). Recently, he was transferred to the Department of Civil Engineering
at FAU. He taught a computer application course for engineers for many
years at both RIT and FAU. This experience has given him the knowledge
to write or coauthor several textbooks on the subject, including Numerical
and Analytical Methods with MATLAB, Numerical and Analytical Methods
with MATLAB for Electrical Engineers, and Introduction to Analytical Methods
with MATLAB for Engineers and Scientists, all published by CRC Press. This
textbook, MATLAB Essentials: A First Course for Engineers and Scientists,
is for a first course in computer applications on the MATLAB platform
for engineers and scientists. He  has also written several papers for the
International Journal of Mechanical Engineering Education (IJMEE).

xv
1
Computer Programming with MATLAB®
for Engineers

1.1 Introduction
Most, if not all, engineering companies use computers in one way or the
other. Many employ computer programmers to solve company-specific
problems. These companies may also purchase or license software pack-
ages such as C/C++ or MATLAB®, and install the programs on their
computer systems to enable their programmers and engineers to efficiently
solve company-specific problems. The field of engineering, in particular,
lends itself to analytical and numerical solutions due to the highly math-
ematical nature of the field. Analytical and numerical methods invariably
involve writing computer code to solve a problem of interest. Mathematical
methods for solving many types of engineering problems use concepts from
linear algebra, root extraction of polynomial and transcendental equations,
integration, curve fitting, differential equations, and so on. MATLAB, using
a variety of analytical and numerical methods, has created built-in functions
that enable the user to readily employ these mathematical methods.
However, the user needs to know some programming techniques to effec-
tively make use of these built-in functions. Many examples involving the
use of these built-in functions are covered in this textbook.

1.2 Computer Usage in Engineering

Some of the ways that the computer is used in engineering are as follows:
1. Solving mathematical models of physical phenomena
2. Storing and reducing experimental data
3. Controlling machine operations
4. Communicating with other engineers and technicians on a particular
project

1
2 MATLAB® Essentials

This textbook is mostly concerned with item 1, that is, solving mathematical
models of physical phenomena.
The engineer’s interest lies in

1. Designing new products or improving existing ones

2. Improving manufacturing efficiency
3. Minimizing cost and power consumption
4. Maximizing yield and return on investment
5. Minimizing time to market
6. Research on developing new products

These can be accomplished by

1. Full-scale experiments. May be prohibitively expensive.

2. Small-scale model experiments. Still very expensive, and extrapola-
tion is frequently questionable.
3. A mathematical model that is the least expensive and faster. It can
provide more detailed answers and different cases under differ-
ent conditions and can be run quickly. If there is confidence in a
mathematical model, it will be used in preference to experiment.

1.3 Mathematical Model

Physical phenomena are described by a set of governing equations. Numerical
methods are frequently used to solve the set of governing equations, since
closed-form solutions for many types of problems are not available. Even
when closed-form solutions are available, the solution may be sufficiently
complicated that the computer is needed to calculate the desired answer.
Numerical methods invariably involve the computer. The computer per-
forms arithmetic operations upon discrete numbers in a defined sequence
of steps. The sequence of steps is defined in the program. A useful solution
is obtained if

1. The mathematical model accurately represents the physical phe-

nomena; that is, the model has the correct governing equations.
2. The numerical method is accurate.
Note: If the governing equations are not correct, the solution will be
worthless regardless of the accuracy.
3. The numerical method is programmed correctly.
4. This book is mainly concerned with items (2) and (3).
Computer Programming with MATLAB® for Engineers 3

1.4 Computer Programming

The advantage of using the computer is that it can carry out many cal-
culations in a fraction of a second; at the time of this writing, computer
speeds are measured in teraflops (trillions of floating-point operations per
second). However, to leverage this power, we need to write a set of instruc-
tions, that is, a program or script. For the problems of interest in this book,
the digital computer is only capable of performing arithmetic, logical, and
graphical operations. Therefore, arithmetic procedures must be developed
for evaluating integrals, determining roots of a transcendental equation,
solving a system of linear equations, solving differential equations, and so
on. The arithmetic procedure usually involves a set of algebraic equations.
A computer solution for such problems involves developing a computer
program that defines a step-by-step procedure for obtaining an answer
to the problem of interest. The method of solution is called an algorithm.
Depending on the particular problem, we might write our own algorithm,
or, as we shall see, we can also use the algorithms built into a package like
MATLAB in order to carry out well-known algorithms for solving many
types of mathematical problems.

1.5 Components of a Typical Desktop/Laptop Computer System

1. Input devices, typically includes a keyboard and mouse, but may
also include a touch-screen, a microphone, or a similar device.
Input devices provide a mechanism for humans to provide data and
instructions to the computer.
2. A Central Processing Unit (CPU) consisting of a Control Unit,
an Arithmetic Logic Unit (ALU), and registers. The Control Unit
fetches instructions from memory, executes the instructions, and
then returns the results to memory. The ALU performs arithmetic
and logical operations. Registers are high-speed local memory
locations, and are used to provide operands and store results from
the ALU.
3. A Memory/Storage Unit in which data and instructions are stored.
There are two types of memory: Main Memory and Secondary
Memory. Main Memory is used for temporary storage of programs
and data, and is commonly implemented with Dynamic Random
Access Memory (DRAM) devices. Items in Main Memory are not
saved when the computer is shut off.
4 MATLAB® Essentials

4. Secondary Memory stores data permanently. The Secondary Memory

commonly consists of
a. Hard drive: Provides semipermanent storage of programs and
data. It is usually internal to the computer. It has a large storage
capacity (terabytes).
b. Optical drive (DVD): Stores programs, data, video onto a CD, or
a DVD disk for permanent storage.
c. Flash drive: Stores programs and data onto a removable flash
memory stick, which can be used to transfer programs and data
from one computer to another.
5. Output devices, typically include a monitor or a printer, but may
also include speakers, a projector, a VR headset, or a similar device.
Output devices provide a mechanism for humans to receive data,
sound, and images from the computer.
6. Network interface allows the computer to send/receive data to other
computers in the vicinity (a local area network, or LAN) or around
the workplace (a wide area network, or WAN). Typical applications
that utilize the network include file transfer, e-mail, World Wide
Web, and streaming audio/video.
7. The operating system (OS) provides a unified environment for soft-
ware to utilize and control the computer. The OS manages storage in
order to enable the creation of files that are organized in folders and
are stored to a C drive. The OS also controls booting the computer and
instructs the display to show the desktop, and receives signals from
the mouse in order to move the cursor. Common operating systems
include Windows, MacOS, and Linux.

MATLAB utilizes all of the computer components described above. Users

utilize a keyboard and mouse to write and execute scripts. The OS saves
the scripts to storage and loads the MATLAB executable program into the
memory. MATLAB specifies instructions for execution by the CPU, and
subsequent display of results to the monitor or printer. The network allows
download of documentation, access to remote MATLAB servers, and soft-
ware updates.
The memory of a computer is an ordered sequence of storage locations
called memory cells. Each memory cell has an address indicating its relative
position in memory. The memory cell is a collection of smaller units called
bytes. A byte is the amount of storage required to store a single character
(letter, number, or symbol). A byte is a collection of smaller units called
bits. A bit takes on the value of 0 or 1, and is therefore suited for the binary
system of numbers. Generally, there are eight bits to a byte. Each character
or set of characters, or value, is represented by a particular pattern of zeros
and ones. The computer can retrieve or store a value.
Computer Programming with MATLAB® for Engineers 5

1.6 Overview of Programming Languages

A program is a list of instructions to be carried out by the computer. There
are two types of software:

1. System Software: Performs tasks required for the operation of the

computer, such as Windows 10, Windows 7, Unix, Linux, and so on.
2. Application Software: Written to perform particular tasks for the per-
son using the computer. These would include programs such as
Microsoft’s Office, AutoCAD, MATLAB, and so on. Programs writ-
ten by individuals would also be classified as Application Software.

There are different levels of computer languages. However, the computer

can only execute programs in machine language, which is considered the
lowest level. All higher level language instructions must be translated into
machine language. A sequence of machine language consists of a collection
of zeros and ones. Higher level languages include the following:
MATLAB, FORTAN, Basic, C/C++, Pascal, and so on. These higher level
languages allow as to write programs in a more familiar and understandable
manner than a program in machine language.

1.7 Why MATLAB?

MATLAB was originally written by Dr. Cleve Moler at the University of
New Mexico, Albuquerque, NM in the 1970s and was commercialized by
MathWorks, Natick, MA in the 1980s. It is a general purpose numerical
package that allows complex equations to be solved efficiently, and sub-
sequently generate tabular or graphical output. Although there are many
numerical packages available to engineers, many are very highly focused
toward a particular application, for example, ANSYS for modeling struc-
tural problems using the finite element method. As of the time of this
writing, MATLAB R2016b runs natively on Microsoft Windows, Apple
Mac OS, and Linux. In this textbook, we will assume that you are running
MATLAB on your local machine in a Microsoft Windows environment. It
should be straightforward for non-Windows users to translate the usage
descriptions to their preferred environment. In any case, these differences
are largely limited to the cosmetics and presentation of the program, and
not the MATLAB commands themselves. All versions of MATLAB (on any
platform) use the same command set, and the Command Window on all
platforms should behave identically.
6 MATLAB® Essentials

MATLAB is offered with accompanying toolboxes at additional cost to the

user. A wide variety of toolboxes are available in fields such as statistics,
optimization, control systems, and so on. However, in this textbook, we will
concentrate on teaching basic elements in computer coding on the MATLAB
platform and on fundamental numerical concepts without requiring the
purchase of any additional toolboxes.

1.8 Programming Methodologies

There are many methodologies for computer programming, but the tasks at
hand boil down to

1. Studying the problem to be programmed including the geometry of

the problem.
2. Listing the algebraic equations specified in the problem statement.
The equations will be based on the known physical phenomena.
3. Selecting the most efficient computer code and numerical method
for obtaining a solution to the problem of interest.
4. Creating an outline or a flow chart for the program flow (today, not
many textbooks on MATLAB recommend creating a flow chart).
5. Writing the program using the list of algebraic equations and the
outline or flow chart.
6. Debugging the program by running it and fixing any syntax errors
(programming language errors).
7. Examining the solution to see if it makes sense.
8. Refining and further debugging the algorithm and program flow.

Experienced programmers often omit some of these steps (or do them in

their head), but the overall process resembles any engineering project:
design, create a prototype, test, and iterate the process until a satisfactory
product is achieved.

1.9 MATLAB Programming Language

MATLAB may be considered a programming or scripting language unto itself,
but like every programming language, it has the following core components:

1. Data types, that is, integers, floating-point numbers, strings, vectors,

and matrices.
Computer Programming with MATLAB® for Engineers 7

2. Operators and built-in functions (e.g., commands for addition, subtrac-

tion, multiplication, division, trigonometric functions, and log function).
3. Control flow directives for making decisions and performing
repeated operations (e.g., loops, alternate paths, and functions).
4. Input/output (“I/O”) commands for receiving input from a user
or a file and for generating output to a file or to the screen (e.g., read
and print statements).

MATLAB borrows many constructs from other languages. For example, the
for and while loops and the fprintf commands are from the C program-
ming language (or its descendents C++). However, the biggest difference is
that the basic element in MATLAB is a matrix, thus providing the ability to
manipulate large amounts of data with a terse syntax, and allowing for the
solution of complicated problems in just a few lines of code. In addition,
MATLAB is also very rich in presentation functions to display sophisticated
plots and graphs.

1.10 Building Blocks in Writing a Computer Program

Most engineering computer programs will include some or all of the following
building blocks in program development:

1. Variable types, scalars, vectors, and matrices

2. Assignments (which, in most cases in this book, is an arithmetic
statement)
3. Input/Output statements
4. Loop statements
5. Conditional Operators
6. Functions (built-in and self-written)

Example programs containing these program building blocks are given

throughout this book.

1.11 Example Programs

The example programs in this book may be downloaded from the publisher’s
website at https://www.crcpress.com/MATLAB-Essentials-A-First-Course-
for-Engineers-and-Scientists/Bober/p/book/9781138032378. Students may
8 MATLAB® Essentials

then run the example programs on their own computer and see the results. It
may also be beneficial for students to type-in a few of the sample programs
(along with some inevitable syntax and typographical errors), thereby giving
the student the opportunity to see how MATLAB responds to program errors
and subsequently learn what they need to do to fix the problem.

REVIEW 1.1
1. List several ways engineers use the computer.
2. List several areas of interest for engineers.
3. List several methods that can be used in the design of new
products.
4. Which method mentioned in item 3 is the least expensive?
5. List several components of a typical desktop/laptop computer
system.
6. Name several computer languages used today and in the past
by Engineers.
7. What is the lowest level computer language and what number-
ing system does it use?
8. For engineers, what is the principle advantage of using MATLAB
over several of the other computer programming languages?
9. List several recommendations in developing a computer
program for solving a particular problem.
10. List several building blocks available in developing a program
in MATLAB or in most other Engineering Software Platforms
such as C or C++.
2
MATLAB® Fundamentals

2.1 Introduction
MATLAB® is a software program for numeric computation, data analysis,
and graphics. One advantage that MATLAB has for engineers over program-
ming languages such as C or C++ is that the MATLAB program includes
functions that numerically solve

1. Large systems of linear algebraic equations.

2. Roots of transcendental and polynomial equations.
3. One- and two-dimensional definite Integrals.
4. A system of first-order ordinary differential equations.
5. Statistical problems.
6. Optimization problems.
7. Control systems problems.
8. Many other types of problems encountered in engineering.

MATLAB also offers toolboxes (which must be purchased separately) that

are designed to solve problems in specialized areas.
In this chapter, we first familiarize the reader with some of the basic ele-
ments of the MATLAB platform. This allows the reader to shortly learn to
do computations in the Command Window and to write and run a script
in MATLAB. This is followed by discussing the basic building blocks in
constructing a computer program (script) for solving mathematical- and
engineering-type problems on the MATLAB platform. These building blocks
are applicable in any programming language, the syntax may be different,
but the concept is the same. Recall at the end of Chapter 1, the building
blocks that was mentioned and which are covered in this book are

1. Variable types, scalars, vectors, and matrices.

2. Assignments (including arithmetic statements).
3. Input/Output statements.

9
10 MATLAB® Essentials

4. Loop statements.
5. Conditional Operators (leading to alternate paths in the program).
6. Functions (built-in and self-written).

Items (1)–(4) are covered in this chapter as well as some of the elementary
built-in functions of item (6). Items (5) and (6) are covered in Chapter 3.
Examples of MATLAB programs that solve various types of mathematical
problems, many of which are related to engineering-type problems, are
covered throughout this book.

2.2 MATLAB’s Desktop

Mathworks, the company that developed MATLAB, normally update their
version of MATLAB every six months. In this textbook, Sections 2.2 and 2.3
discusses the MATLAB desktop windows and how to construct a script in
MATLAB based on MATLAB version R2016b.
Under Microsoft Windows, MATLAB may be started via the Start Menu
or by clicking on the MATLAB icon on the desktop. Upon startup, a
new window will open containing the MATLAB desktop (not to be con-
fused with the Windows desktop), and one or more MATLAB windows
will open within the MATLAB desktop (see Figure 2.1 for the default
configuration).
The main windows are the Command Window, Current Folder, and
Workspace. You can customize the MATLAB windows that appear upon
startup by clicking on Layout in the Toolstrip and checking (or unchecking)
the windows that you wish to appear on the MATLAB desktop. Figure 2.1
shows the Command Window (in the center), the Current Folder Window
(on the left), the Workspace Window (on the right), and a Long Narrow box
containing the Path to the Current Folder (just below the Toolstrip and just
above the Command Window). MATLAB designates this Long Narrow
box as the Current Folder Toolbar. These windows and the Current Folder
Toolbar are summarized as follows:

• Command Window: In the Command Window, you can enter com-

mands and data, make calculations, and print results. You can
write a script in the Command Window and execute the script.
However, writing a script directly into the Command Window is
discouraged because it will not be saved, and if an error is made,
MATLAB® Fundamentals 11

FIGURE 2.1
MATLAB desktop windows for the three-column option.

the entire script must be retyped. By using the up arrow (↑) key
on your keyboard, the previous command can be retrieved (and
edited) for reexecution.
• Current folder toolbar: This Toolbar gives the path to the Current
Folder. To run a MATLAB script, the script needs to be in the folder listed
in this Toolbar.
• Current Folder Window (on the left): This window lists all the files in
the Current Folder whose path is listed in the Current Folder Toolbar.
By double clicking on a file in this window, the file will open within
MATLAB.
• Workspace Window: This window will be on the right for a three-
column option (see Figure 2.1) or below the Current Folder Window
for the two-column option (see Figure 2.2). The two- or three-column
option can be selected from the layout options in the Toolstrip.
The  Workspace Window contains all the commands entered into
the Command Window.
• Editor Window: To open this window, click on the New Script icon
in  the Toolstrip in MATLAB’s desktop (see Figure 2.1). This will
open the Editor Window (see Figure 2.3). This window may be used
to create, edit, and execute MATLAB scripts (also called programs).
Figure 2.4 contains the script for Example 2.1.
12 MATLAB® Essentials

FIGURE 2.2
MATLAB desktop windows for two-column option.

FIGURE 2.3
Editor Window just above the Command Window.
MATLAB® Fundamentals 13

FIGURE 2.4
Script for Example 2.1.

2.3 Constructing a Script (Program) in MATLAB

In the first few examples the reader is asked to type-in several commands in
the Command Window. Subsequent examples involve creating scripts in the
Editor Window. The following list summarizes the steps for writing a script
in MATLAB:

1. If available, start the MATLAB program by double-clicking on the

MATLAB icon on the Window’s desktop. If not available, go to
the Window’s Start Menu, click on All Programs, find the MATLAB
program among the list of available programs, and double-click
on it. This will open up the MATLAB desktop.
2. Click on the New Script icon in the Toolstrip in MATLAB’s desktop. This
brings up a new Editor Window just above the Command Window (see
Figure 2.3).
3. Type your program into the Editor Window.
14 MATLAB® Essentials

4. When you are finished typing in the program, save the script by click-
ing on the Save icon in the Toolstrip (see Figure 2.3). A dialog box will
open in which you are to select the folder (left column), and in which
you are to  type-in the name of the script in the File Name Dialog
Box (see Figure 2.5). By default, your program will be saved with the
.m extension. It is best to use a folder that contains only your own
MATLAB scripts.
5. You may then run the script in the Editor Window by clicking on the
arrow located just above the Run icon in the Toolstrip (see Figure 2.6).
This icon is a Save and Run Command. Note: In the Editor Window,
the arrow is green.
Alternatively, you can run the script from the Command Window
by typing the script name (without the .m extension) after the MATLAB
prompt (>>). For example, if the program has been saved as heat.m,
then type heat after the MATLAB prompt (>>), as shown below:

>> heat

Select folder Type in name of script

FIGURE 2.5
Select folder (left column) and type in the name of the script in the file name dialog box.
MATLAB® Fundamentals 15

FIGURE 2.6
Save and Run icon in the toolstrip.

6. If you try to run your script and your script is not in the Current
Folder whose path is listed in the Current Folder Toolbar, a dialog
box will appear giving you the option of changing the folder listed
in the Current Folder Toolbar to the folder containing your script
(see  Figure 2.7). If a path to the folder containing the script has
already been established, click on the Change Folder button.
7. If you need additional help on getting started, you can click on the
Help icon in the Toolstrip in MATLAB’s desktop (see Figure 2.2).
If you are in the Editor Window, click on Home (upper left) to
get back to MATLAB’s desktop. In the window that opens (see
Figure 2.8) you can type-in an item of interest in the search box, or
you can click on the MATLAB option that brings up the window
shown in Figure 2.9.
8. Whenever you write a script, it is good practice to add comment lines
at the beginning of the script describing what the program is about.
This is accomplished by placing a % sign in front of a statement in the
script. Example:

% This program plots velocity vs. time.

16 MATLAB® Essentials

FIGURE 2.7
Dialog box for changing folder or path.

FIGURE 2.8
Help window.
MATLAB® Fundamentals 17

FIGURE 2.9
Topics in the MATLAB Help window.

REVIEW 2.1
1. What are the two alternative ways to start the MATLAB
program?
2. What are the windows in the MATLAB’s default desktop?
3. It is best to write a MATLAB script (program) in the Editor
Window. From MATLAB’s default desktop, how does one
open the Editor Window?
4. After you have completed writing a script in the appropriate
window, what is the next step?
5. Name two ways to execute a script.
6. What happens if you try to run a script and the folder contain-
ing the script is not listed in the Current Folder Toolbar?
7. In MATLAB, what is the file name extension for saved scripts?
8. How does one establish a comment line in a script?
18 MATLAB® Essentials

2.4 Variable Names and Types

• Variable names must start with a letter.
• Can contain letters, digits, and the underscore character (no spaces).
• Can be of any length, but must be unique within the first 19 characters.
NO T E : Do not use a variable name that is same as a file name, a MATLAB function
name or a self-written function name.
A variable can be a scalar (A = 3.5), a vector (A = [2 4 6 8]), or a matrix

 1 3  
A= 
 6 5  


A scalar has just one value in the computer memory, whereas a vector of n
elements will have n values in the computer memory, and a matrix of n rows
and m columns will have n × m values in the computer memory.
To make it easier to follow ones program, it is best to use variable names that are
similar to the variables used in a problem statement.
MATLAB command names and variable names are case sensitive.
Numerical variables can be either an integer (no decimal point) or a float-
ing point number (one with a decimal point). Integers can be stored in either
8 bits (numbers less than 127 or greater than −127), 32 bits (numbers less
than 32,767 or greater than −32,767) or 64 bits of memory. In MATLAB, the
default for floating point numbers is double precision that requires 64 bits of
memory. You can specify variables to be single precision that only requires
32 bits of memory. Calculations carried out in single precision are faster than
carrying out calculations in double precision. For all problems in this text-
book, time of execution is not a problem. Numbers larger than approximately
3.4 × 10 38 or less than − 3.4 × 10 38 requires double precision.
When defining a variable, either in the Command Window or in a script,
you should place a semicolon after the variable definition when you do not
want the command echoed to the screen. In the absence of a semicolon,
the defined variable appears on the screen. For example, if you entered the
following command in the Command Window:

>> A = [3 4 7 6]

In the Command Window, you would see

A =
3 4 7 6
>>
MATLAB® Fundamentals 19

Alternatively, if you add the semicolon after the command statement, then
your command is entered but there is nothing printed to the screen, and the
prompt immediately appears for you to enter your next command:

>> A = [3 4 7 6];
>>

2.5 Assignment Operator

The assignment operator is of the form

Variable name = an expression

In most cases in this book the expression will be an Arithmetic Statement

involving constants, Arithmetic Operators, variables, MATLAB functions
and self-written functions. The one exception would be when we are deal-
ing with characters and strings. Many MATLAB built-in functions that may
be used in an arithmetic statement are discussed later. The way it works is
that the Arithmetic Logic Unit in the computer will determine a value of
the  expression on the right-hand side of the = sign and replace the value
of the variable on the left-hand side of the = sign with the value determined
by the expression on the right-hand side of the = sign. For example,
Suppose you had the following assignments:

x=10; y=20;
x=x+y;

What the computer does is to determines the value of x+y, which equals
30, and replaces the original value of x, which is 10, with the new value
of 30. Although, algebraically, the expression x=x+y does not make sense,
since that would make y=0, it does make sense in programming language.
Although, algebraically you can write x+y=20, you cannot do this in the
programming language. There needs to be a single variable on the left-hand
side of the = sign.

NO T E : In an Arithmetic Statement, all variables on the right-hand side of the equal

sign must be previously defined (given a numerical value) in the program.

You may have noticed that in the variable assignments above that we placed
a semicolon at the end of the assignment. This avoided the variables from
being echoed to the screen.
20 MATLAB® Essentials

2.5.1 Arithmetic Operators

The Arithmetic Operators used for addition, subtraction, multiplication,
division, and exponentiation are listed below:

+ Addition
− Subtraction
* Multiplication
/ Division
^ Exponentiation

For Arithmetic Statements that contain several Arithmetic Operators and

parentheses, there is a specific order that is used in evaluating the arithmetic
expression. First, the Arithmetic Logic Unit, going from left to right, searches
for parentheses, if it finds them, it will carry out the operations inside all the
parentheses in the following order: exponentiation, multiplication and divi-
sion, and addition and subtraction. It then returns to the beginning of the
arithmetic statement and carries out the operations in the same order listed
above.
Knowing this order may help you in deciding where parentheses are
required when you write arithmetic statements. Suppose you had an expres-
sion y = c 2m, you might be tempted to write the expression in the MATLAB
Command Window (after defining c and m) as

clc;
c = 36.0; m = 3.0;
y = c/2*m

This would give the wrong answer for y. MATLAB would divide c by 2
and multiply the result by m. The correct ways to write the expression are

y = c/(2*m) or c/2/m

In the first expression, MATLAB will first carry out the expression within
the parentheses, so that the 2*m becomes one number, and then c is divided
by this one number. In the second expression, there are no parentheses, so
MATLAB, proceeding from left to right, will calculate c/2, then divide the
result by m. Try typing these expressions in the Command Window and
observe the two different answers you get for y.
To display a variable value, just type the variable name without the
semicolon, and the variable will appear on the screen.
Try typing these commands into the Command Window and verify the
results:
MATLAB® Fundamentals 21

clc;
x = 5; y = 10; z = x + y <enter>
w = x – y <enter>
z = y/x <enter>
z = x*y <enter>
u = x^2 <enter>

2.6 Some MATLAB Features, Commands,

Special Items, and Built-in Functions
2.6.1 Trigonometric and Other Useful Functions
Whenever you write a script, you should add comment lines to the script that
describes what the script is about. You do this by entering the percent sign (%)
at the beginning of the line. Example:
% This script determines the velocity of a free falling body …
You may also add a comment after a particular command.
Whenever you write a script, it is a good practice to clear out variables that
are left in the workspace from previous programs, since there could be a conflict
between the variables used in the present program with those used in a previ-
ous program. You may also wish to clear contents in the Command Window, so
that only results from the present program appear in the Command Window.
You may accomplish this by placing the following commands at the beginning
of your script (after your comment lines describing what your script is about).
clear; removes all variables and data from the work space.
clc; clears the Command Window.
If you wish to clear a graphics window, use
clf; clears the Graphics Window.
If you find that your program is running in an infinite loop, you can halt the
program by hitting the ctrl and C keys simultaneously, that is,
ctrl-C aborts a program that may be running in an infinite loop.
Commands are case sensitive. Use lowercase letters for commands.
The quit command or exit command terminates MATLAB.
The save command saves variables or data in the work space of the
Current Folder. The data file name will have the .mat extension.
The basic data structure in MATLAB is a matrix.
A matrix is surrounded by brackets and may have an arbitrary number of
rows and columns; for example, the matrix
1 3
A=
6 5 
22 MATLAB® Essentials

may be entered into MATLAB as

>> A = [ 1 3 <enter>
6 5 ]; <enter>
or
>> A = [ 1 3 ; 6 5 ]; <enter>

where the semicolon within the brackets indicates the start of a new row
within the matrix. In the above expression for matrix A, row 1 are the elements
1 and 3, row 2 are the elements 6 and 5, column 1 are the elements 1 and 6, and
column 2 are the elements 3 and 5.
A matrix of 1 row and 1 column is a scalar. Example:

>> A = [ 3.5 ];

Alternatively, MATLAB also accepts A=3.5 (without brackets) as a scalar.

A matrix consisting of 1 row and several columns, or 1 column and several
rows is considered a vector. Example:
>> A = [ 2 3 6 5 ] (row vector)
>> B = [ 2
3
6
5 ] (column vector)

We can convert a column vector to a row vector by using the transpose

symbol. Suppose we enter B as a column vector in the Command Window
then write B'. We would see the following in the Command Window:
>> B=[2
3
6
5];
>> B'
ans =
2 3 6 5
>>

A matrix can be defined by including a second matrix as one of the elements.

Example:
>> B = [ 1.5 3.1 ];
>> C = [ 4.0 B ]; (thus C = [ 4.0 1.5 3.1] )

You can select a specific element of the vector c as follows:

If C = [ 4.0 1.5 3.1], then
>> b = C(2)
MATLAB® Fundamentals 23

gives b = 1.5 .
1 3
If A =  , then
6 5 
>> b = A(2,2)
gives b = 5.

You can also define a vector by specifying each element in the vector, for
example:

A(1)=1, A(2)=3, A(3)= 5, A(4)=7, then vector A=[1 3 5 7].

The element number must be an integer.

This concept is very important and is used in many examples that follow.

2.6.1.1 Special Values

One special value in MATLAB is ans, it is the last computed unassigned
result to an expression typed in the Command Window. For example, if we
typed in the following assignments in the Command Window, we can see
MATLAB’s response.

>> x=5; y=10;

>> x+y
ans =
15
>>

MATLAB has a built-in value for the variable π. Its symbol is pi, and it should
be used in place of 3.14 whenever π appears in an arithmetic statement.
Typing pi in the Command Window gives

>> pi
ans =
3.1416
>>

The display default is four places, but it is carried to many more places in
memory.
If you had an expression in which you accidentally divided by zero, MATLAB
would respond with the infinity value, ∞, with the symbol, inf. Example:

>> x=10; y=0;

>> z=x/y
z =
Inf
>>
24 MATLAB® Essentials

2.6.1.2 Trigonometric Functions

There are many engineering examples that involve the trigonometric func-
tions. Similar to other computer programs, MATLAB has functions that
evaluate the trigonometric functions. The most frequently used are shown
below.

MATLAB’s Function Trigonometric Function Name

sin( ) sine
cos( ) cosine
tan( ) tangent

The arguments of these trigonometric functions are in radians. However, the

arguments can be made in degrees if a d is placed after the function name,
such as sind(x). In all of the trigonometric functions, you may use simple
arithmetic in the arguments of the function.
Try typing these statements in the Command Window and use your calcu-
lator to verify the results.

clc;
x = 50/180*pi; y = sin(x) <enter>
z = cos(pi/2) <enter>

The answer should be 0, but with round off error it gives 6.1232e-17.

w = tan(pi/4) <enter>
x = 45/180*pi; y2 = sin(x) <enter>
z2 = cos(x) <enter>
x1 = sind(50) <enter>
y1 = cosd(90) <enter>
w1 = tand(45) <enter>

2.6.1.3 Inverse Trigonometric Functions

MATLAB’s Function Trigonometric Function Name

asin( ) Inverse sine
acos( ) Inverse cosine
atan( ) Inverse tangent

Since the values of the sine and cosine functions vary from −1 to +1. The input
arguments to the asin( ) and acos( ) functions should be from −1 to +1.
The results will be in radians. The value of the tan function can be anywhere
from (−∞ to + ∞ ), so the input argument to the atan( ) function can be any
number, but the result will be in radians.
MATLAB® Fundamentals 25

Try typing these statements into the Command Window and use your
calculator to verify the results:

clc;
x = asin(0.5); xd = x*180/pi <enter>
y = acosd(-1.0) <enter>
z = atand(1.732) <enter>
z = atan(1.0); zd = z*180/pi <enter>

2.6.1.4 Exponential, Square Root, and Error Functions

MATLAB’s Function Mathematics Function Name

exp( ) Exponential (e( ), e ≈ 2.7183)
log( ) Natural logarithm
log10( ) Common (base 10) logarithm
sqrt( ) Square root
erf( ) Error function

Try typing these statements into the Command Window and use your
calculator to verify the results:

clc;
x = 2.5; y = exp(x) <enter>
z = log(y) <enter>
w = sqrt(x) <enter>
u = log10(100) <enter>

Suppose we had a problem involving the following arithmetic statement that

we needed to evaluate:

 k 2 
 c 
y = cos  −  t
 m  2m  
 

To make it easier to write the MATLAB statement corresponding to the above

arithmetic statement, we could break up the argument of the cos function as
follows (type the following in the Command Window):

k = 200; c = 5; m = 25; t = 5;
arg = sqrt(k/m - (c/(2*m))^2);
y = cos(arg*t )
26 MATLAB® Essentials

2.6.1.5 Complex Numbers

Complex numbers may be written in two forms: Cartesian, for example:

z = x + yj;

The x part is considered the real part of the complex number and the y part
is considered as the imaginary part of the complex number.
The complex number can also be expressed in polar form, for example:

z = r * exp(j*theta).

MATLAB allows the use of i and j for −1. Programmers who have expe-
rience with FORTRAN, the programming language that was commonly
used in engineering many years ago, frequently used i and j as integer loop
variables.
In this book we do not deal with complex numbers very often, but when
we do, we will use j for −1. Also, there are many examples in this book
where i is used as an integer loop variable.

2.6.2 Other Special Values

MATLAB’s Function Math Function Name

abs( ) Absolute value (magnitude)
conj( ) Complex conjugate
imag( ) Imaginary part of a complex number
real( ) The real part of a complex number

Try typing these statements into the Command Window and use your calcu-
lator to verify the results:

clc;
z1 = 1 + j; z2 = 2*exp(j*pi/6)=2*(cos(pi/6)+j sin(pi/6));
y = abs(z1) <enter>
w = real(z2) <enter>
v = imag(z2) <enter>

2.6.2.1 Other Useful MATLAB Functions

size(X) Gives the size (number of rows and the number of columns of
matrix X).
x' Transposes a matrix or vector, rows become columns and
columns become rows.
length(X) For vectors, length(X) gives the number of elements in X.
(Continued)
MATLAB® Fundamentals 27

linspace(X1,X2,N) Generates N points between X1 and X2.

sum(X) For vectors, sum(X) gives the sum of the elements in X. For
matrices, sum(X) gives a row vector containing the sum of
the elements in each column of the matrix.
max(X) For vectors, max(X) gives the maximum element in X. For
matrices, max(X) gives a row vector containing the maximum
in each column of the matrix. If X is a column vector, it gives
the maximum absolute value of X.
min(X) Same as max(X) except it gives the minimum element in X.
mean(X) The mean of a vector, also known as the average, equals the
sum of the vector elements divided by the number of elements
in the vector. For vectors, mean(X) gives the mean value of
the vector X. For matrices, mean(X) gives a row vector
containing the mean value in each column of the matrix X.
sort(X) For vectors, sort(X) sorts the elements of X in ascending
order. For matrices, sort(X) sorts each column in the matrix
in ascending order.
factorial(n) n ! = 1 × 2 × 3 × ... × n
mod(x,y) Modulo operator gives the remainder resulting from the
division of x by y. For example, mod(13,5) = 3, that is, 13 ÷ 5
gives 2 plus remainder of 3 (2 is discarded). Another example:
mod(n,2) gives zero if n is an even integer and one if n is an
odd integer.

Try typing these statements into the Command Window:

NO T E : The repeat of A is not necessary, but it will make it easy to see the
results of n, y, z, w, and u.

clc;
A = [ 2 15 6 18 ]; n = length(A) <enter>
A <enter> y = max(A) <enter>
A <enter> z = sum(A) <enter>
A <enter> w = mean(A) <enter>
A <enter> u = sort(A) <enter>
A = [ 2 15 6 18; 15 10 8 4; 10 6 2 3 ]; <enter>
A' <enter>
A <enter> x = max(A) <enter>
A <enter> y = sum(A) <enter>
A <enter> w = mean(A) <enter>
A <enter> u = sort(A) <enter>
A <enter> z = size(A) <enter>
w = mod(21,2) <enter>
u = mod(20,2) <enter>

A list of the complete set of elementary math functions can be obtained by

typing help elfun in the Command Window.
28 MATLAB® Essentials

2.6.2.2 Colon Operator (:)

The colon operator may be used to

1. Create a new matrix from an existing matrix; examples:

 5 7 10 
 
if A =  2 5 2 
 1 3 1 

5
 
then x = A(:,1) gives x =  2 
 1 

The colon in the expression A(:,1) implies all the rows in matrix A,
and 1 implies column 1.

 7 10 
 
x = A(:,2:3) gives x =  5 2 
 3 1 

The first colon in the expression A(:,2:3) implies all the rows in A,
and the 2:3 implies columns 2 and 3.
We can also write

y = A(1,:) that gives y = [ 5 7 10 ]

The 1 implies the first row and the colon implies all the columns.
2. Colon operator can also be used to generate a series of numbers
(as  in a for loop, which is discussed later) or to create a vector.
The format is
n = starting value: step size: final value.
If the step size is omitted, the default step size is one. Example:
n = 1:8 gives n =  1 2 3 4 5 6 7 8  .
To increment in steps of 2 use
n = 1:2:7 gives n =  1 3 5 7 

Exercise

E2.1. Type the following matrix in the Command Window. Assume that the
first, second, and third columns represent the vector variables of altitude,
z, temperature, T, and density, rho respectively: (a) use the colon operator
MATLAB® Fundamentals 29

to define the vector variables, (b) determine the mean values of altitude,
temperature, and density, (c) determine the length of vector z, and (d)
determine the size of matrix A. Print the results to the Command Window.

 0 288.15 1.2252 
 
1000 281.65 1.1118 
 
 2000 275.15 1.0065 
A= 
 3000 268.65 0.9091 
 
 4000 262.15 0.8191 
 
 5000 255.65 0.7360 


2.6.2.3 Preallocation of a Matrix

Sometimes it is necessary to preallocate a matrix of a given size. This can be
done by defining a matrix of all zeros or ones; Examples:

0 0 0 
 
A = zeros(3) = 0 0 0  (3 rows, 3 columns)
 
0 0 0 

0 0
 
B = zeros(3:2) = 0 0  (3 rows, 2 columns)
 
0 0 

REVIEW 2.2
1. List at least two conditions in selecting a name for a variable.
2. Finish the following statement. An arithmetic statement may
involve …
3. What can be said about the variables that appear on the right
side of an arithmetic statement?
4. List the Arithmetic Operators in MATLAB.
5. What is the order in which an arithmetic statement will be
carried out?
(Continued)
30 MATLAB® Essentials

REVIEW 2.2 (Continued)

6. What is MATLAB’s command for
a. π.
b. e.
c. ln.
d. Sine function in radians.
e. Sine function in degrees.
f. sin−1 function.
g. The number of elements in a vector.
h. The size of a matrix (the number of rows and columns).
i. The sum of the elements in a vector.
j. The maximum element in a vector.
k. Preallocating the size of a 3 × 3 matrix.
7. What is the purpose of placing a semicolon at the end of a com-
mand statement or a variable assignment?
8. What is the command that will clear the Command Window?
9. What is the basic data structure in MATLAB?
10. Name two functions of the colon operator.

2.7 MATLAB Output

To display a vector X, just type X without the semicolon, and vector X will be
printed to the screen. For example, first define X,

>> X = [0 1 2 3 4 5];
>>

Now enter X without the semicolon.

>> X

The following will be displayed on the screen:

X =
0 1 2 3 4 5
>>
MATLAB® Fundamentals 31

2.7.1 The disp Command

The disp command prints only the items that are enclosed within the paren-
theses, which can be a variable or alphanumeric information. Alphanumeric
information must be enclosed by singe quotation marks. Example (assum-
ing that vector X above has already been entered in the Command Window)
type in

disp(X); disp(' m/s');

The following will be displayed on the screen:

0 1 2 3 4 5
m/s
>>

2.7.2 The fprintf Command

The fprintf command prints formatted text to the screen or to a file.
Example:

>> V = 2.2; clc;

fprintf('The velocity is %f m/s \n', V);

The following will appear on the screen:

The velocity is 2.200000 m/s

The \n in the above command tells MATLAB to move the cursor to the
next line.
MATLAB also has a tab command. It is \t; this command tells MATLAB
to move the cursor several spaces along the same line.
The %f refers to a formatted floating point number that is assigned to vari-
able V, and the default is 6 decimal places. The command fprintf uses
format strings based on the C programming language. You can specify the
number of spaces allotted for the printed variable as well as the number of
decimal places by using %8.2f. This will allow eight spaces for the variable
to two decimal places. You can also just specify the number of decimal places
for the variable and let MATLAB decide the number of spaces allotted for the
printed variable. For example, to specify three decimal places use %.3f. The
variable will be printed out to three decimal places, but MATLAB will decide
the number of spaces for the variable. However, to create neat looking tables,
it is best to specify the number of spaces in the format statement that allows
for several spaces between variables in adjacent columns, such as %10.3f.
32 MATLAB® Essentials

Other formats:

%i or %d Used for integers

%f Used for a floating point number (one with a decimal point)
%e Scientific notation (e.g., 6.02e23), default is 6 decimal places
%g Automatically uses the briefest of %f or %e format
%s Used for a string of characters
%c Used for a single character

Unlike C, the format string in MATLAB’s fprintf must be enclosed by

single quotation marks (and not double quotes).

2.7.3 Printing to a File

It is often useful to print the results of a MATLAB program to a file, possibly
for inclusion in a report. In addition, program output that is printed to a file
can be subsequently edited within the file, such as aligning or editing column
headings in a table. Before you can print to a file, you need to open a file for
printing with the command, fopen. The syntax for fopen is

fo = fopen('filename','w')

Thus, fo is a pointer to the file named filename, and the w indicates that
there will be writing to the file. The fo can be replaced by a name selected
by the programmer. To print to filename use

fprintf(fo,'format',var1,var2,..);

where the format string contains the text format for var1, var2, and so on.
Try typing the following example script in the Editor Window, save the
script in the folder that you have chosen for your MATLAB scripts (this
becomes the current folder), then run the script (see Section 2.3). The output
should go to the file named output.txt, which should be located in the same
folder as the script that produced it. To see the results, open the output file,
as described below. If you wish, you can edit the results and also print the
results by clicking on the print command in the Toolstrip.

Example 2.1
% Example_2_1.m
% This program is an example for printing to a file.
clear; clc;
V=12; % velocity
F=50.2; % force
fo=fopen('output.txt','w');
fprintf(fo,'V=%4i m/s, F = %5.2f N \n',V, F);
fclose(fo);
------------------------------------------------------------------------
MATLAB® Fundamentals 33

Program Results:
V= 12 m/s, F = 50.20 N
---------------------------------------------------------------------

The extension on the output file should be .txt (otherwise when you try
to open the file, MATLAB will start the import wizard). The resulting
output file will be saved and listed in the Current Folder. You can open
the file by double clicking on the output.txt file listed in the Current Folder
column on the left (see Figure 2.10). Alternatively, you can open the file by
clicking on the Open icon in the Toolstrip that brings up a screen listing
all the .m files in the Current folder. In the box labeled File name, type
in *.txt. This will bring up a screen listing of all the files with the exten-
sion .txt in the Current folder as shown in Figure 2.11. To open the file of
interest, double click on the name of the output file (in this example, the
file name is output.txt).
In earlier versions of MATLAB, you would not be able to open the output
file without having included the fclose(fo) statement in the program. But
it is still a good practice to include the fclose statement after all the output
statements in the program, or at the end of the program itself.

FIGURE 2.10
Searching for the output.txt file in Current Folder Column.
34 MATLAB® Essentials

FIGURE 2.11
Opening all files with .txt extension.

2.8 Simple Plot Commands

MATLAB provides many different types of plots. Clicking on the PLOTS tab
in MATLAB’s desktop graphically lists the various types of plots that are
available (see Figure 2.1). The commands for creating linear plots, semilog
plots, and log-log plots are as follows:

Plot(x,y) Linear plot of y versus x

Semilogx(x,y) Semilog plot (log scale for x-axis, linear scale for y-axis)
Semilogy(x,y) Semilog plot (linear scale for x-axis, log scale for y-axis)
Loglog(x,y) Log-log plot (log scale for both x- and y-axes)

Unless you wish to plot a single point, the arguments in the plot command must
be vectors. In addition, the vectors need to be of the same length. If the argu-
ments in the plot command are scalars, the plot commands will produce just
a single point.

2.8.1 Linear Plot

Suppose we have a relationship of V = f (t) and we have created the following
vectors (V(1) occurs at t(1), V(2) occurs at t(2), etc.).

t = 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 

V =  −20.2 − 21.0 −19.4 −14.7 − 6.2 6.9 25.4 50.0 81.4 120.4 167.8 
MATLAB® Fundamentals 35

In the following script we plot V versus t using the plot(t,V) command. We

will assume that t is in seconds and V is in meters/second.
We can label the t-axis, v-axis, and add a title and a grid with the following
commands:
xlabel('t(s)'),
ylabel('V(m/s)'),
title('V vs. t'),
grid;

Example 2.2
% Example_2_2.m
% The vectors t and V are entered into the program.
% Then a plot of V vs. t is created.
% To plot V vs. t both variables need to be vectors
% of the same length.
clear; clc;
t = [0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0];
V = [−20.2 -21.0 -19.4 -14.7 -6.2 6.9 25.4 50.0 81.4 120.4...
167.8];
% Create the plot of V vs. t.
plot(t,V), xlabel('t(s)'), ylabel('V(m/s)'), title('V vs. t'), grid;
------------------------------------------------------------------------

Program Results:
See Figure 2.12.
------------------------------------------------------------------------

V vs. t
200

150

100
V(m/s)

50

−50
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
t(s)

FIGURE 2.12
Plot of V versus t.
36 MATLAB® Essentials

REVIEW 2.3
1. Name two commands that will result in printing to the screen.
2. What is the command that will move the cursor to the next
line?
3. What is the format that will print a floating point variable to 10
spaces and to three decimal points?
4. What is the format that will print a floating point variable in
scientific notation to 12 spaces and to four decimal points?
5. What are the commands necessary to print to a file?
6. What is the command to create a plot of y versus x and what
type of variable must x and y be?
7. What are the commands that will label the x- and y-axis and
provide a title to a plot?

If a program involves creating more than one plot, you need to include the
statement figure after each plot command (except the last), otherwise only
the last plot will appear.
You may have noticed that the script for Example 2.2, contained comment lines
that described what the script is about. This is a good practice and should be imple-
mented every time you write a script.

2.9 Loops
2.9.1 The for Loop
The for loop command provides the means to repeat a series of statements
with just a few lines of code. In MATLAB, in many cases, one can avoid the
use of the for loop and achieve the same result. However, the method used
in MATLAB to achieve this may not be available in many other computer
platforms. Since we are emphasizing the computer programming building
blocks that are applicable in most, if not all, programming languages, we will
exclusively use the for loop method in the first few chapters of this book.
Syntax:

for loop variable = starting value: step size: final value

The step size may be omitted, and then MATLAB will take the step size to
be 1. Although the loop variable need not be an integer, in most cases in
this book, it will be an integer. That is because, we frequently use the loop
MATLAB® Fundamentals 37

variable to select or create an element of a vector or a matrix. Elements of

a vector are identified by an index which must be an integer. In most other
platforms, if you wish a variable to be an integer, you need to designate that
variable as an integer. This is not the case in MATLALB. MATLAB looks
at the context in which the variable is used and knows when to consider the
variable as an integer.
As an example, we will take the index variable as m, the starting value as
1, omit the step size and take the final value as 20, then our for loop will be

for m = 1:20
statement;
.
.
.
statement;
end

MATLAB sets the index m to 1, carries out the statements between the for
and end statements, then returns to the top of the loop, changes m to 2 and
repeats the process. After the process has been carried out 20 times the pro-
gram exits the loop without further executing any of the statements within
the loop.
NO T E : There is no semicolon after the for and end statements.
All statements that are not to be repeated should not be within the for loop. For
example, table headings that are not to be repeated should be outside the for loop.
Also notice that statements within the for loop are indented. MATLAB does
this to make it easier to read and debug a script containing for loops. You
can have MATLAB to do final indenting by highlighting your entire script
and then entering Ctl-I.

Example 2.3
% Example_2_3.m
% This program is an example of the use of a for loop in which
% the indices of the for loop select an element of a vector.
% The indices must be an integer. In the for loop expression,
% MATLAB will take i as an integer if there is no decimal point
% in the assignment of variable i. But you can also specify i
% by the assignment int8(i) or int32(i) before it is used.
% The assignment length(t) specifies the number of elements
% in the vector t.
% The program creates a table of y1 and y2 vs t.
% 0 <= t <= 10 in steps of 0.5.
clear; clc;
% Table headings:
fprintf(' t y1 y2 \n');
fprintf('-------------------------------\n');
t=0:0.5:10;
38 MATLAB® Essentials

for i=1:length(t)
y1=t(i)^2/10;
y2=t(i)^3/100;
fprintf('%5i %10.3f %10.3f \n',i,y1,y2);
end
-----------------------------------------------------------------------

Program Results:
t y1 y2
-------------------------------
0.0 0.000 0.000
0.5 0.025 0.001
1.0 0.100 0.010
1.5 0.225 0.034
2.0 0.400 0.080
. . .
. . .
8.0 6.400 5.120
8.5 7.225 6.141
9.0 8.100 7.290
9.5 9.025 8.574
10.0 10.000 10.000
>>
-----------------------------------------------------------------------

In the above example, we selected an element of vector t by the loop vari-

able i. But we did not make y1 and y2 as vectors. Thus, we would not be able
to plot y1 and y2 versus t.
When there is a large output to the Command Window, you might wish to
separate the Command Window from the Editor Window. You can do this
by clicking on the down arrow within the little circle in the black section of
the Command Window and selecting the undock option in the dropdown
window (see Figure 2.13).

FIGURE 2.13
How to undock Command Window from Editor Window?
MATLAB® Fundamentals 39

In the next example the loop variable is not an integer, and thus we would
not be able to use the loop variable to select an element of a vector.

Example 2.4
% Example_2_4.m
% In this example the loop variable is x which is not an integer.
% Thus, the loop variable x could not be used to select an element
% of a vector. The range of x is from -0.9 to +0.9 in steps of 0.1.
clear; clc;
% print the table headings outside of the 'for' loop:
fprintf(' x y1 y2 \n');
fprintf('--------------------------------\n');
for x = -0.9:0.1:0.9
y1=x/(1-x);
y2=y1^2;
fprintf('%5.2f %10.3f %10.3f \n',x,y1,y2);
end
% fprintf('\n %5.2f \n',x)
----------------------------------------------------------------------------

Program Results:
x y1 y2
--------------------------------
-0.90 -0.474 0.224
-0.80 -0.444 0.198
-0.70 -0.412 0.170
-0.60 -0.375 0.141
-0.50 -0.333 0.111
. . .
. . .
0.50 1.000 1.000
0.60 1.500 2.250
0.70 2.333 5.444
0.80 4.000 16.000
0.90 9.000 81.000
>>
----------------------------------------------------------------------------

You might think that the statement for x = −0.9:0.1:0.9 would create a
vector x. However, that is not the case. The process starts by setting x = −0.9.
As the program progresses back to the start of the for loop, old values of x
are overwritten by the new value of x. Try adding the statement

fprintf(' %5.2f \n',x);

at the end of the for loop (by removing the % sign before the fprintf
statement in the above program) and rerunning the program. See that
you only get the last value of x, which is 0.9. Now type x =−0.9:0.1:0.9 in
the Command Window without the semicolon. See that x is now a vector.
40 MATLAB® Essentials

Example 2.5
In this example we will calculate the position and velocity of a free falling body
in a gravitational field (neglecting drag) as a function of time, t. See Figure 2.14. The
governing equations are based on Newton’s second law and can be found in any
university physics textbook.
The governing equations are

V = Vo − g t (2.1)

g t2
y = Vo t − (2.2)
2

where:
V is the velocity
y is the position pointing upward
g is the acceleration of gravity
t is the time

The following MATLAB program calculates and plots V, and y versus t, for 0 ≤ t ≤ 2 s
2
=
in steps of 0.1 s. We have taken Vo 10= m s , and g 9.81 m s . We will print a table
consisting of t, V, and y at every other time step. In addition, we will plot V versus t
and y versus t. Finally, we will determine the approximate maximum height reached
by the free falling body.

t=0 x
O

FIGURE 2.14
Ball in a gravitational field.
MATLAB® Fundamentals 41

% Example_2_5.m
% This program calculates the velocity and position of a free
% falling body vs. time.
% The velocity, V = Vo-gt
% The position, y = Vo*t-0.5*g*t^2
% The initial velocity, Vo=10 m/s, g=9.81 m/s^2
% The output goes to a file named output.txt.
% Plots of y vs. t and V vs. t are made.
% The approximate maximum height reached by the body is determined.
clear; clc;
Vo=10.0; g=9.81;
fo=fopen('output.txt','w');
t=0:0.1:2;
for i=1:length(t)
V(i)=Vo-g*t(i);
y(i)=Vo*t(i)-0.5*g*t(i)^2;
end
plot(t,V), xlabel('t(s)'), ylabel('V(m/s)'), title('V vs. t'), grid;
figure;
plot(t,y), xlabel('t(s)'), ylabel('y(m)'), title('y vs. t'), grid;
ymax=max(y);
fprintf(fo,'The approximate maximum height reached by the body =');
fprintf(fo,' %8.3f m \n',ymax);
% Table headings
fprintf(fo,'t(s) V(m/s) y(m) \n');
fprintf(fo,'--------------------------\n');
for i=1:2:length(t)
fprintf(fo,'%6.2f %10.2f %10.2f \n',t(i),V(i),y(i));
end
---------------------------------------------------------------------------

Program Results:

The approximate maximum height reached by the body = 5.095 m

t(s) V(m/s) y(m)
--------------------------
0.00 10.00 0.00
0.20 8.04 1.80
0.40 6.08 3.22
0.60 4.11 4.23
0.80 2.15 4.86
1.00 0.19 5.09
1.20 -1.77 4.94
1.40 -3.73 4.39
1.60 -5.70 3.44
1.80 -7.66 2.11
2.00 -9.62 0.38
---------------------------------------------------------------------------
See Figures 2.15 and 2.16.
---------------------------------------------------------------------------
42 MATLAB® Essentials

y vs. t
6

4
y(m)

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
t(s)

FIGURE 2.15
y versus t for ball in a gravitational field.

V vs. t
10

2
V(m/s)

−2

−4

−6

−8

−10
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
t(s)

FIGURE 2.16
V versus t for ball in a gravitational field.
MATLAB® Fundamentals 43

FIGURE 2.17
Preallocation message when running Example 2.5.

In running Example 2.5, you may have noticed a small orange line just to the
right of the vertical ladder (see Figure 2.17). If you use your mouse to point
on the orange line you would get the following message: “The variable ‘V’
appears to change size with every loop iteration (within a script). Consider
preallocating for speed.” This would be very important when the number
of repeats in the loop is very large; otherwise, it is not important. Although
MATLAB recommends, but does not require, the preallocation of the size of
the vector or matrix that is being generated, other programs such as C/C++
do require it. To preallocate the size of the vector that is being generated, use
MATLAB’s zeros function. In the above example, 21 v values and 21 y values
will be generated. So add the following statements before the for loop:
v=zeros(21,1) and y=zeros(21,1).

2.9.2 The While Loop

An alternative to the for loop is the while loop. If an index in the program
is required, the use of the while loop statement (unlike the for loop state-
ment) requires that the program generate its own index, as shown in the
following example:
n = 0;
while n < 10
n = n+1;
y = n^2;
end

In the while loop, MATLAB will carry out the statements between the
while and end statements as long as the condition in the while statement is
satisfied. In the above example, when n = 10, none of the commands within
44 MATLAB® Essentials

the while loop will be executed and the program goes to the next command
after the end statement. Note that the statement “n = n+1” above does not
make sense algebraically, but does makes sense in the MATLAB language.
The “=” operator in MATLAB (as in many computer languages) is the assignment
operator that tells MATLAB to fetch the contents in the memory cell containing the
variable n, put its value into the arithmetic unit of the CPU, increment the variable n
by 1, and put the new value back into the memory cell designated for the variable n.
Thus, the old value of n has been replaced by the new value for n.
In this example, we will use a simpler version of Example 2.5, but this time
we will use the while loop instead of the for loop.

Example 2.6
% Example_2_6.m
% This program calculates the velocity and position of a free
% falling body vs. time.
% The program uses a while loop in place of the for loop.
% The velocity, V = Vo-gt
% The position y = Vo*t-0.5*g*t^2
% Vo=10 m/s, g=9.81 m/s^2
% The output goes to a file named output.txt.
clear; clc;
Vo=10.0; g=9.81;
fo=fopen('output.txt','w');
% Table headings
fprintf(fo,'t(s) V(m/s) y(m) \n');
fprintf(fo,'-------------------------\n');
t=0; V=0; y=0;
while t<=2
fprintf(fo,'%6.2f %10.2f %10.2f \n',t,V,y);
t=t+0.2;
V=Vo-g*t;
y=Vo*t-0.5*g*t^2;
end
----------------------------------------------------------------------------

Program Results:
t(s) V(m/s) y(m)
-------------------------
0.00 10.00 0.00
0.20 8.04 1.80
0.40 6.08 3.22
0.60 4.11 4.23
0.80 2.15 4.86
1.00 0.19 5.09
1.20 -1.77 4.94
1.40 -3.73 4.39
1.60 -5.70 3.44
1.80 -7.66 2.11
2.00 -9.62 0.38
----------------------------------------------------------------------------
Compare results obtained from Examples 2.7 and 2.8.
Are they the same?
MATLAB® Fundamentals 45

REVIEW 2.4
1. What is the objective in using a for loop?
2. What is the syntax of a for loop?
3. Should table headings that are not to be repeated be inside a
for loop?
4. If the index of a for loop is used to select an element of a vector
or a matrix, what variable type should the for loop index be?
5. What other statement type can be used to create a loop?
6. What is the major difference between a for loop and a while
loop?

Exercises

E2.2. The motion of a piston in an internal combustion engine is shown in

Figure 2.18a and b.
The piston’s position, s, as seen from the crankshaft center is determined to be

s(t) = r cos( 2πωt) + b 2 − r 2 sin 2 (2πωt) (2.3)

where:
b is the length of the piston rod
r is the radius of the crankshaft
ω is the rotational speed of the crankshaft in revolutions per second

Piston
Displacement
Piston rod

Journal b
s

θ
r

(a) Rotation of crankshaft (b)

FIGURE 2.18
(a) Piston configuration and (b) piston position variables.
46 MATLAB® Essentials

Develop a MATLAB program that determines s versus t for 0 ≤ t ≤ 0.01 s.

Use 20 subdivisions on the t domain. Take r = 9 cm, ω = 100 revolutions per
second, and b = 14 cm.

1. Create a table of s versus t and print the results to both the screen
and to file.
2. Create a plot of s versus t.

E2.3. The position, y, of a mass in a mass-spring-dashpot system (see

Figure 2.19) is given by

  2   2  
 c   k  c    k  c  
y = exp  − t   A sin − t + B cos − t
 2m   m  2 m    m  2 m    (2.4)
    
 

Take
m = 25.0 kg
c is the damping factor = 5.0 N-s/m
k is the spring constant = 200.0 N/m
A = 5.0 m
B = 0.25 m

1. Determine y(t) for 0 ≤ t ≤ 10 seconds in steps of 0.1 seconds.

2. Create a table of y versus t every 1 second and print the results to the
screen.
3. Create a plot of y versus t.

Unstretch
–
y0
Position Static
Equilibrium ky cy′
– y
y position
–
y+–
y0 = –
y
w M M

w w

FIGURE 2.19
Mass-spring-dashpot system.
MATLAB® Fundamentals 47

6m

Center of hoop

Vo

ϑo
3.048 m

1.98 m y

Floor
x

FIGURE 2.20
Basketball player shooting the basketball.

E2.4. A basketball player shoots the ball when he is 6 m from the center of the
hoop as shown in Figure 2.20. The ball is released at a velocity, Vo , and makes
angle ϑo = 40° with the horizontal. Using Newton’s second law and the initial
conditions and neglecting the drag on the basketball, we can determine the
following equations for the (x, y) position of the ball as a function of time, t.

x = Vo cos(ϑo )t (2.5)

g 2
y = yo + Vo sin(ϑo )t − t (2.6)
2

Take the (x, y) position of the center of the hoop to be (xf , yf ) = (6.0 m, 3.048 m),
yo = 1.98 m, and ϑo = 40º.

1. Determine the time, tf , which it takes for the ball to reach the center
of the hoop. Time, t, equals zero when the ball leaves the  player’s
hands.
2. Determine the velocity, Vo , that will result in the ball reaching the
center of the hoop at time tf .
3. Create a table consisting of t, x, y for 0 ≤ t ≤ t f in steps of t f 10. Carry
variables to 4 decimal places. Print the table to an output file, includ-
ing tf and Vo .
4. Create a plot of y versus x.

Hint: Solve Equation 2.5 for Vo and substitute the expression for Vo into
Equation 2.6, giving an expression involving t, x, and y. In that expression,
set t = tf , x = x f , and y = yf . In the resulting equation, tf is the only unknown.
Use this expression in your MATLAB program to solve for tf .
48 MATLAB® Essentials

E2.5. A small sphere moving though a fluid at a slow velocity will have a
drag force acting on it, which is described by Stokes’ Law. The sphere could
be a dust particle or a raindrop moving in air, or a ball bearing moving in oil.
The drag force described by Stokes’ Law is

D = 6πRµV (2.7)

where:
D is the drag
R is the radius of the sphere
µ is the viscosity of the fluid
V is the velocity of sphere

Let us consider a steel ball bearing dropped in oil (see Figure 2.21) with an
initial velocity of zero. The ball bearing will drop with a varying velocity
until it reaches a final velocity (terminal velocity, VT). The forces acting on
the ball bearing are the gravitational force, W, buoyancy force, B, and the
drag force, D. The buoyancy force is equal the weight of the fluid displaced.
The equations for W and B are

W = ρsteel υg (2.8)

B = ρoil υg (2.9)

where:
ρsteel is the mass density of steel
ρoil is the mass density of oil
υ is the volume of sphere = 4 3 πR3
g is the gravitational constant

y B

FIGURE 2.21
Ball bearing in oil.
MATLAB® Fundamentals 49

The terminal velocity occurs when

W −B−D = 0 (2.10)

By applying Newton’s second law to the sphere we can determine V and V T

that are
W −B
VT = (2.11)
6πRµ

 −
6πRµ g
t
V = VT  1 − e W  (2.12)
 
 

Take µ = 3.85 (N − s)/m 2 , R = 0.01 m, ρsteel = 7910 kg / m 3 , ρoil = 899 kg/m 3,

g = 9.81 m/s 2
Create a MATLAB program that will

1. Determine the weight of the ball bearing, W.

2. Determine the buoyancy force, B.
3. Determine the terminal velocity, V T .
4. Use a while loop to determine V as a function of time, for 0 ≤ t ≤ 0.3 s
in steps of 0.01 s.
5. Create and print to a file values for W, B, and V T and a table contain-
ing t and V.
6. Plot V versus t.

E2.6. The voltage in a parallel resistance, inductor, and capacitor (RLC) circuit
(see Figure 2.22) is given by

  2   2  
 1   1  1  1  1
v = exp  − t   A exp    − t  + B exp    − t   (2.13)
 2 RC    2 RC  LC    2RC  LC 
    
t=0

iR iL iC
+ + +
Io(t) vC
vR R vL L C

− − −

FIGURE 2.22
A parallel resistance, inductor, and capacitor (RLC) circuit.
50 MATLAB® Essentials

Take R = 10 ohm, L = 1.0e-2 henry, C = 1.0e-6 farad, A = 6.0 V, and B = −8.9 V.

1. Determine v(t) for 0 ≤ t ≤ 5.0 × 10 −4 seconds. Use 100 subdivisions on

the time domain.
2. Print out a table of v versus t every fourth subdivision.
3. Create a plot of v versus t.

2.10 Input
Engineers who carry out tests on a piece of equipment may need to enter
data into an existing computer program for analysis. There are several com-
mands that may be used to enter data from a data file. These are the load,
dlread, and fscanf commands. We will discuss them one at a time. The
analysis program may also ask the user to input data from the keyboard.
To enter data from the key board, use MATLAB’s input command, which
makes the program interactive.

2.10.1 The Load Command

One of the commands that allow the user to enter data from a file is the load
command. The data file is likely to only contain numbers and would have a
specific name. The syntax for the command is

load filename.txt

The input file must have the same number of columns in each row and must be in
the same folder as the program loading the data file. The data file should only have
numbers. Suppose we had the following data file named atm_properties and
we wish to enter the data into a program. Here, the first column is altitude
in meters, the second column is temperature in degrees Kelvin (K), the third
column is pressure in Pascal (Pa), and the fourth column is density in
kilogram/meter3 (kg/m3).

0 288.15 1.0133e+005 1.2252

1000 281.65 8.9869e+004 1.1118
2000 275.15 7.9485e+004 1.0065
3000 268.65 7.0095e+004 0.9091
4000 262.15 6.1624e+004 0.8191
5000 255.65 5.4002e+004 0.7360
MATLAB® Fundamentals 51

N O T E : Before you can run the following example (Example 2.9), you need to
create the data file shown on the previous page. To do this, copy the data,
then go to MATLAB and open up a new script window and paste the data
into the new script window. Then click on the Save icon and save the file as
atm_properties.txt. To save the file as a .txt file, click on the down arrow in
the Save as type box and select All files (*.*).
Alternatively, you can open up a new script window and type in the data,
and then follow the instructions described above. Try typing or copy-
ing the following program into the Editor Window in MATLAB and
running it.

Example 2.7
% Example_2_7.m
% This program uses the load command to load the data in the file
% named atm_properties.txt into this program. Column 1 is the
% altitude, column 2 is the temperature, column 3 is the pressure
% and column 4 is the density. Altitude is in meters (m),
% temperature is in degrees Kelvin (K), pressure is in Pascal (Pa)
% and density is in (kg/m^3).
% The program also is an example of using the colon operator to
% create the vectors z,T,p and rho.
% The program also demonstrates the use of the fprintf command.
% The program prints out elements of the vectors z,T,p and rho.
clear; clc;
load('atm_properties.txt');
% establishing variable names to scanned file.
z=atm_properties(:,1);
T=atm_properties(:,2);
p=atm_properties(:,3);
rho=atm_properties(:,4);
fprintf('z(m) T(K) p(Pa) rho(m^3/kg) \n');
fprintf('-------------------------------------------\n');
for j=1:length(z)
fprintf('%5.0f %8.2f %10.3e %8.4f \n',z(j),T(j),p(j),rho(j));
end
------------------------------------------------------------------------

Program Results:
z(m) T(K) p(Pa) rho(m^3/kg)
-------------------------------------------
0 288.15 1.013e+05 1.2252
1000 281.65 8.987e+04 1.1118
2000 275.15 7.949e+04 1.0065
3000 268.65 7.010e+04 0.9091
4000 262.15 6.162e+04 0.8191
5000 255.65 5.400e+04 0.7360
>>
---------------------------------------------------------------------------
52 MATLAB® Essentials

2.10.2 The dlmread Command

An alternative to the load command is the dlmread command. This com-
mand will read an ASCII delimited file. All data in the file must be numeric.
In this example, the entire data file in atm_properties.txt is specified as a matrix
Y consisting of six rows and four columns. Then the colon operator is used to
create vectors z, T, p, and rho. To demonstrate the use of the dlmread com-
mand, we will modify Example 2.7 by replacing the lines starting with

load (atm_properties.txt);

and ending with

rho = atm_properties(:,4);

with

Y=dlmread('atm_properties.txt');
z=Y(:,1);
T=Y(:,2);
p=Y(:,3);
rho=Y(:,4);

The modification of Example 2.7 (omitting the comment lines) would be

clear; clc;
Y=dlmread('atm_properties.txt');
% establishing variable names to scanned file.
z=Y(:,1);
T=Y(:,2);
p=Y(:,3);
rho=Y(:,4);
fprintf(' z(m) T(K) p(Pa) rho(m^3/kg) \n');
fprintf('------------------------------------------------------\n’);
for j=1:length(z)
fprintf('%5.0f %8.2f %10.3e %8.4f \n',z(j),T(j),p(j),rho(j));
end

2.10.3 fscanf Command

Students who have a background in C/C++ may use the fscanf com-
mand to enter data into a program. The commands necessary to this are
shown below.

A = zeros(n, m);
fi = fopen('filename.txt','r');
[A] = fscanf(fi,'%f',[n,m]);
MATLAB® Fundamentals 53

where n × m is the number of elements in the data file. The 'r' in the fopen
statement tells MATLAB that this file is for reading in data. The n × m
matrix is filled in column order. Thus, rows become columns and columns
become rows.
The following example program enters the data in atm_properties.txt into
the program.
NO T E : Before you can run program Example_2_8.m, the data file atm_properties.
txt had to be created.

Example 2.8
% Example_2_8.m
% This program uses fscanf command to load the data in the file
% named atm_propeties.txt into this program. Column 1 is the
% altitude, column 2 is the temperature, column 3 is the pressure
% and column 4 is the density. Altitude is in meters (m)
% temperature is in degrees Kelvin (K), pressure is in Pascal (Pa)
% and density is in (kg/m^3).
clear; clc;
fi = fopen('atm_properties.txt','r');
% Print A to the screen and see that columns of the data file
% became rows.
A = fscanf(fi,'%f',[4,6])
% establishing variable names to scanned file.
z=A(1,:);
T=A(2,:);
p=A(3,:);
rho=A(4,:);
fprintf('z(m) T(K) p(Pa) rho(m^3/kg) \n');
fprintf('-------------------------------------------\n');
for j=1:length(z)
fprintf('%5.0f %8.2f %10.3e %8.4f \n',z(j),T(j),p(j),rho(j));
end
------------------------------------------------------------------------

Program Results:
z(m) T(K) p(Pa) rho(m^3/kg)
-------------------------------------------
0 288.15 1.013e+05 1.2252
1000 281.65 8.987e+04 1.1118
2000 275.15 7.949e+04 1.0065
3000 268.65 7.010e+04 0.9091
4000 262.15 6.162e+04 0.8191
5000 255.65 5.400e+04 0.7360
>>
--------------------------------------------------------------------------
54 MATLAB® Essentials

2.10.4 The input Command

The MATLAB command that the programmer can use to have the user
enter data from the keyboard is the input command. The program should
pause and move the cursor to the Command Window (without providing
a prompt sign) waiting for the user to enter the data requested. However,
in MATLAB version R2016A, the cursor stays in the Editor Window. This is
a bug in this version of the MATLAB program. This bug was eliminated
in MATLAB version R2016B. This was not a problem in earlier versions of
MATLAB. The use of the input command makes the program interactive.
Suppose that you are the programmer and you wish to have the user enter
a matrix named z, from the keyboard, use

Z = input('Enter matrix Z enclosed by brackets \n')

The user will see the following on the screen:

Enter matrix Z enclosed by brackets

If you are using MATLAB version R2016A, you will need to click on the
Command Window to enter the data. If you are not using MATLAB version
R2016A, the user can type information in the Command Window without
having to first click in the Command Window. The user should then type in
something like

[ 5.1 6.3 2.5; 3.1 4.2 1.3 ]

 5.1 6.3 2.5 

Thus, Z =  .
 3.1 4.2 1.3 

Note that the argument to input command is a character string enclosed

by the single quotation marks. The character string will be printed to the
screen as shown above. If the response to the input statement is a charac-
ter or a string, you need to enclose the character or the string with single
quotation marks. However, you can avoid this requirement by entering a
second argument of 's' to the input command as shown in the following
statement:

response = input('Print Z to a file? (y/n):\n', 's')

In this case, the user can respond with either a y or n (without single quotation
marks). An example using this concept will be given in Chapter 3.
MATLAB® Fundamentals 55

E2.7. Write a MATLAB program that uses the input command to enter the
following three vectors:

Z = [ 0 1000 2000 3000 4000 5000 ]

T = [ 288.1 281.6 275.1 268.6 262.1 255.6 ]

rho = [1.2252 1.1118 1.0065 0.9091 0.8191 0.7360 ]

Z is altitude in m, T is temperature in K, and rho is density in kg/m3.

Then plot T versus Z and rho versus Z.

REVIEW 2.5
1. Name four commands that can be used in a script to input data
into the workspace.
2. Which of the four commands makes the program interactive?

2.11 More on MATLAB Graphics

2.11.1 The figure Command
As mentioned earlier, if a program involves creating more than one plot, you
need to include the statement figure after each plot command (except the
last), otherwise only the last plot will appear. The following example pro-
gram produces two separate plots.

Example 2.9
% Example_2_9.m
% This program creates two separate plots.
% First y1=t^2/10 is plotted with 0 <= t <= 10,
% then y2=t^3/100 is plotted over the same t range.
% To plot y1 and y2 vs. and t, they need to be made vectors.
clear; clc;
t=0:0.5:10;
for n=1:length(t)
y1(n)=t(n)^2/10;
y2(n)=t(n)^3/100;
end
plot(t,y1), xlabel('t'), ylabel('y1'), grid, title('y1 vs. t');
figure;
plot(t,y2), xlabel('t'), ylabel('y2'), grid, title('y2 vs. t');
------------------------------------------------------------------------
56 MATLAB® Essentials

Program Results:
See Figure 2.23a and b.
------------------------------------------------------------------------

y1 vs. t
10

5
y1

0
0 1 2 3 4 5 6 7 8 9 10
(a) t
y2 vs. t
10

5
y2

0
0 1 2 3 4 5 6 7 8 9 10
(b) t

FIGURE 2.23
(a) Plot of y1 versus t and (b) plot of y2 versus t.
MATLAB® Fundamentals 57

2.11.2 Multiple Plots

Suppose in matrix A, shown below, we wished to plot column 2 versus
column 1, column 3 versus column 1, and column 4 versus column 1.

 t1 y1 z1 w1 
t y2 z2 w2 
A=
2

    
 
tn yn zn wn 

We could let T = A( : , 1 ), Y = A( : , 2 ), Z = A( : , 3 ), and W = A( : , 4 ), giving

 t1   y1   z1   w1 
t  y  z  w 
T =  , Y =  , Z =  , W =  
2 2 2 2

     

       
tn   yn   zn   wn 

Then to plot Y versus T, Z versus T and W versus T all on the same graph,
we would write,

plot(T,Y,T,Z,T,W);

Of course, we could have avoided the additional steps by writing

plot(A(:,1),A(:,2),A(:,1),A(:,3),A(:,1),A(:,4))

To identify which curve goes with which variable, you can add text to the
plot with the command,

text(x,y,'text statement');

where (x, y) are the coordinates on the graph where the text statement
will start.
Multiple curves on the same graph can be distinguished by color coding
the curves.
Available color types:

black 'k'
blue 'b'
green 'g'
red 'r'
cyan 'c'
yellow 'y'
58 MATLAB® Essentials

Multiple curves on the same graph can also be distinguished by using

different types of lines.
Available line types:

solid (default)
dashed '--'
dashed-dot '-.'
dotted ':'

Alternatively, you can create a marker plot of discrete points (without a line)
by using one of these marker styles:

point '.'
plus '+'
star '*'
circle 'o'
x-mark 'x'
diamond 'd'

The legend command may also be used in place of the text command to
identify the curves. The format for the legend command is

legend('text1', 'text2')

The legend box may be moved by clicking on the box and dragging it to the
desired position.
You can also change the axis limits in a plot by using the command

axis([xmin xmax ymin ymax])

(See Example 2.11)

Example 2.10
The following example illustrates a multiple plot program:

% Example_2_10.m
% This program creates a simple table and a multiple plot.
% First a table of y1=t^2/10 and y2=t^3/100 is created.
% To plot y1, y2 vs. and t, they need to be made vectors.
% y1 and y2 vs. t are plotted on the same graph.
clear; clc;
t=0:10;
for n=1:length(t)
y1(n)=t(n)^2/10;
y2(n)=t(n)^3/100;
end
MATLAB® Fundamentals 59

% By making t, y1 and y2 as vectors, their values can be printed

% outside the for loop that created them.
% Column headings
fprintf(' t y1 y2 \n');
fprintf('----------------------------------\n');
for n=1:length(t)
fprintf('%8.1f %10.2f %10.2f \n',t(n),y1(n),y2(n));
end
% Create the plot, y1 as a solid line, y2 as a dashed line.
% Note: the variables t, y1,y2 need to be vectors in the plot
% command.
plot(t,y1,t,y2,'--');
xlabel('t'), ylabel('y1,y2'), grid, title('y1 and y2 vs. t');
% Plot identification is also established by adding text to the plot.
text(6.5,2.5,'y2');
% In the above statement, 6.5 is the abscissa position and 2.5 is
% the ordinate position where the 'y1' label will be positioned.
text(4.2,2.4,'y1'),
% We can also use the legend command to identify the curves
legend('y1','y2');
-------------------------------------------------------------------------

Program Results:
t y1 y2
-------------------------------------
0.0 0.0000 0.0000
1.0 0.1000 0.0100
2.0 0.4000 0.0800
3.0 0.9000 0.2700
4.0 1.6000 0.6400
5.0 2.5000 1.2500
6.0 3.6000 2.1600
7.0 4.9000 3.4300
8.0 6.4000 5.1200
9.0 8.1000 7.2900
10.0 10.0000 10.0000
>>
See Figure 2.24.
------------------------------------------------------------------------

2.11.3 The hold on Command

In the above example, we used a single plot command to plot both y1 and
y2, that is, plot(t,y1,t,y2,'--'). However, we could also have plot-
ted both y1 and y2 on the same graph by plotting each separately with
the command hold on between the plots. We would be superimposing
the second plot onto the first plot. To do this, replace the plot command
plot(t,y1,t,y2,'--')with
plot(t,y1);
hold on
plot(t,y2);
xlabel('t'), ylabel('y1,y2'), grid, title('y1 and y2 vs. t');
60 MATLAB® Essentials

y1 and y2 vs. t
10

6
y1,y2

3
y1 y2
2
y1
1 y2

0
0 1 2 3 4 5 6 7 8 9 10
t

FIGURE 2.24
Plots of y1 and y2 on the same graph.

Example 2.11
The following example illustrates the plotting of trigonometric functions. The exam-
ple also illustrates that simple arithmetic can be used in the arguments of the trig-
onometric functions. This is also true for other built-in MATLAB functions. If the
resulting curves are not smooth than we would need to use more points to properly
display the curves.
% Example_2_11.m
% This script calculates both sin(2x/3), sin(2x/3)^2
% and cos(2x/3+pi) for -pi <= x <= pi. The x domain is subdivided
% into 50 subdivisions. The script plots the 3 functions and
% determines the absolute maximum values of the vectors fsin, fsinsq
% and fcos and prints those values to the screen.
clear; clc;
x=-pi:2*pi/50:pi;
for i=1:length(x)
fsin(i)=sin(2*x(i)/3);
fsinsq(i)=sin(2*x(i)/3)^2;
fcos(i)=cos(2*x(i)/3+pi);
end
MATLAB® Fundamentals 61

fsin_max=max(abs(fsin)); fcos_max=max(abs(fcos));
fsinsq_max=max(fsinsq);
fprintf('fsin_max=%10.5f, fcos_max=%10.5f \n',fsin_max, fcos_max);
fprintf('fsinsq_max=%10.5f \n',fsinsq_max);
plot(x, fsin, x,fcos,'--',x, fsinsq,'-.'), xlabel('x'),
ylabel('fsin, fcos, fsinsq'), grid,
title('fsin, fcos, fsinsq vs. x'), legend('fsin','fcos','fsi nsq');
-------------------------------------------------------------------------

Program Results:
From the Command Window:

fsin_max= 0.99978, fcos_max= 1.00000

fsinsq_max= 0.99956
>>

See Figure 2.25.

-------------------------------------------------------------------------

fsin, fcos, fsinsq vs. x

1

0.8

0.6

0.4
fsin,fcos,fsinsq

0.2

−0.2

−0.4

−0.6
fsin
fcos
−0.8 fsinsq
−1
−4 −3 −2 −1 0 1 2 3 4
x

FIGURE 2.25
Plot of fsin, fcos, and fsinsq versus x.
62 MATLAB® Essentials

2.11.4 Plotyy Command

Suppose we have two functions of the same variable but the numerical range
of the two functions differ significantly and we would like to display the
functions on the same plot. This can be done using the plotyy function.
In the next example we plot position, y, and velocity, V of a free falling body
vs. time on the same graph (see Example 2.6).

Example 2.12
% Example_2_12.m
% This script is a modification of Example 2.5. In this script
% both y and V axes are plotted on the same graph. The y axis
% is on the left side and the V axis is on the right side.
clear; clc;
Vo=10.0; g=9.81; t=0:0.1:2;
for i=1:length(t)
V(i)=Vo-g*t(i); y(i)=Vo*t(i)-0.5*g*t(i)^2;
end
plotyy(t,y,t,V), xlabel('t(s)'), title('y vs. t and V vs. t'), grid,
yyaxis left, axis([0 2 0 10]), ylabel('y(m)'), text(0.32,2.5,'y');
yyaxis right; axis([0 2 -10 10]), ylabel(' V(m/s)'), text(0.6,5.0,'V');
------------------------------------------------------------------------

Program Results:
See Figure 2.26.
----------------------------------------------------------------------

y vs. t and V vs. t

10 10

6
V
4

2
V(m/s)
y(m)

5 0

−2

−4
y
−6

−8

0 −10
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
t(s)

FIGURE 2.26
Plot of y versus t and V versus t on the same graph.
MATLAB® Fundamentals 63

2.11.5 The subplot Command

Suppose you want to plot each of several curves as a separate plot, but all
on the same page. The subplot command provides the means to do so. The
command subplot(m,n,p) breaks the page into an m by n matrix of small
plots, and p selects the matrix position of the plot. The subplot command
is a positioning command and not a plot command. The following example
demonstrates the use of the subplot command.

Example 2.13
% Example_2_13.m
% This program is an example of the use of the subplot command.
% Values of y1, y2, y3 and y4 are constructed as
% vectors. Separate plots of y1 vs. t, y2 vs. t, y3 vs. t,
% and y4 vs. t are plotted on the same page.
clc; clear;
t=0:0.5:10;
for n=1:length(t)
y1(n)=t(n)^2/10;
y2(n)=sin(pi*t(n)/10);
y3(n)=exp(t(n)/2);
y4(n)=sqrt(t(n));
end
subplot(2,2,1),
plot(t,y1), grid, title('y1 vs. t'), xlabel('t''), ylabel('y1');
subplot(2,2,2),
plot(t,y2), grid, title('y2 vs. t'), xlabel('t'), ylabel('y2');
subplot(2,2,3),
plot(t,y3), grid, title('y3 vs. t'), xlabel('t'), ylabel('y3');
subplot(2,2,4),
plot(t,y4), grid, title('y4 vs. t'), xlabel('t'), ylabel('y4');
----------------------------------------------------------------------------

Program Results:
See Figure 2.27.
---------------------------------------------------------------------------

2.11.6 Bar Charts

I find that bar charts appear more often in business related topics than in
engineering topics. For example, you might wish to compare several compa-
nies yearly profit percentages. A convenient way to do this is by the use of a
bar chart. The syntax the bar charts are

bar(y)
bar(x,y)
bar(___, width)
bar(___, style)
bar(___, color)
64 MATLAB® Essentials

y1 vs. t y2 vs. t
10 1

5 0.5
y1

y2
0 0
0 5 10 0 5 10
t t
y3 vs. t y4 vs. t
150 4

3
100
2
y3

y4

50
1

0 0
0 5 10 0 5 10
t t

FIGURE 2.27
Plots of y1, y2, y3, and y4 versus t on the same page.

See MATLAB help for more examples on bar charts.

Example 2.14
In this example, we compare the profit percentage gained for year 2015 for Companies
A, B, C, D, E, F, and G. These are given in Table 2.1

% Example_2_14.m
% This script is an example of creating a bar chart.
% The script lists and plots the percentage gains in profits for
% several companies for the year 2015.
clear; clc;
y = [2.51 -0.13 3.16 4.72 1.2 6.5 3.8];
bar(y,0.4);
ylabel('% profit, year 2015');
title('1=CO.A, 2=CO.B, 3=CO.C, 4=CO.D, 5=CO.E, 6=CO.F, 7=CO.G');
----------------------------------------------------------------------

Program Results:
See Figure 2.28.
-----------------------------------------------------------------------
MATLAB® Fundamentals 65

TABLE 2.1
Companies Profits Percentage Gain for 2015
Company % Profit Gain for Year 2015

A 2.51
B −0.13
C 3.16
D 4.72
E 1.60
F 6.50
G 3.80

1 = CO.A, 2 = CO.B, 3 = CO.C, 4 = CO.D, 5 = CO.E, 6 = CO.F, 7 = CO.G

7

5
% profit, year 2015

−1
1 2 3 4 5 6 7

FIGURE 2.28
Bar chart. Percent profit gain for several companies for year 2015.

2.11.7 Pie Charts

As with bar charts, pie charts appear more often in business and finances than
in engineering. For example, you might wish to know what percentage of your
investments are in U.S. stocks, foreign stocks, mutual funds, bonds, and cash.
The syntax for the pie chart for this example could be

X = 1:5;
labels = {'US stocks','foreign stocks','mutual funds',...
'bonds','cash'};
pie(X,labels)
66 MATLAB® Essentials

Example 2.15
Suppose we wished to plot the percentage of several different types of investments
made by a particular investor. Table 2.2 gives the percentage of different types of
investments for that individual.
The program follows:

% Example_2_15.m
% This program gives the percentage of various types of investments
% made by a particular individual. The percentages are displayed in
% a pie chart.
clear; clc;
x=[43 12 15 20 10];
labels = {'US stocks','foreign stocks','mutual funds',...
'bonds','cash'};
pie(x, labels);
-----------------------------------------------------------------------

Program Results:
See Figure 2.29.
-------------------------------------------------------------------------

TABLE 2.2
Investment Percentages
Investments Percentage

US stocks 43
Foreign stocks 12
Mutual funds 15
Bonds 20
Money market 10

cash

bonds
US stocks

mutual funds

foreign stocks

FIGURE 2.29
Pie chart for several different types of investments.
MATLAB® Fundamentals 67

Example 2.16
• The sprintf Command
The sprintf is the same as fprintf command except that it returns the print data
as a MATLAB variable rather than writing to the Command Window or to a file.
When plotting several different plots on the same page, you may wish to vary the
titles of the plots depending on the specific variable defined in the program. This is
demonstrated in the following example:

% Example_2_16.m
% This program is an example of the use of the subplot and
% the sprintf commands.
% Plots of y=sin(k*pi*t/L) for several values of k are created and
% plotted on the same page.
clc; clear;
t=0:0.1:10;
k=[2 4 6 8];
L=10;
for m=1:length(k)
for n=1:length(t)
y(n)=sin(k(m)*pi*t(n)/L);
end
subplot(2,2,m), plot(t,y), xlabel('t'), ylabel('y'), grid,
title(sprintf('y vs.t, k=%3i \n',m));
end
-------------------------------------------------------------------------
Program Results:
See Figure 2.30.
-----------------------------------------------------------------------
y vs. t, k = 1 y vs. t, k = 2
1 1

0 0
y

−1 −1
0 5 10 0 5 10
t t
y vs. t, k = 3 y vs. t, k = 4
1 1

0 0
y

−1 −1
0 5 10 0 5 10
t t

FIGURE 2.30
Four plots of y versus t, all on one page. Each plot is for a different value of k.
68 MATLAB® Essentials

Exercises

E2.8. This exercise involves plotting the temperatures of a spherical object

dropped into a fluid contained in a vertical circular cylinder. The tempera-
ture variation of both the sphere, T_sphere, and the fluid, T_ fluid, are given as
a function of time, t, in the following three vectors:

t = [ 0 0.04 0.08 0.12 0.16 0.20 0.24 0.28 0.32 0.36 0.40 0.44 0.48 0.52 0.56 0.60 ]

T _ sphere = [150 124 104 89 77 67 60 54 49 46 43 41 39 38 37 36 ]

T_ fluid = [20.0 22.9 25.2 27.0 28.3 29.5 30.3 31.0 31.9 32.2 32.5 32.7 32.8 32.9]

In MATLAB, create a plot of both T_sphere and T_ fluid versus t on the same
graph, t is in seconds and T_sphere and T_ fluid are in degrees C.

E2.9. This exercise involves the x position and x component of the velocity,
u, of a package dropped from an airplane as a function of time, t. These vari-
ables are specified in vectors x and u and t, respectively.

t = [ 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 ]

x = [ 0.0 23.6 44.7 63.6 80.6 95.8 109.5 121.7 132.7 142.4 150.9]

u = [ 50.0 44.6 39.9 35.8 32.2 28.9 25.9 23.1 20.6 18.2 15.9]

Create a MATLAB program that will plot x versus t and u versus t as two
separate plots, but both on the same page.

2.11.8 Greek Letters and Mathematical Symbols

Greek letters and mathematical symbols can be used in xlabel, ylabel,
title, and text by spelling out the Greek letter and preceding it with a ‘\’
(backslash character). Thus, to display ω, use \omega, and to display β, use
\beta.

Example:
ylabel('\omega'), title('\omega vs. \beta'), text(10,5,'\omega');

For an additional list of Greek symbols and some special characters, see
Appendix A. You may also occasionally need to print a “'” character in your label
or title. In this case, use a double-quotation mark as shown here '' to escape the
single-quote character in your string. Thus, to generate the plot title “Signal 'A'
vs. Signal 'B'”, you would type

title('Signal ''A'' vs. Signal ''B''')

MATLAB® Fundamentals 69

2.11.9 Interactively Annotating Plots

As an alternative to adding the xlabel, ylabel, and title commands
into your program, you can create the plot, then click on the Insert
option in the menu bar in the plot window and choose X Label from
the dropdown menu. This will highlight a box in which you can type in
the abscissa variable name. You can repeat this process for the Y Label
and the Title of the plot. Other options available in the Insert Menu
are TextBox, Text Arrow, Arrow, and others. When you click any one
of these options, a cross-hair will appear and you can then move the item
to the location where you want it to appear, then left-click the mouse to
fix the location. You can then type in the desired text. To remove the out-
lines of a TextBox, place the cursor in the TextBox and right-click the
mouse. This will bring up a dropdown menu, then select Line Style,
and then left-click on none. This will remove the lines from the TextBox.

2.11.10 Saving Plots

To save a plot, click on the File in the plot window and select the Save
option from the dropdown menu. This produces a window where you can
enter a file name. The disadvantage of this method is that if you decide to
rerun the script, the items that you manually inserted will not be saved. If
you wish to copy the figure into a report, you can click on Edit in the plot
window, and then select Copy Figure from the dropdown menu. You can
then paste the figure into your report. If you need a monochrome version
of your plot (for best reproduction on a photocopier), you can make all of
your curves black by choosing File from the task bar menu, then select-
ing Export Setup from the dropdown menu. This will open a window in
which you need to click on Rendering, and change the Colorspace to
black and white.
There are many more options available in the plot window, however we
leave it to the student to explore it further.

REVIEW 2.6
1. When there is more than one function plotted on a graph, what
are the ways to identify which curve goes with which function?
2. What is the name of the function that will allow you to plot
several graphs on one page?
3. How does one enter Greek symbols into a plot?
4. What are the commands that will allow you to enter text onto
a plot once the plot has been created?
70 MATLAB® Essentials

Projects
P2.1. A tennis player on serve wishes to place the tennis ball close to the
outside line of the service box when the ball hits the ground (see Figure P2.1a
and b). The horizontal distance from the point where the ball leaves the
racket to where the ball hits the ground is 19.33 m. The vertical distance, yo ,
above the ground when the ball leaves the racket is 2.36 m. The angle that the
ball makes with the horizontal on leaving the players racket is 5.7° pointing
down. Neglecting drag, the governing equations describing the motion of
the ball are

x = Vo cos(ϑ) t (P2.1a)

g 2
y=− t − Vo sin(ϑ) t + yo (P2.1b)
2

In the above equations, x and y are in (m), t is in (s) and Vo is in (m/s).

Let (xf , yf ) be the x and y positions where the ball hits the ground and tf the
time when this occurs.

1. Determine the time, tf . Time, t, equals zero when the ball leaves the
racket.
2. Determine the velocity, Vo , that will result in the ball reaching the
ground at time tf .

Ball hits ground

Ball path

Ball leaves
racket
Net
(a)
y
5.7° Ball leaves racket

Ball hits ground

Vo yo
0.914 m Net
x
12.569 m
19.33 m
(b)

FIGURE P2.1
Tennis player on serve: (a) plan view and (b) vertical view.
MATLAB® Fundamentals 71

3. Create a table consisting of t, x, y for 0 ≤ t ≤ t f in steps of t f 10.

Carry variables to 4 decimal places. Print the table to an output file,
including tf and Vo .
4. Create a plot of y versus x.
5. Determine the height of the tennis ball at the position of the net.

P2.2. A batter in a baseball game hits a ball to right center field. The ball leaves
the bat at a 30° angle with the horizontal at a speed of Vo and at a height of
1.5 m above the ground. The center fielder is 71 m from home plate and the
angle that the horizontal line connecting the center fielder with home plate
makes angle of 10° with the horizontal path of the ball, see  Figure P2.2a
and b. The center fielder sees the direction of the fly ball and starts to run
toward the path of the ball at an average speed, S and 0.5 s after the ball is
hit. The center fielder catches the ball when it is just 1.8 m above the ground
and 91 horizontal meters from the initial position of the ball as it leaves the bat.
Neglecting drag, the governing equations describing the motion of the ball are

x = Vo cos(ϑ ) t (P2.2a)

g 2
y=− t − Vo sin(ϑ ) t + yo (P2.2b)
2

1. What is the time of flight, tf , when the ball is caught?

2. What is the initial velocity, Vo , of the ball when it leaves the bat?

Position of Ball is caught

center fielder
at t = 0

10°
71 m 91 m

(a) Baseball leaves bat

y Vo Ball is caught
Baseball
leaves bat 30°
Ground
1.5 m 1.8 m
x
91 m
(b)

FIGURE P2.2
Batter hitting baseball: (a) Plan view and (b) vertical view.
72 MATLAB® Essentials

3. What is the average speed, S, of the center fielder as a runs to catch

the ball.
4. Create a table consisting of t, x, y for 0 ≤ t ≤ t f in steps of t f 10. Carry
variables to 4 decimal places. Print the table to the Command
Window, include tf , Vo , and S.
5. Create a plot of y versus x.

P2.3. Although atmospheric conditions vary from day-to-day, it is convenient

for design purposes, to have a model for atmospheric properties as a function
of altitude. The U.S. Standard Atmosphere, modified in 1976, is such a model.
For altitudes less than or equal to 11,000 m, the governing equations for the
air temperature, pressure, and density are as follows:

g
 λz  λ R
p = po  1 −  (P2.3a)
 To 

T = To − λz (P2.3b)

p
ρ= (P2.3c)
RT

where:
z is the altitude
To = 288.15 K (the temperature at z = 0)
po = 1.01325 × 105 Pa (the pressure at z = 0)
R = 287 J/(kg-K) (the gas constant for air)
g = 9.81 m/s2 (the gravitational constant for air)
λ = 0.0065 K/m (the lapse rate)
ρ is the air density (kg/m3)

Calculate atmospheric properties of temperature, T, pressure, p, and density, ρ,

every 1000 m from z = 0 (sea level) to z = 11,000 m and print the results to a
file in a table format. Also plot T versus z, p versus Z, and ρ versus z as three
separate plots, all on the same page.

P2.4. The properties of specific volume, v, and pressure, p, as a function of

temperature, T, for carbon dioxide based on the Redlich–Kwong Equation
of state are given in Table P2.1:
Plot v versus T and p versus T as two separate plots.
MATLAB® Fundamentals 73

TABLE P2.1
Specific Volume and Pressure versus Temperature
T (K) v (m3/kmol) p (bar)

350 0.28 7.65

400 0.32 8.57
450 0.36 9.16
500 0.40 9.55
550 0.44 9.81
600 0.48 10.00
650 0.52 10.14
700 0.56 10.24
750 0.60 10.31

P2.5. In this project, we consider two cars on a collision course (see

Figure P2.3). Each car’s initial position and the angle its path makes with the
x-axis is specified below. The speed of car1 is also specified.
Initial position of car1: x1 = 500 m, y1 = 100 m, and V1 = 40 m/s. Car1 moves
in a straight line that makes an angle of 60° with the x-axis.
Initial position of car2: x2 = 2000 m, y2 = 200 m. Car2 moves in a straight line
and makes an angle of 45° with the (−x) axis.
The collision coordinates are (xc , yc). See Figure P2.3.
Car paths
1200
(xc, yc)

1000

800

600
y

Car1 path Car2 path

400

200 (2000, 200)

(500, 100)
0
500 1000 1500 2000
x

FIGURE P2.3
Two cars on a collision path.
74 MATLAB® Essentials

We can determine the coordinates of the collision point by writing the

equation for the tangent of each line, solving each equation for yc , equating
the two yc expressions, then solving for xc , as shown below.
yc − y1 yc − y2
= tan(60°) , = tan(45°) (P2.5)
xc − x1 x2 − xc

Solving each equation for yc gives

yc = y1 + ( xc − x1 ) × tan(60°) (P2.5)

yc = y2 + ( x2 − xc ) × tan(45°) (P2.5)

Equating the two yc values gives

y1 + ( xc − x1 ) × tan(60°) = y2 + ( x2 − xc ) × tan(45°) (P2.5)

Solving for xc gives

y2 − y1 + x1 tan(60°) + x2 tan(45°)
xc = (P2.5)
tan(60°) + tan(45°)

The distance travelled by each car from the initial state to the collision state is

d1 = ( xc − x1 )2 + ( yc − y1 )2 = V1 tc (P2.5)

d2 = ( x2 − xc )2 + ( yc − y2 )2 = V2 tc (P2.5)

where tc = time of collision. Equating the tc from both the above equations give

d1 d2 d
= → V2 = V1 2 (P2.5)
V1 V2 d1

On line 1:
x(t) = x1 + V1 cos(60°) t , y(t) = y1 + V1 sin(60°) t (P2.5)

On line 2:
x(t) = x2 − V2 cos( 45°)t , y(t) = y2 + V2 sin( 45°) t (P2.5)

Create a MATLAB program that will do the following.

1. Create a plot of the intersecting lines of lengths d1 and d2.

NO T E : You only need to specify two points on the line to plot the line.

2. Determine V2 that will cause the collision to take place.

MATLAB® Fundamentals 75

3. Take t = 0: tc /5: tc and plot the two lines and the two car’s positions
at ti , shown as small circles, all on the same graph.

P2.6. A formula describing the fluid level, h(t), in a tank as the fluid dis-
charges through a small circular orifice (see Figure P2.4) is

Cd A o
h = ho − 2g t (P2.6)
2 AT

where:
Cd is the discharge coefficient
ho is the fluid level in the tank at time, t = 0
Ao is the circular area of the orifice having diameter d
AT is the circular cross-sectional area of the tank having diameter D

Create a MATLAB program that will

1. Determine vectors h versus t, for 0 ≤ t ≤ 80 s.

2. Create a table containing 20 values of t and h (every fourth time step)
and print the table to a file and print the file.
3. Create a plot of h versus t and print the plot.

Use the following parameters:

h o = 0.3 m, the tank diameter, D = 0.8 m, the orifice diameter, d = 0.05 m,
g = 9.81 m/s 2 and Cd = 0.7.

P2.7. When a fluid flows through a pipe there is a pressure drop that is pro-
portional to the pipes length (see Figure P2.5). For a pipe having a circular
cross section, the pressure drop, p1 − p2 [1] is given by

Vo

FIGURE P2.4
Fluid discharging through a small orifice.
76 MATLAB® Essentials

1 2

FIGURE P2.5
Fluid flow through a pipe.

ρV 2 L
p1 − p2 = f (P2.7)
2 D
where:
ρ is the fluid density (kg/m 3 )
V is the average fluid velocity in the pipe (m/s)
D is the pipe diameter (m)
L is the pipe length between points 1 and 2 (m)
f is the friction factor

The friction factor has been determined by experiment. For smooth pipes a
formula that approximates the experimental data is [5]

f = (1.82 log 10 Re − 1.64)−2 (P2.7)

where:
ρVD
Re = (Reynolds number) (P2.7)
µ

and µ = Absolute fluid viscosity (N − s m 2 ).

Develop a MATLAB program that will calculate
1. f versus Re.
2. V versus Re.
3. p1 − p2 versus Re.
4. Plot log(Re) on the x-axis and f on the y-axis (semilog plot). Take

Re = [5.0e3 7.5e3 1.0e4 2.5e4 5.0e4 7.5e4 1.0e5 2.5e5 5.0e5 7.5e5 1.0e6
2.5e6 5.0e6 ].
Take ρ = 1000 kg/m 3 , L = 50 m , D = 0.15 m , and µ = 1.52 × 10 −3 N − s m 2

P2.8. The positioning of a piston in an internal combustion engine is shown

in Figure 2.18a and b. The piston’s position, s, as seen from the crank shaft
center can be determined by the Law of cosines, that is,

b 2 = s2 + r 2 − 2s r cos θ (P2.8a)
MATLAB® Fundamentals 77

or

s2 − (2 r cos θ ) s + ( r 2 − b 2 ) = 0 (P2.8b)

where:
b is the length of the piston rod
r is the radius of the crankshaft

Equation P2.8b is a quadratic equation in s and therefore

s=
1
2 ( )
2r cos θ + 4r 2 cos 2 θ − 4(r 2 − b 2 ) = r cos θ + r 2 (cos 2 θ − 1) + b 2

or

s = r cos θ + b 2 − r 2 sin 2 θ (P2.8c)

The piston is constrained to move in the vertical direction and its position, s,
varies as the crankshaft rotates. The angle, θ, varies with time, t, and can be
expressed in terms of the rotational speed, ω, of the crankshaft. The angle ν
is thus given by
θ = 2πωt (P2.8d)

where ω is in revolutions per second. Substituting Equation P2.8d into

Equation P2.8c gives

s(t) = r cos( 2πωt) + b 2 − r 2 sin 2 (2πωt) (P2.8e)

The piston velocity, V, can be obtained by taking the derivative of Equation

P2.8e with respect to time giving

2πωr 2 sin( 2πωt)cos(2πωt)

V(t) = −2πωr sin( 2πωt) − (P2.8f)
b 2 − r 2 sin 2 (2πωt)

1. In MATLAB, create a matrix consisting of s versus t and V versus t,

for 0 ≤ t ≤ 0.02 seconds. Use 50 subdivisions on the t domain. Take
r = 9 cm, ω = 100 revolutions per second, and b = 14 cm. Plot s versus
t and V versus t as two separate plots.
2. Using MATLAB’s max function and the matrix obtained in part (1),
determine the approximate maximum velocity and print out those
values to the screen.
3. Plot on a single page s versus t for ω = [ 50 100 150 200 ] revolutions
per second.
78 MATLAB® Essentials

P2.9. This project involves plotting the oscillatory motion of a mass in a

mass-spring-dashpot system (see Figure 2.19). The governing equation
for the position, y, of the mass measured from the equilibrium position
depends on the values of the spring constant, k, the damping factor, c, and
the mass, m.
If, k m > (c 2 m)2, then the mass motion will be damped oscillations and
the governing equation describing the motion is

  k 2   k 2 
 c   c    c  
y = exp  − t   A cos  −  t + B sin  −   t  (P2.9)
 2m    m  2m    m  2m   
    

The coefficients A and B are determined by initial conditions, which is

beyond the scope of this textbook. Given the following parameters:

N N−s c A
m = 25 kg , k = 200 , c=5 , A = 0.5 m , B = ×
m m 2m 2
k  c 
− 
m  2m 

For a complete derivation of Equation P2.9 see Project P2.5 in [3,4].

Determine y(t) for 0 ≤ t ≤ 20 seconds in steps of 0.1 seconds.
The envelope of the solution graph for this case is given by

 c 
yenv = ± A exp  − t
 2m 

Plot y versus t and yenv versus t on the same graph.

P2.10. In this project we consider the voltage, v, of a parallel RLC circuit when
at t = 0, the switch is opened. See Figure 2.22. The governing equation for v,
depends on the values of R, L, and C.
If (1 2RC )2 < 1 LC , then the solutions are decaying sinusoids over time
(underdamped) and the governing equation for v is

  1 2   1 2 
 1   1    1  
v = exp  − t   A cos  −  t + B sin  −   t  (P2.10)
 2RC    LC  2RC    LC  2RC   
    

For a complete derivation of Equation P2.10 see Project P2.7 in [3,5]. The
coefficients A and B are to be determined by initial conditions, which are
beyond the scope of this book.
MATLAB® Fundamentals 79

Create a MATLAB program that will calculate and plot v(t) for
0 ≤ t ≤ 500 µsec in steps of 5 μsec using the following parameters:

R = 100 Ω, L = 1 mH, C = 1 μF, A = 6.0000 V, and B = −0.9608 V

P2.11. This project involves determining the rate that heat, q, which is trans-
ferred into a house per unit surface area from a section of the exterior walls
shown in Figure P2.6. The wall consists of plaster board, insulation, wood
sheathing, and brick.
The governing equation describing the rate that heat is transferred, q, into
a house from the wall section [6] shown in Figure P2.6 is

To − Ti
q= (P2.11)
1 L1 L2 L3 L4 1
+ + + + +
ho k1 k2 k3 k4 hi

where:
ho is the outside convective heat transfer coefficient
hi is the inside convective heat transfer coefficient
k1 is the thermal conductivity of brick
k2 is the thermal conductivity of wood sheathing
k3 is the thermal conductivity of insulation
k4 is the thermal conductivity of plaster board

Temperature profile
To

Inside
Outside

Outside brick
Ti
L1 L2 L3 L4

Wood sheathing Insulation

Plaster board

FIGURE P2.6
Heat flow through exterior wall.
80 MATLAB® Essentials

Create a MATLAB program that determines the rate that heat flows into a
house per unit surface area due to the section of the wall described above.
Take To to vary from 25°C to 40°C in steps of 1.0°C and Ti = 20°C. Create
a table containing To and q, include table headings with units. Also, plot q
versus To .
The thickness, L, of each material follows:

=L4 1=
.3 cm, L=
3 10 cm, L2 1=
.3 cm and L1 0.7 cm

The thermal conductivity, k, of each of the materials follows:

W W W
k4 = 0.48 , k3 = 0.05 , k2 = 0.11
m−C m−C m−C

and

W W W
k1 = 0.69 , ho = 56.8 2 and hi = 11.4 2
m−C m −C m −C

For the interior temperature to remain constant, an air-conditioning system

must remove heat at the rate that heat enters the house from the outside as
well as any heat that is generated in the interior.

References
1. Bober, W., Kenyon, R.A., Fluid Mechanics, John Wiley & Sons, New York, 1980.
2. Bober, W., The use of the Swamee-Jain formula in pipe network problems,
Journal of Pipelines, 4, 315–317, 1984.
3. Bober, W., Introduction to Numerical and Analytical Methods with MATLAB for
Engineers and Scientist, CRC Press, Boca Raton, FL, 2014.
4. Thomson, W. T., Theory of Vibration with Applications, Prentice Hall, Englewood
Cliffs, NJ, 1972.
5. Bober, W., Stevens, A., Numerical and Analytical Methods with MATLAB for
Electrical Engineers, CRC Press, Boca Raton, FL, 2012.
6. Holman, J.P., Heat Transfer, 9th Ed., McGraw-Hill, New York, 2002.
3
Conditional Operators, Built-in
Functions with Vector Arguments,
MATLAB®’s Interp1 Function,
and Some Scalar and Vector Operations

3.1 Introduction
In this chapter, we cover the next building block in basic programming,
one that exists in most programming languages and that is the Conditional
Operators. The first conditional operator discussed is the if-else com-
mand. Next we cover the if-elseif-else ladder, followed by a description
of the Switch Group. Also, in this chapter, we discuss working with built-
in functions with vector arguments, MATLAB®’s interp1 function and
some scalar and vector operations, including element-by-element operations.

3.2 Conditional Operators and Alternate Paths

3.2.1 The if Command Provides Two Alternate Paths
Syntax:

if logical expression
statement;

statement;
else
statement;

statement;
end

81
82 MATLAB® Essentials

If the logical expression is true, then only the upper set of statements is
executed. If the logical expression is false, then only the bottom set of state-
ments is executed.
Logical expressions are of the form

a == b; a <= b;
a < b; a >= b;
a > b; a ~= b; (a not equal to b)

Logical expressions have only two values: true or false.

Compound logical expressions

a > b && a ~= c (a > b and a ≠ c)

a > b || a < c (a > b or a < c )

The following example illustrates the use of both the if command and the
input command.
In Example 3.1, we use the fprintf command just before the input
NO T E 1 :
command, to provide the user with more directions than can be given by the
input command itself.

NO T E 2 : When the input command is executed, the Run icon in the Editor
Window changes to a Pause icon (see Figure 3.1).

Example 3.1
% Example_3_1.m
% This program uses the input command and an if statement to
% determine if the output is to go to the screen or to a file. The
% variables y1 and y2 are made vectors so that these variables can
% be printed outside the for loop that created them. As vectors,
% they can also be plotted.
clear; clc;
t=0:0.5:5;

FIGURE 3.1
MATLAB’s menu push button figure.
Conditional Operators, Built-in Functions with Vector Arguments 83

for j=1:length(t)
y1(j)=t(j)^2/10;
y2(j)=t(j)^3/100;
end
fprintf('Do you wish to print the output to \n');
fprintf('the screen or to a file? \n');
response=input('Enter s for screen or F for file \n','s');
% Note, since we entered 's' in the input statement, you do not
% need to enclose your answer in single quotation marks.
if response=='s'
% Table headings:
fprintf(' t y1 y2 \n');
fprintf('---------------------------\n');
for j=1:length(t)
fprintf('%3.1f %10.3f %10.3f \n', ...
t(j),y1(j),y2(j));
end
end
if response=='F'
fo=fopen('output.txt','w');
% Table headings:
fprintf(fo,' t y1 y2 \n');
fprintf(fo,'---------------------------\n');
for j=1:length(t)
fprintf(fo,'%3.1f %10.3f %10.3f \n',...
t(j),y1(j),y2(j));
end
end
--------------------------------------------------------------------------------------------------
Program Results (either from the screen or from the file “output.txt”):

t y1 y2
---------------------------
0.0 0.000 0.000
0.5 0.025 0.001
1.0 0.100 0.010
1.5 0.225 0.034
2.0 0.400 0.080
2.5 0.625 0.156
3.0 0.900 0.270
3.5 1.225 0.429
4.0 1.600 0.640
4.5 2.025 0.911
5.0 2.500 1.250
-----------------------------------------------------------

3.2.2 The if-elseif-else Command Provides More than Two

Alternate Paths
Syntax:
if logical expression 1
statement(s);
elseif logical expression 2
statement(s);
84 MATLAB® Essentials

elseif logical expression 3

statement(s);
else
statement(s);
end

The if-elseif-else ladder works from top down. If the top logical expres-
sion is true, the statements related to that logical expression are executed,
and the program will leave the ladder. If the top logical expression is not
true, the program moves to the next logical expression. If that logical expres-
sion is true, the program will execute the group of statements associated
with that logical expression and leave the ladder. If that logical expression is
not true, the program moves to the next logical expression and continues the
process. If none of the logical expressions are true the program will execute
the statements associated with the else statement. The else statement is
not required. In that case, if none of the logical expressions are true, no state-
ments within the ladder will be executed.
--------------------------------------------------------------

Suppose in Example 3.1, we had more than two choices. For example, we had the
choice of printing to the screen, printing to a file, exiting the program, or opening
the edit window to create a new program. The following program uses the if-
elseif-else ladder and the input statement to decide which choice to select.

Example 3.2
% Example_3_2.m
% First the script determines y1 and y2 as functions of t. The
% script then uses the input command and the if-elseif ladder to
% determine if the program prints the results to the screen,
% prints the results to a file, exits the program or opens the edit
% window to create a new script.
clear; clc;
t=0:0.5:5;
for j=1:length(t)
y1(j)=t(j)^2/10;
y2(j)=t(j)^3/100;
end
fprintf('If you wish to print the output to the screen enter S \n');
fprintf('If you wish to print the output to a file enter F \n');
fprintf('if you wish to close the program enter C \n');
fprintf('If you wish to open the edit window enter E \n');
fprintf('Enter your response without single quotation marks \n');
response=input('Enter S, F, C or E \n','s');
if response=='S'
% Table headings:
fprintf(' t y1 y2 \n');
fprintf(' ------------------------------\n');
Conditional Operators, Built-in Functions with Vector Arguments 85

for j=1:length(t)
fprintf('%5.1f %10.3f %10.3f \n',t(j),y1(j),y2(j));
end
elseif response=='F'
fo=fopen('output.txt','w');
% Table headings:
fprintf(fo,' t y1 y2 \n');
fprintf(fo,'-------------------------------\n');
for j=1:length(t)
fprintf(fo,'%5.1f %10.3f %10.3f \n',t(j),y1(j),y2(j));
end
fclose(fo);
elseif response=='C'
quit;
elseif response=='E'
edit;
end
------------------------------------------------------------------------

3.2.3 The break Command

The break command may be used with an if statement to end a loop;
Example:

for m = 1:20
statement(s);
if m > 10
break;
end
end

In the above example, when m becomes greater than 10, the program
leaves  the for loop and moves on to the next statement outside the for
loop.
----------------------------------------------------------------

Frequently when we deal with problems involving material properties, we

will find that there exist tables that describe several material properties of
several different substances. For example, in Thermodynamics, there are
textbooks that contain tables of the thermodynamic properties of specific
volume, internal energy, and entropy of saturated water, ammonia, refriger-
ant 22, and so on as a function of temperature. Suppose we have a problem
involving a material property that lies between table values. The simplest
way to determine the material property would be to interpolate between
table values. If we assume that the properties vary linearly between table
values, then we can use linear interpolation. Suppose we have a table of y
as a function of x and we wish to determine the value of y at x, where x lies
86 MATLAB® Essentials

between table values x1 and x2 , then the general linear interpolation formula,
based on similar triangles, in terms of y and x is

( y2 − y1 )× ( x − x1 )
y = y1 + (3.1)
x2 − x1

where y1 and y2 are the values of y at x1 and x2 , respectively.

Let us consider the Atmospheric problem described in Example 2.7. In that
example, the atmospheric properties of temperature, pressure, and density
as a function of altitude were specified in the file named atm_properties.txt,
which is shown below. Here, the first column is altitude (m), the second col-
umn is temperature (K), the third column is pressure (Pa), and the fourth
column is density (kg/m3).

0 288.15 1.0133e+005 1.2252

1000 281.65 8.9869e+004 1.1118
2000 275.15 7.9485e+004 1.0065
3000 268.65 7.0095e+004 0.9091
4000 262.15 6.1624e+004 0.8191
5000 255.65 5.4002e+004 0.7360

The next example is an interactive program, where the user is asked to enter
an altitude at which he/she wishes to know the atmospheric properties. The
program uses the if-elseif ladder to determine the closest surrounding
altitude to the entered altitude. It then calculates the atmospheric properties
of temperature, pressure, and density by linear interpolation. In MATLAB,
the easiest and most efficient way to solve this interpolation problem is to use
MATLAB’s interp1 function. However, if you do not have MATLAB avail-
able, but have availability a different computer programming platform used
by engineers, then you would probably need to solve the interpolation prob-
lem by one of the two examples listed below. The second one, Example 3.4 is
more efficient than Example 3.3 and should always be used in preference to
Example 3.3. We give Example 3.3 as a demonstration of the use of the if-
elseif ladder. Later we will use the MATLAB’s interp1 function to solve
the problem.

Example 3.3
% Example_3_3.m
% This program loads data from a file named atm_properties.txt
% The program asks the user to enter an elevation at which atmospheric
% properties are to be determined by linear interpolation.
% The altitude range is from 0 to 5000 m.
Conditional Operators, Built-in Functions with Vector Arguments 87

% Then the atmospheric properties are printed to the screen.

% The program uses the if-elseif ladder to select the closest interval
% to the entered altitude. The properties in this interval will be used
% in the interpolation formula.
% Temperature is in degrees Kelvin (K), pressure is in Pascal (Pa) and
% density is in kg/m^3.
clear; clc;
load atm_properties.txt
% establishing variable names to loaded data.
zt=atm_properties(:,1);
Tt=atm_properties(:,2);
pt=atm_properties(:,3);
rhot=atm_properties(:,4);
fprintf('Enter the altitude at which atmospheric properties \n');
z=input('are to be determined. Altitude range is from 0 to 5000 m \n');
if z>=zt(1)&& z<zt(2)
z1=zt(1); z2=zt(2); T1=Tt(1); T2=Tt(2); p1=pt(1); p2=pt(2);
rho1=rhot(1); rho2=rhot(2);
elseif z>=zt(2)&& z<zt(3)
z1=zt(2); z2=zt(3); T1=Tt(2); T2=Tt(3); p1=pt(2); p2=pt(3);
rho1=rhot(2); rho2=rhot(3);
elseif z>=zt(3)&& z<zt(4)
z1=zt(3); z2=zt(4); T1=Tt(3); T2=Tt(4); p1=pt(3); p2=pt(4);
rho1=rhot(3); rho2=rhot(4);
elseif z>=zt(4)&& z<zt(5)
z1=zt(4); z2=zt(5); T1=Tt(4); T2=Tt(5); p1=pt(4); p2=pt(5);
rho1=rhot(4); rho2=rhot(5);
elseif z>=zt(5)&& z<zt(6)
z1=zt(5); z2=zt(6); T1=Tt(5); T2=Tt(6); p1=pt(5); p2=pt(6);
rho1=rhot(5); rho2=rhot(6);
end
T=T1+(T2-T1)*(z-z1)/(z2-z1);
p=p1+(p2-p1)*(z-z1)/(z2-z1);
rho=rho1+(rho2-rho1)*(z-z1)/(z2-z1);
fprintf('T=%6.2f(K), p=%10.4e(Pa) rho=%6.4f(kg/m^3) \n',T,p,rho);
---------------------------------------------------------------------------

Program Results:
Enter the altitude at which atmospheric properties
are to be determined. Altitude range is from 0 to 5000 m
4380
T=259.68(K), p=5.8728e+04(Pa) rho=0.7875(kg/m^3)
>>
------------------------------------------------------------------------------

Whenever one gets the results of a program, it is prudent to examine the

results to see if they make sense. In this case, do the obtained properties lie
within the proper interval?
An alternative to loading the data in the file atm_properties.txt into the above
script is to enter the data directly into the program as vectors. To accomplish
this, replace the following lines in Example 3.3
88 MATLAB® Essentials

load atm_properties.txt
% establishing variable names to loaded data.
zt=atm_properties(:,1);
Tt=atm_properties(:,2);
pt=atm_properties(:,3);
rhot=atm_properties(:,4);

with
zt=[0 1000 2000 3000 4000 5000];
Tt=[288.15 281.65 275.15 268.65 262.15 255.65];
pt=[10.133 8.9869 7.9485 7.0095 6.1624 5.4002]*1.0e+004;
rhot=[1.2252 1.1118 1.0065 0.9091 0.8191 0.7360];

A more efficient way to solve the problem with far fewer lines of code
is to use a single for loop and an if statement to determine the closest
interval to the entered altitude by the user, thus, reducing the number
of lines in the program. This is demonstrated in the following example.
This becomes important when the number of conditions in the program
is large.

Example 3.4
% Example_3_4.m
% This program enters the data shown in atm_properties.txt directly
% into the program as vectors.
% The program then asks the user to enter an elevation at which the
% atmospheric properties are to be determined by linear interpolation.
% The atmospheric properties are then printed to the screen.
% The program uses a for loop and a compound if statement to determine
% the closest interval to the entered altitude. The properties in
% this interval will be used in the interpolation formula.
% Temperature is in degrees Kelvin (K), pressure is in Pascal (Pa) and
% density is in kg/m^3.
clear; clc;
zt=[0 1000 2000 3000 4000 5000];
Tt=[288.15 281.65 275.15 268.65 262.15 255.65];
pt=[10.133 8.9869 7.9485 7.0095 6.1624 5.4002]*1.0e+004;
rhot=[1.2252 1.1118 1.0065 0.9091 0.8191 0.7360];
fprintf('Enter the altitude at which atmospheric properties \n');
z=input('are to be determined. Altitude range is from 0 to 5000 m \n');
for i=1:length(zt)-1
if z>=zt(i)&& z<zt(i+1)
z1=zt(i); z2=zt(i+1); T1=Tt(i); T2=Tt(i+1);
p1=pt(i); p2=pt(i+1); rho1=rhot(i); rho2=rhot(i+1);
break;
end
end
T=T1+(T2-T1)*(z-z1)/(z2-z1);
p=p1+(p2-p1)*(z-z1)/(z2-z1);
rho=rho1+(rho2-rho1)*(z-z1)/(z2-z1);
fprintf('T=%6.2f(K), p=%10.4e(Pa) rho=%6.4f(kg/m^3) \n',T,p,rho);
-------------------------------------------------------------------------------
Conditional Operators, Built-in Functions with Vector Arguments 89

Program Results:
Enter the altitude at which atmospheric properties
are to be determined. Altitude range is from 0 to 5000 m
1350
T=279.38(K), p=8.6235e+04(Pa) rho=1.0749(kg/m^3)
>>
--------------------------------------------------------------------------------------

3.2.4 The switch Command

In some cases, the Switch Group may be used as an alternative to the if-
elseif-else ladder.
Syntax:

switch(var)
case var1
statement(s);
case var2
statement(s);
case var3
statement(s);
otherwise
statement(s);
end

where var takes on the possible values var1, var2, var3, and so on.
If var equals var1, those statements associated with var1 are executed
and the program leaves the Switch Group. If var does not equal var1, the
program tests if var equals var2, and if yes, the program executes those
statements associated with var2 and leaves the Switch Group. If var does
not equal any of var1, var2, and so on, the program executes the state-
ments associated with the otherwise statement. If var1, var2, and so on
are strings, they need to be enclosed by single quotation marks. It should
be noted that var cannot be a logical expression, such as var1 > = 80.
The following example illustrates the use of the Switch Group in a
MATLAB program.

Example 3.5
% Example_3_5.m
% This program is a test of the switch statement.
clear; clc;
var = 'a';
x = 5;
switch(var)
case 'b'
z = x^2;
case 'a'
z = x^3;
90 MATLAB® Essentials

otherwise
z=0;
end
fprintf(' z = %6.1f \n',z);
--------------------------------------------------------------------------

Program Results:
z = 125.0
>>
-----------------------------------------------------------------------

3.2.5 MATLAB’s menu Function

MATLAB’s menu function allows the user to select from several choices by
the use of push buttons on a graphical display. Each item listed in the menu
is given a number according to its position in the menu list. The top item
in the menu display is given number 1, the second from the top is given
number 2, and so on. In the following example, the user is prompted to click
on one of the gases listed in the menu display (see Figure 3.2). The program
then determines the density of the gas selected based on the ideal gas law.
Pressure and temperature are specified.

Example 3.6
% Example_3_6.m
% This program uses MATLAB's menu function and the if-elseif ladder
% to determine the density of a gas by the ideal gas law.
% The gas is selected by the user by clicking on a push button in
% the menu display.
clear; clc;
p=2*1.013e+5; T=350.0;
Rt=[287 2077 4121 297 260];
k= menu('choose a gas','air','helium','hydrogen','nitrogen','oxygen');
if k==1
R=Rt(1);
fprintf('The gas is Air \n');
elseif k==2
R=Rt(2);
fprintf('The gas is Helium \n');
elseif k==3
R=Rt(3);
fprintf('The gas is hydrogen \n');
elseif k==4
R=Rt(4);
fprintf('The gas is Nitrogen \n');
elseif k==5
R=Rt(5);
fprintf('The gas is Oxygen \n');
end
rho=p/(R*T);
fprintf('The density, rho, is based on the ideal gas law \n');
fprintf('T=%5.1fK p=%10.4ePa rho=%7.4fkg/m^3 \n',T,p,rho);
----------------------------------------------------------------------------
Conditional Operators, Built-in Functions with Vector Arguments 91

FIGURE 3.2
Menu display.

The menu display that pops up (upper left corner of the screen) is shown in Figure 3.2.
I clicked on the nitrogen button and got the following result:

Program Results:
The gas is Nitrogen
The density, rho, is based on the ideal gas law
T=350.0K p=2.0260e+05Pa rho=1.9490kg/m^3
>>
-------------------------------------------------------------------------------

REVIEW 3.1
1. What statement is frequently used to establish two conditional
paths?
2. What series of statements is used to establish several condi-
tional paths?
3. List the various types of logic statements that can be used with
the if-else and if-elseif-else ladder.
4. Is the else statement required with either the if-else or the
if-elseif–else ladder?
5. What statement group and a MATLAB’s function are alterna-
tives to the if-elseif-else ladder?
92 MATLAB® Essentials

TABLE 3.1
Score Range for Letter Grade
Letter Grade Score Range
A 100 to 90
B 89 to 80
C 79 to 70
D 69 to 60
F Less than 60

Exercises

E3.1. The numerical grades on an exam are listed in the vector labeled scores.
Scores = [92 85 73 83 75 85 65 76 96 84 76 81 55 76 94 65 84 62 78 80 74 62 84
76 70 88 74 82 70 86]. Each score will be assigned a letter grade as indicated
in the Table 3.1.
Write a MATLAB program that will determine the number of grades in
each letter grade category and plot the result as a bar chart (see Example 2.14).

E3.2. Repeat Example 3.6, but this time use the switch Statement instead of
the if-elseif ladder.

3.3 Working with Built-in Functions with Vector Arguments

MATLAB allows the built-in functions such as sin( ), cos( ), exp( ), and
so on as well as functions in general to have vectors as arguments. The result
will also be a vector. This is demonstrated in the next example.

Example 3.7
% Example_3_7.m
% This program demonstrates that if the argument in a built in
% function, such as MATLAB's sine function, is a vector, the result
% will also be a vector.
clear; clc;
% Define vector x;
x=0:30:360;
% Let y1 be the sine of a vector x where x is in degrees.
% Running sind with vector x as an argument will return a vector:
y1 = sind(x);
% Thus, y1 is a vector.
% Let y2(n) be the sine of the nth element of x. We will use a for
% loop to calculate each value y2(n) and then compare y1 and y2.
Conditional Operators, Built-in Functions with Vector Arguments 93

for n=1:length(x)
y2(n)=sind(x(n));
end
% Table headings
fprintf(' x y1 y2 \n');
fprintf('------------------------------------\n');
for n=1:length(x)
fprintf('%5.1f %8.5f %8.5f \n',x(n),y1(n),y2(n))
end
-----------------------------------------------------------------------------

Program Results:
x y1 y2
------------------------------------
0.0 0.00000 0.00000
30.0 0.50000 0.50000
60.0 0.86603 0.86603
90.0 1.00000 1.00000
120.0 0.86603 0.86603
150.0 0.50000 0.50000
180.0 0.00000 0.00000
. . .
. . .
-------------------------------------------------------------------------------

In the generated output, does y1 = y2?

We see that in some scripts, we could replace the use of a for loop by using
a vector argument in many built-in or self-written functions, which produces
a vector result, thus reducing the number of lines in the script. This concept
was demonstrated in the above Example 3.7.

3.4 MATLAB’s interp1 Function

MATLAB has a function named interp1 that performs interpolation.
The syntax for interp1 is

Yi = interp1(X,Y,Xi)

where X and Y are a set of known (x, y) data points and Xi is the set of x
values at which the set of y values, Yi, are to be determined by linear inter-
polation. Arrays X and Y must be of the same length. Note: If Xi is a vector,
then Yi will also be a vector. The function interp1 can also be used for
interpolation methods other than linear interpolation, and this is covered in
Chapter 9 on Curve Fitting. In the next example, we modify Example 3.4 by
using MATLAB’s interp1 function to interpolate for the atmospheric proper-
ties at an altitude entered from the keyboard.
94 MATLAB® Essentials

Example 3.8
% Example_3_8.m
% This program enters the data shown in atm_properties.txt directly
% into the program as vectors.
% The program then asks the user to enter an elevation at which the
% atmospheric properties are to be determined by linear interpolation.
% The atmospheric properties are then printed to the screen.
% The program uses MATLAB's interp1 function to do the interpolation.
% Temperature is in (K), pressure is in (Pa) and
% density is in (kg/m^3).
clear; clc;
zt=[0 1000 2000 3000 4000 5000];
Tt=[288.15 281.65 275.15 268.65 262.15 255.65];
pt=[10.133 8.9869 7.9485 7.0095 6.1624 5.4002]*1.0e+004;
rhot=[1.2252 1.1118 1.0065 0.9091 0.8191 0.7360];
fprintf('Enter the altitude at which atmospheric properties \n');
z=input('are to be determined. Altitude range is from 0 to 5000 m \n');
T=interp1(zt,Tt,z);
p=interp1(zt,pt,z);
rho=interp1(zt,rhot,z);
fprintf('T=%6.2f(K), p=%10.4e(Pa) rho=%6.4f(kg/m^3) \n',T,p,rho);
---------------------------------------------------------------------------
Program Results:
Enter the altitude at which atmospheric properties
are to be determined. Altitude range is from 0 to 5000 m.
1350
T=279.38(K), p=8.6235e+04(Pa) rho=1.0749(kg/m^3)
>>

Are the results the same as those obtained in Example 3.4?

---------------------------------------------------------------------------

The next example demonstrates the use of interp1 function for interpolat-
ing for internal energy of a refrigerant at temperatures specified in vector T2.
The output from interp1 will also be a vector.

Example 3.9
% Example_3_9.m
% This program uses MATLAB's function interp1 to interpolate for
% the internal energy, u, as a function of temperature, T, of
% an unspecified refrigerant.
% Measured values of u in (kJ/kg) vs. T in degrees (C)
% are specified in vectors ut and Tt respectively.
% The temperatures at which the internal energy is to be determined
% are specified in vector T2.
% The program also creates a plot of u vs. T and includes points of
% u at temperature T2.
clear; clc;
Tt=-20:10:90;
ut=[217.86 224.97 232.24 239.69 247.32 255.12 263.10 271.25 ...
279.58 288.08 296.75 305.58];
fprintf('This program interpolates for the internal energy, u at \n');
fprintf('a specified temperature T. \n');
Conditional Operators, Built-in Functions with Vector Arguments 95

fprintf('The allowable temperature range is -20 to +90 C. \n\n');

T2=[-12 6 24 32 64 82];
u=interp1(Tt,ut,T2);
fprintf(' T2(C) u(kJ/kg) \n');
fprintf('--------------------------\n');
for i=1:length(T2)
fprintf(' %6.1f %8.3f \n',T2(i),u(i));
end
plot(Tt,ut,T2,u,'o' );
xlabel('T(C)'), ylabel('u(kJ/kg)'), title('u vs. T'), grid;
-----------------------------------------------------------------------------------------

Program Results:
This program interpolates for the internal energy, u at
a specified temperature T.
The allowable temperature range is -20 to +90 C.
T2(C) u(kJ/kg)
--------------------------
-12.0 223.548
6.0 236.710
24.0 250.440
32.0 256.716
64.0 282.980
82.0 298.516
>>

See Figure 3.3

--------------------------------------------------------------------------

u vs. T
310
ut vs. Tt
300 u vs. T2

290

280

270
u(kJ/kg)

260

250

240

230

220

210
−20 0 20 40 60 80 100
T(C)

FIGURE 3.3
Internal energy, u, as a function of temperature, T, of an unspecified refrigerant.
96 MATLAB® Essentials

3.5 Some Scalar and Vector Operations

3.5.1 Addition of a Scalar and a Vector
The result of a scalar added to a vector is that the scalar is added to each ele-
ment of the vector. Example:

Z = 5 + [2 4 6 8] = [7 9 11 13]

3.5.2 Multiplication of a Scalar Times a Vector

The result of a scalar multiplied by a vector is that each element of the vector
is multiplied by the scalar. Example:

Z = 5 ∗ [2 4 6 8] = [10 20 30 40]

3.5.3 Addition and Subtraction of Two Vectors of the Same Length

Given A = [ a1 a2 a3 ] and B = [ b1 b2 b3 ], then

C = A + B gives C = [( a1 + b1 ) ( a2 + b2 ) ( a3 + b3 )]
D = A − B gives D = [( a1 − b1 ) ( a2 − b2 ) ( a3 − b3 )]

The addition or subtraction of two vectors of different lengths is not

defined.

3.5.4 Element-by-Element Operations

Given two vectors of the same length, we can perform element-by-element
multiplication, division, and exponentiation in MATLAB with the .*, and
./, and . ^ operators.
Given A = [ a1 a2 a3 ] and B = [ b1 b2 b3 ], then
C = A . ∗ B = [ a1b1 a2b2 a3b3 ] ,

a a2 a3  ,
D = A ./ B =  1
 b1 b2 b3 

= =
E A . ^ B [ a1b1 a2 b2 a3 b3 ] = power(A,B)

We see that the element-by-element operation results in a vector that is the

same length as the vectors that are involved in the operation.
Conditional Operators, Built-in Functions with Vector Arguments 97

Example 3.10
To demonstrate the above relations, copy the following script and run the program
and observe the result.

% Example_3_10.m
% This program demonstrates some scalar and vector operations.
clear; clc;
s=5; a=[1 5 9]; b=[2 6 12]; c=s+a; d=s*b; e=a+b; f=a-b; g=a.*b;
h=a./b;
fprintf('s=%3i \n',s);
fprintf('a= %3i %3i %3i \n',a);
fprintf('\n');
fprintf('c=s+a\n');
fprintf('c= %3i %3i %3i \n',c);
fprintf('\n');
fprintf('s=%4i \n',s);
fprintf('b= %3i %3i %3i \n',b);
fprintf('d=s*b \n');
fprintf('d= %3i %3i %3i \n',d);
fprintf('\n');
fprintf('a= %3i %3i %3i \n',a);
fprintf('b= %3i %3i %3i \n',b);
fprintf('e=a+b \n');
fprintf('e= %3i %3i %3i \n',e);
fprintf('\n');
fprintf('a= %3i %3i %3i \n',a);
fprintf('b= %3i %3i %3i \n',b);
fprintf('f=a-b \n');
fprintf('f= %3i %3i %3i \n',f);
fprintf('\n');
fprintf('a= %3i %3i %3i \n',a);
fprintf('b= %3i %3i %3i \n',b);
fprintf('g=a.*b \n');
fprintf('g= %3i %3i %3i \n',g);
fprintf('\n');
fprintf('a= %3i %3i %3i \n',a);
fprintf('b= %4i %4i %4i \n',b);
fprintf('h=a./b \n');
fprintf('h= %8.4f %8.4f %8.4f \n',h);
fprintf('\n');
a=[2 3 4]; b=[2 2 2];
k=a.^b;
fprintf('a= %3i %3i %3i \n',a);
fprintf('b= %4i %4i %4i \n',b);
fprintf('k=a.^b \n');
fprintf('k= %4.1f %4.1f %4.1f \n',k(1),k(2),k(3));
--------------------------------------------------------------------------

Program Results:
s= 5
a= 1 5 9
c=s+a
c= 6 10 14
98 MATLAB® Essentials

s= 5
b= 2 6 12
d=s*b
d= 10 30 60

a= 1 5 9
b= 2 6 12
e=a+b
e= 3 11 21

a= 1 5 9
b= 2 6 12
f=a-b
f= -1 -1 -3

a= 1 5 9
b= 2 6 12
g=a.*b
g= 2 30 108

a= 1 5 9
b= 2 6 12
h=a./b
h= 0.5000 0.8333 0.7500

A= 2 3 4
B= 2 2 2
k=A.^B
k= 4.0 9.0 16.0
>>
--------------------------------------------------------------------------------

3.5.5 Operation of Two Vector Functions

If a script involves a mathematical operation of two vector functions (such
as a product of two vector functions), then the operation will require an
element-by-element operation. In Example 3.9, we compute the product of
two vector functions, both directly and indirectly by using a for loop and
multiplying the elements of each vector. We then compare the results.

Example 3.11
% Example_3_11.m
% This example illustrates element-by element operation
% of two vector functions
clear; clc;
x = 0:30:180;
% y1 is the product of two vector functions
y1 = sind(x).* cosd(x);
fprintf(' x y1 y2 \n');
fprintf('---------------------------------------\n');
for n=1:length(x)
% y2(n) is the product of the elements of the two functions.
Conditional Operators, Built-in Functions with Vector Arguments 99

y2(n) = sind(x(n)) * cosd(x(n));

fprintf('%5.1f %8.5f %8.5f \n',x(n),y1(n),y2(n));
end
--------------------------------------------------------------------------------

Program Results:
x y1 y2
-----------------------------------
0.0 0.00000 0.00000
30.0 0.43301 0.43301
60.0 0.43301 0.43301
90.0 0.00000 0.00000
120.0 -0.43301 -0.43301
150.0 -0.43301 -0.43301
180.0 -0.00000 -0.00000
>>
--------------------------------------------------------------------------------

We see that the two different methods for computing y1 and y2 give the
same answer.

REVIEW 3.2

1. If y = 3.0 * A and A is a vector, what can you say about y?

2. If y = 3.0 * sin(x) and x is a vector, what can you say about y?
3. If vector C = A + B, what must be true about vectors A and B.
4. What is the result of the multiplication of two vectors of the
same length, say A and B, and how must it be programmed?
5. What is the name of MATLAB’s function that does interpolation?
6. What are the inputs to MATLAB’s interpolation function?
7. What are the outputs from MATLAB’s interpolation function?

Projects
P3.1. Though atmospheric conditions vary from day-to-day, it is convenient
for design purposes, to have a model for atmospheric properties with alti-
tude. The U.S. Standard Atmosphere, modified in 1976, is such a model. The
model consists of two types of regions, one in which the temperature varies
linearly with altitude, and the other is a region where the temperature is a
constant (see Figure P3.1).
100 MATLAB® Essentials

× 104 z versus T
8
Region 7
7

Region 6
6

5
Region 5

4
z

Region 4
3
Region 3
2
Region 2
1
Region 1
0
−80 −70 −60 −50 −40 −30 −20 −10 0 10 20
T

FIGURE P3.1
Temperature versus altitude of U.S. Standard Model Atmosphere.

The temperature and approximate pressure relations are as follows:

1. For a region where the temperature varies linearly

gi
 λ i ( zi +1 − zi )  λ i R
p = pi  1 −  (P3.1)
 Ti 

T = Ti − λ i ( zi +1 − zi ) (P3.2)

where:
zi is the altitude at the beginning of region i, i = 1, 2, …, 7
zi+1 is the altitude at the end of region i and at the beginning of
region i+1
(pi , Ti) is the pressure and temperature at the beginning of region i
λ i is the lapse rate in region i
R is the air gas constant = 286.9 N − m kg − K
gi is the gravitational constant in region i. Although g varies
slightly with altitude we take g to be a constant in the region
evaluated at zi , otherwise the above expression for p would
be a lot more complicated than the one shown above
Conditional Operators, Built-in Functions with Vector Arguments 101

2. For a region where the temperature is constant (λ = 0)

T = Ti (P3.3)

 g ( z − zi ) 
p = pi exp  − i  (P3.4)
 RTi 

For each region, the governing equation for g is

 2z 
g = go  1 − (P3.5)
 Ro 

where:
go = 9.810 m/s2
Ro = 6.3714e+6 m
z is the altitude at the beginning of the region

Create a MATLAB program that will

1. Determine the values of Ti , pi , and gi , i = 1, 2, …, 8

2. Construct a table filling in the unknown values listed in Table P3.1.

TABLE P3.1
U.S. Standard Atmosphere Property Table
Regional Properties of U.S. Standard Atmosphere
Region zi (km) Ti (K) p i (Pa) λ i (K/m) g i (m/s2)
0 288.15 101325 — 9.810
1 — — — 0.0065 —
11.0 — — — —
2 — — — 0.0000
20.0 — — — —
3 — — — −0.001 —
32.0 — — — —
4 — — — −0.0028 —
47.0 — — — —
5 — — — 0.0000 —
51.0 — — — —
6 — — — 0.0028 —
71.0 — — — —
7 — — — 0.0020 —
84.9 — — — —
102 MATLAB® Essentials

P3.2. This project is a modification of Example 3.8. Instead of making the

program interactive, enter the following altitudes, z2, at which the atmo-
spheric properties of T, p, ρ are to be determined by linear interpolation using
MATLAB’s interp1 function.

z2 = [ 680 1250 2360 3685 4320 4865]

Print the results to a file in a table format.

P3.3. The properties of specific volume, v, and pressure, p, as a function of

temperature, T, for carbon dioxide based on the Redlich–Kwong Equation of
state are given in the tables below:
NO T E : 1 bar = 105 N/m2
Write a MATLAB program that will do the following:

1. Construct three separate vectors containing the carbon dioxide vari-

ables of T, v, and p.
2. Print Table P3.2 to the screen (with table headings).
3. Determine v and p at temperatures T2 using MATLAB’s interp1
function. Take

T 2 = [ 367 634 420 587 742 ]

4. Print to the screen in a table format (with table headings) values of v

and p at temperatures T2.

TABLE P3.2
Properties of Carbon Dioxide Based on
the Redlich–Kwong Equation of State
T (K) v (m3/kmol) p (bar)
350 0.28 7.65
400 0.32 8.57
450 0.36 9.16
500 0.40 9.55
550 0.44 9.81
600 0.48 10.00
650 0.52 10.14
700 0.56 10.24
750 0.60 10.31
Conditional Operators, Built-in Functions with Vector Arguments 103

TABLE P3.3
Refrigerant Properties
T (°C) v (m3/kg) u (kJ/kg)
−35 0.34235 244.33
−30 0.35369 217.58
−20 0.36513 222.43
−10 0.37649 227.37
−5 0.35778 232.42
0 0.39901 237.57
10 0.41019 242.82
20 0.42133 248.16
25 0.43243 253.61
30 0.44348 259.16

P3.4. The measured properties of a refrigerant are shown in Table P3.3. Create
a MATLAB program that will repeatedly ask the user if he/she wishes to
determine the refrigerant properties. If the answer is Y for yes, then ask
the user to enter a temperature at which the refrigerant properties are to
be determined. Use MATLAB’s interp1 function to evaluate the refrigerant
properties, and then print the result to the screen.
P3.5. This project involves the mass-spring-dashpot system (see Figure 2.19).
If the mass is displaced from its equilibrium position and released, the sub-
sequent motion of the mass will depend on the values of m, k, and c. To sim-
plify the writing of the governing equations, we will define

2
k  c 
arg = − 
m  2m 

If arg > 0, then the displacement, y, as a function of time, t, will be

 ct 
y = exp  − (
 × A cos ( arg t ) + B sin ( arg t )
 2m 
)

If arg < 0, then the displacement, y, as a function of time, t, will be

 ct 
y = exp  − (
 × A exp ( − arg t ) + B exp ( − − arg t )
 2m 
)
104 MATLAB® Essentials

where exp( x) = e x
We wish to consider four cases.

a. m = 75 kg, k = 85 N/m, c = 200 N-s/m.

b. m = 75 kg, k = 150 N/m, c = 40 N-s/m.
c. m = 75 kg, k = 50 N/m, c = 150 N-s/m.
d. m = 75 kg, k = 200 N/m, c = 20 N-s/m.

Create a MATLAB program that will plot the motion of y as a function of t for
0 ≤ t ≤ 20 s in steps of 0.1 s. To make the initial displacement equal to 0.5 m
and initial velocity equal to zero, use the following values for A and B.
For the case arg > 0, take

Ac
A = 0.5 m, B =
2 m arg

For the case arg < 0, take

 c  c
0.5  − arg +  − arg −
 2 m ,B= A 2m
A=
2 − arg c
− arg +
2m
4
Self-Written Functions and MATLAB®’s
fminbnd Function

4.1 Introduction
In this chapter, we cover the self-written function, which is the last of the
building blocks that we mentioned earlier. Self-written functions are useful
if you have a complicated program and wish to break it down into smaller
segments. Also, if a series of statements is to be used many times, it is con-
venient to place them in a function. Many MATLAB® functions (such as
fminbnd, fzero, quad, and ode45) require a self-written function to define
the problem of interest. Self-written functions are equivalent to subroutines
in most programming languages, but in MATLAB they are usually stored in
separate files instead of the main program (though small functions can be
defined in the same file as your main script, as described in the section on
anonymous functions). The function file name must be saved as a .m file.
MATLAB has a function that determines the relative minimum or relative
maximum of a single variable function. The function is fminbnd and is
covered in this chapter. This MATLAB function requires the user to write a
self-written function, either as a separate .m file or as an anonymous func-
tion within the main script.

4.2 Self-Written Function

MATLAB has a template for writing a function (see Figure 4.1), which can be
accessed by clicking on the down arrow under new in the toolstrip in the Editor
Window and selecting function from the dropdown menu. To create your
function, you would need to edit the function template and make the desired
changes. However, it is just as easy to open a new script window and type in
your desired function and save it with the function name and as a .m file. The
function template is of the form
function [output arguments] = function_name (input arguments)

105
106 MATLAB® Essentials

FIGURE 4.1
MATLAB’s function template.

TABLE 4.1
Example Function Definitions
Function Definition Line Function File Name
function [p,T] = atm(z,rho) atm.m
function ex = exf (x) efx.m
function[] = output(x,y) output.m

The first executable statement in the function file must start with the word function.
Some example function definitions are shown in Table 4.1.
If the function has more than one output value, then the output variables
must be in brackets. If there is only one output value, then no brackets are
necessary. If there are no output values, use empty brackets.

1. Input and Output Arguments:

Input and output arguments in the function may be either scalars, vectors,
or matrices. The input arguments defined in the calling program passes
information to the function, where it is most often used in one or more
arithmetic statements. The input arguments in the calling program need to
be defined before the function is called. The output arguments determined
in the function passes information to the calling program.
Self-Written Functions and MATLAB®’s fminbnd Function 107

2. Variables Defined and Manipulated inside the Function:

Variables defined and manipulated inside the function are local to the
function. This means that the only communication between the calling
program and the function is through the input and output arguments
of the function. The exception to this is when a global statement is
contained in both the calling program and in the function.

In the next example, the input command is again used to ask the user
to enter an altitude from the keyboard. Earlier, it was mentioned that
MATLAB version R2016a has a bug. When the input command is exe-
cuted, the cursor remains in the Editor Window instead of moving to the
Command Window. So the user needs to click in the Command Window
before entering an altitude. This bug will be fixed in the next version of
MATLAB.

Example 4.1, Part A

% Example_4_1_parta.m
% The program asks the user to enter an elevation at which the
% atmospheric properties are to be determined by linear interpolation.
% The function atm_fun contains atmospheric properties every
% thousand meters and does the interpolation by MATLAB's
% interp1 function. The properties at the entered elevation is
% passed on to the calling program where it is printed to the
% Command window.
% In this example input and out variables are scalars.
% Temperature is in degrees Kelvin (K), pressure is in Pascal (Pa) and
% density is in kg/m^3.
clear; clc;
fprintf('Enter the altitude at which atmospheric properties are \n');
z=input('to be determined. Altitude range is from 0 to 5000 m \n');
[T,p,rho]=atm_fun(z);
fprintf('z=%6.1f(m) T=%6.2f(K) ',z,T);
fprintf('p=%10.4e(Pa) rho=%6.4f(kg/m^3) \n',p,rho);
--------------------------------------------------------------------

Example 4.1, Part B

% atm_fun.m
% This function works with Example 4.1, part A
function [T,p,rho]=atm_fun(z)
zt=[0 1000 2000 3000 4000 5000];
Tt=[288.15 281.65 275.15 268.65 262.15 255.65];
pt=[10.133 8.9869 7.9485 7.0095 6.1624 5.4002]*1.0e+004;
rhot=[1.2252 1.1118 1.0065 0.9091 0.8191 0.7360];
T=interp1(zt,Tt,z);
p=interp1(zt,pt,z);
rho=interp1(zt,rhot,z);
--------------------------------------------------------------------
108 MATLAB® Essentials

Program Results:
Enter the altitude at which atmospheric properties
are to be determined. Altitude range is from 0 to 5000 m
3400
z=3400.0(m), T=266.05(K), p=6.6707e+04(Pa), rho=0.8731(kg/m^3)
>>
-----------------------------------------------------------------------------------------------

We now want to modify Example 4.1, so that the input and output variables
to the function are vectors. We only need to modify Example 4.1, Part A. We
do not need to modify the function atm_fun.

Example 4.2, Part A

% Example_4_2_parta.m
% This program specifies vector z as an input to function atm_fun.
% The function atm_fun contains atmospheric properties every
% thousand meters and does the interpolation by MATLAB's
% interp1 function at each element of vector z.
% The output variable of T, p and rho from atm_fun will be vectors.
% These vectors are passed on to the calling program where it is
% printed to the screen in table format.
% Temperature is in degrees Kelvin (K), pressure is in Pascal (Pa) and
% density is in kg/m^3.
clear; clc;
z=[1250 2560 3480 4360];
[T,p,rho]=atm_fun(z);
fprintf(' z(m) T(K) p(Pa) rho(kg/m^3) \n');
fprintf('---------------------------------------------------\n');
for i=1:length(z)
fprintf('%6.1f %6.2f %10.4e %6.4f \n',...
z(i),T(i),p(i),rho(i));
end
----------------------------------------------------------------------

Example 4.2, Part B is the same as Example 4.1, Part B

-------------------------------------------------------------------------------
Program Results:
z(m) T(K) p(Pa) rho(kg/m^3)
---------------------------------------------------
1250.0 280.02 8.7273e+04 1.0855
2560.0 271.51 7.4227e+04 0.9520
3480.0 265.53 6.6029e+04 0.8659
4360.0 259.81 5.8880e+04 0.7892
>>
---------------------------------------------------------------------------------

The following example demonstrates that the names of the arguments in the
calling program need not be the same as those in the function. It is only
the order of the argument list in the calling program that needs to be in the
Self-Written Functions and MATLAB®’s fminbnd Function 109

same order as the argument list defined in the function. This feature is use-
ful when a function is to be used with several different scripts, each script
having different variable names, but each of the variables names correspond
to variables in the function. This concept is used in all of MATLAB’s built-in
functions.
Let us modify Example 4.2, Part A, and name it Example 4.3, Part A.

Example 4.3, Part A

% Example_4_3_parta.m
clear; clc;
z=[1250 2560 3480 4360];
[T,p,rho]=atm_fun2(z);
fprintf('\n This output is from Example 4.3, Part A \n');
fprintf(' z(m) T(K) p(Pa) rho(kg/m^3) \n');
fprintf('-------------------------------------------------\n');
for i=1:length(z)
fprintf('%6.1f %6.2f %10.4e %6.4f \n',...
z(i),T(i),p(i),rho(i));
end
-----------------------------------------------------------------------

Now we will modify function atm_fun and name it atm_fun2.

% atm_fun2.m
% This function works with Example 4.3, part A
function [A,B,C]=atm_fun2(x)
alt=[0 1000 2000 3000 4000 5000];
Temp=[288.15 281.65 275.15 268.65 262.15 255.65];
pres=[10.133 8.9869 7.9485 7.0095 6.1624 5.4002]*1.0e+004;
dens=[1.2252 1.1118 1.0065 0.9091 0.8191 0.7360];
A=interp1(alt,Temp,x);
B=interp1(alt,pres,x);
C=interp1(alt,dens,x);
fprintf('This output is from atm_fun2 \n');
fprintf(' x(m) A(K) B(Pa) C(kg/m^3) \n');
fprintf('------------------------------------------------\n');
for i=1:length(x)
fprintf('%6.1f %6.2f %10.4e %6.4f \n',...
x(i),A(i),B(i),C(i));
end
---------------------------------------------------------------------------

Program Results:
This output is from atm_fun2
x(m) A(K) B(Pa) C(kg/m^3)
------------------------------------------------
1250.0 280.02 8.7273e+04 1.0855
2560.0 271.51 7.4227e+04 0.9520
3480.0 265.53 6.6029e+04 0.8659
4360.0 259.81 5.8880e+04 0.7892
110 MATLAB® Essentials

This output is from Example 4.3, Part A

z(m) T(K) p(Pa) rho(kg/m^3)
-------------------------------------------------
1250.0 280.02 8.7273e+04 1.0855
2560.0 271.51 7.4227e+04 0.9520
3480.0 265.53 6.6029e+04 0.8659
4360.0 259.81 5.8880e+04 0.7892
>>
------------------------------------------------------------------------------

Comparing results, we see that the names of the arguments in the calling program
need not be the same as those in the function. In the calling program, the names of the
input and output arguments are z, T, p, and rho. In the function, atm_fun2, the names
of the input and output arguments are x, A, B, and C. Looking at the results, we see
that x = z, A = T, B = p, and C = rho. It is only the order of the argument list in the
function that needs to be in the same order as the argument list in the calling program.
-------------------------------------------------------------------------------

4.3 Anonymous Functions

Sometimes it is more convenient to define a function inside your script rather
than in a separate file. For example, if a function is brief (perhaps a single
line) and unlikely to be used in other scripts, then the anonymous form of a
function can be used. This will save you from having to create another .m file.
The syntax for an anonymous function is

funhandle = @(arg_list) (function expression)

A function handle is a MATLAB value that provides a means of calling a

function indirectly. An example of an anonymous function is

fh = @(x,y)(y*sin(x)+x*cos(y));

In the above expression, MATLAB lists the @ sign as a function handle cre-
ation, fh is the function handle, the (x,y) defines the input arguments to
the function, and (y*sin(x)+x*(cos(y)) is the function. Anonymous
functions may be used in a script or in the Command Window.
Example: In the Command Window, type-in the following two lines:

>> fh = @(x,y)(y*sin(x)+x*cos(y));
>> w = fh(pi,2*pi)
w =
3.1416
-----------------------------------------------------------------------

Additional information on anonymous functions can be obtained by typing

help function_handle in the Command Window.
Self-Written Functions and MATLAB®’s fminbnd Function 111

Example 4.4
The following example uses an anonymous function that employs the interpola-
tion formula described in Equation 3.1. The script is a modification of Example 3.4
that contains atmospheric data of temperature, pressure, and density every
1000 meters. The script, interactively, asks the user to enter an altitude from the
keyboard.

% Example_4_4.m
% This program interpolates for atmospheric properties T, p and rho at
% an altitude entered from the keyboard.
% Atmospheric properties of temperature, pressure and density
% are specified every 1000 meters. The atmospheric properties at an
% altitude entered from the key board are determined by
% the interpolation formula described in Equation 3.1 and printed to
% the screen. An anonymous function, which avoids creating an extra .m
% file, is used to do the interpolation:
clear; clc;
% anonymous function
yf = @(z,z1,z2,y1,y2) (y1+(z-z1)*(y2-y1)/(z2-z1));
zt=[0 1000 2000 3000 4000 5000];
Tt=[288.15 281.65 275.15 268.65 262.15 255.65];
pt=[10.133 8.9869 7.9485 7.0095 6.1624 5.4002]*1.0e+004;
rhot=[1.2252 1.1118 1.0065 0.9091 0.8191 0.7360];
fprintf('Do you wish to have the atmospheric properties \n');
fprintf('at a specific altitude determined \n');
char=input('enter Y for yes or N for no \n','s');
if char=='N'
fprintf('Good Bye \n');
exit;
end
while char=='Y'
fprintf('Enter the altitude at which atmospheric properties \n');
fprintf('are to be determined. \n');
z=input('Altitude range is from 0 to 5000 m \n');
for i=1:length(zt)-1
if z >= zt(i) && z < zt(i+1)
z1=zt(i); z2=zt(i+1);
T1=Tt(i); T2=Tt(i+1);
T=yf(z,z1,z2,T1,T2);
p1=pt(i); p2=pt(i+1);
p=yf(z,z1,z2,p1,p2);
rho1=rhot(i); rho2=rhot(i+1);
rho=yf(z,z1,z2,rho1,rho2);
fprintf('z=%6.1f(m) T=%7.2f(C)',z,T);
fprintf('p=%12.5e(Pa) rho=%8.5f(kg/m^3) \n',p,rho);
end
end
fprintf('\n');
fprintf('Do you wish to enter another altitude ');
char=input('enter Y for yes or N for no \n','s');
if char=='N'
fprintf('Good Bye \n');
end
end
----------------------------------------------------------------------------
112 MATLAB® Essentials

Program Results:
Do you wish to determine atmospheric properties
at a specific altitude,
enter Y for yes or N for no
Y
Enter the altitude at which atmospheric properties
are to be determined.
Altitude range is from 0 to 5000 m
1480
z=1480.0(m) T= 278.53(C) p= 8.48847e+04(Pa) rho= 1.06126(kg/m^3)

Do you wish to enter another altitude enter Y for yes or N for no

Y
Enter the altitude at which atmospheric properties
are to be determined.
Altitude range is from 0 to 5000 m
3620
z=3620.0(m) T= 264.62(C) p= 6.48430e+04(Pa) rho= 0.85330(kg/m^3)

Do you wish to enter another altitude enter Y for yes or N for no

N
Good Bye
>>
------------------------------------------------------------------------------

REVIEW 4.1
1. When does it seem appropriate to write a self-written function?
2. In writing a self-written function what must be the first word
in the first executable statement?
3. A self-written function usually has both an input and an output.
Where does the input come from? Where does the output go to?
4. If a self-written function has more than one output, how must
the output be presented?
5. How does a self-written function communicate with the calling
program?
6. What can be said about variables in the self-written function
that are not in the input or output arguments of the function
and there are no global statements?
7. Do the variable names in the input and output arguments between
the calling program and the function have to be the same?
8. If a programmer wishes to write a self-written function, but
does not wish to create an additional .m file, what can the pro-
grammer do and what is the constraint?
Self-Written Functions and MATLAB®’s fminbnd Function 113

4.4 MATLAB’s fminbnd

There are times when we might be interested in determining a relative mini-
mum of a single variable function. MATLAB has a built-in function that will
do this for us. The syntax for the function is

[x,FVAL] = fminbnd(FUN,x1,x2)

The function fminbnd determines the relative minimum of a single variable

function in the interval x1<x<x2. The FUN argument is a function handle to
the function that describes the function whose relative minimum we wish
to determine. The arguments x1,x2 give the interval in which the relative
minimum may lie and the output x is the x value at the relative minimum.
FVAL gives the functional value at x. FUN can be a function defined in a
separate .m file or may be defined by an anonymous function or within
the fminbnd function itself with the use of the function handle creator,
@, as shown in Example 4.5. Note that MATLAB does not have a separate
function to find a relative maximum. In order to find a relative maximum,
redefine FUN to return the negative value of the function of interest, and
then use fminbnd to find the relative minimum (see Example 4.5). This is
also an example in which the user needs to write a self-written function in
order to use a MATLAB built-in function.

Example 4.5
Given: y( x ) = x 3 + 5.7 x 2 − 35.1x + 85.176.
Determine the relative minima and maxima.

% Example_4_5.m
% Find the minima and maxima of y = x^3 + 5.7x^2 – 35.1x + 85.176
clc; clear;
% First, plot the function so that we can determine the x range to use
% in fminbnd. Let us assume that the relative minimum lies between
% x1=-10, x2=6.
xf=-10:0.1:6;
y=xf.^3 + 5.7*xf.^2 − 135*xf + 85.176
plot(xf,y), xlabel('x'),ylabel('y'), grid, title('y vs x');
% Next, find the minimum and maximum using MATLAB's anonymous
% function method directly in the fminbnd function.
[xmin,ymin] = fminbnd( @(x) (x^3+5.7*x^2-35.1*x+85.176),-10,6 );
fprintf('xmin=%7.3f ymin=%9.3f \n',xmin,ymin)
% Note: To find a maximum, instead find the minimum of the
% negative of the function.
114 MATLAB® Essentials

[xmax,ymax] = fminbnd( @(x) -(x^3+5.7*x^2-35.1*x+85.176),-10,6);

Ymax=-ymax;
% Print results
fprintf('xmax=%7.3f Ymax=%9.3f \n',xmax,Ymax);
-----------------------------------------------------------------------

Program Results:
xmin= 2.013 ymin= 45.774
xmax= -5.813 Ymax= 285.394
>>

See Figure 4.2.

-----------------------------------------------------------------------

y vs x
300

250

200

150
y

100

50

0
−10 −8 −6 −4 −2 0 2 4 6
x

FIGURE 4.2
Relative maximum and minimum of y versus x.
Self-Written Functions and MATLAB®’s fminbnd Function 115

Projects
P4.1. Before scientific calculators and computers existed, numerical values
for functions such as ln x, ex, cos x, and so on were given in tables. The table
values were determined by power series, such as a Maclaurin or a Taylor
series. For example, the cosine function can be represented by the following
series:

x2 x4 x6 xn
cos x = 1 − + − + ....... ..... valid for [ x 2 < ∞] (P4.1)
2! 4! 6! n!

where n is an even number and x is in radians. The project involves creat-

ing a MATLAB program that contains both a calling program and a func-
tion. The calling program is to continuously ask the user if he/she wishes
to know the value of the cosine at a specific angle ranging from 0 to 360
degrees. If the answer  is N for no, exit the program. If the answer is Y
for yes, the calling program is to ask the user to enter an angle from the
keyboard. The calling program is to convert the entered angle to radians,
which is the x value to be used in Equation P4.1. The calling program then
is to call the function using the x value as an input to the function. The out-
put from the function should be the value of the cosine function evaluated
at x by Equation P4.1. The calling program is then to print the angle, the x
value, and the cosine value to the screen. The calling program is to repeat the
process until the user responds with a N.

P4.2. Repeat Project 4.1, but this time use the series expansion for the sine
function, which is

x x3 x5 xn
sin x = − + − + ..... .... valid for [ x 2 < ∞] (P4.2)
1! 3 ! 5 ! n!

where n is an odd number.

P4.3. In this project we consider the fluid level, h, in a tank, as it discharges

through a small circular hole (orifice) of diameter, d, near the bottom of the
tank (see Figure P2.4, page 75).
The tank has a circular cross section of diameter D. A formula describing
the fluid level, h, in the tank is
2
Cd Ao  C A  g t2
h = ho − 2 g ho t +  d o  × (P4.3a)
AT  AT  2
116 MATLAB® Essentials

where:
Cd is the discharge coefficient
ho is the fluid level in the tank at time, t = 0
Ao is the area of the orifice
AT is the cross-sectional area of the tank

The discharge coefficient, Cd , for a particular tank and orifice is determined

by experiment. Create a MATLAB program that consists of the following:

1. A calling program that calls a self-written function with input

arguments of Ao , AT , ho , and Cd . Take d = 0.0055 m, D = 0.146 m,
ho = 0.288 m, g = 9.81 m/s2 and Cd = 0.6, where d and D are the diam-
eters of the orifice and tank, respectively.
2. A self-written function as a .m file with output argument vectors h and t.

Take t = 0 to 200 seconds in steps of 4 seconds and g = 9.81 m/s2.

In the calling program print to a file a table consisting of h versus t, with
table headings and units. Also plot a graph of h versus t.

P4.4. This project is a variation of Project P4.3.

Create a MATLAB program that determines h(t) by the use of an anony-
mous function for Equation P4.3a. Take Cd = [0.5 0.6 0.7] and t = 0 to
200 seconds in steps of 4 seconds. For each Cd create a table of h versus t,
with table headings. For each Cd create a plot of h versus t, all on the same
page.
Use the same parameters that was used in Project P4.3, that is, take
d = 0.0055 m, D = 0.146 m, ho = 0.288 m, and g = 9.81 m/s2.

P4.5. Several properties of a refrigerant as a function of temperature are

shown below in Table P4.1:
TABLE P4.1
Refrigerant Properties
T(°C) v (m3/kg) u(kJ/kg)
−20 0.31003 206.12
−10 0.34992 224.97
0 0.36433 232.24
10 0.37861 239.69
20 0.39279 247.32
30 0.40688 255.12
40 0.42091 263.10
50 0.43487 271.25
60 0.44879 279.58
Self-Written Functions and MATLAB®’s fminbnd Function 117

We want to determine the properties of the refrigerant at the following

temperatures (−12, 18, 32, 57):

1. Create a data file of the above data.

2. Create a MATLAB program that consists of a calling program and a
self-written function.

Calling program should do the following:

1. Load the data file and create vectors Tt, vt, ut. Also create a vector,
T, that includes the temperatures at which the properties are to be
determined.
2. Call the self-written function using input variables of Tt, vt, ut, and
T and output variables T, v, and u.
3. Print to the Command Window a table consisting of T, v, and u with
table headings and units.
4. Plot vt versus Tt and on the same graph, plot v versus T as small x’s.
5. Plot ut versus Tt and on the same graph, plot u versus T as small x’s.

Self-written function should do the following:

Using the input variables and MATLAB’s interp1 function, interpolate for the
properties of the refrigerant at the temperatures, T, and return those values to
the calling program.

P4.6. This project is a variation of Project P2.5. In that project, we discussed

the pressure drop, p1 − p2, in a circular pipe having a flow rate, Q [1]. We will
repeat the governing equations involved in the process.

ρ V2 L
p1 − p2 = f (P4.6a)
2 D

where:
`ρ is the fluid density (kg/m 3 )
V is the average fluid velocity in the pipe (m/s)
Q = AV is the volume flow rate in the pipe (m 3/s )
D is the pipe diameter (m)
A is the pipe cross-sectional area = (πD2/4) (m2)
L is the pipe length between points 1 and 2 (m)
f is friction factor
118 MATLAB® Essentials

The friction factor has been determined by experiment. For smooth pipes a
formula that approximates the experimental data is [2]

f = ( 1.82 log 10 Re − 1.64 )

−2
(P4.6b)

where:
VD
Re = (Reynolds number) (P4.6c)
υ

 m2  µ
υ is the fluid viscosity  = (P4.6d)
 s  ρ
Develop a MATLAB program that contains both a calling program and a
function that determines the pressure drop, p1 − p2 versus the flow rate, Q. Q
is to vary from 0.001 to 0.02 m3/s in steps of 0.001 m3/s. Properties of D, L,
ρ , υ and Q are to be defined in the calling program. These values should be
made as an input argument to the function. The output from the function is to
be the pressure drop, p1 − p2, in the pipe and returned to the calling program.
In the calling program, a table of p1 − p2 versus Q is to be printed out to the
Command Window, including table headings and units. Use the following
values: ρ = 1000 kg m3, D = 0.16 m, L = 5 m and υ = 1.2 × 10 −6 m 2/s .

P4.7. This project is a modification of Project P3.5. That project involved the
mass motion in a mass-spring-dashpot system. A sketch of such a system is
shown in Figure 2.19. Disturbing the mass position, y, from its equilibrium
position and releasing it with zero velocity, will result in the y position vary-
ing with time, t.
As discussed in Project P3.5, the type of motion that the mass will have
depends on the values of the system properties of m, k, and c, where m is the
mass, k is the spring constant, and c is the damping factor [3].
If, (k/m) > (c/2 m)2 , then the mass motion will be damped oscillations and
the governing equation describing the motion is

  2   2  
 c  k  c  k  c  
y = exp  − t   A cos  −  t  + B sin  −  t  (P4.7a)
 2m   m  2 m    m  2 m  
    
 

The coefficients A and B are determined by initial conditions. Suppose we

take the initial displacement to be 0.5 m. Then

Ac
A = 0.5 m, B =
2 m arg

where arg = k/m − ( c/2m )

2
Self-Written Functions and MATLAB®’s fminbnd Function 119

If, k /m < (c/2m)2 , then the mass motion will be damped exponential
motion and the governing equation will be

 c   
 c 
2
k  
 c 
2 
k  
y = exp  − t   A exp    − t + B exp  −   − t  (P4.7b)
 2 m     2m  m    2m  m  
   

where exp( x) = e x. For this case,

 c  c
0.5  − arg +  − arg −
 2m  2 m
A= ,B= A
2 − arg c
− arg +
2m

If (c/2 m)2 = k/m, then the system is critically damped. For this case, the
solution is

 c 
y = ( A + Bt)exp  − t (P4.7c)
 2m 

For this case,

c
A = 0.5 m, B = A
2m

Construct a MATLAB program that consists of two parts: a calling program

and a function. The calling program is to create four different vectors con-
taining values of m, k, and c. The calling program is to use these vectors as an
input to the function. The function is to determine which of the three equa-
tions to use in calculating vector y as a function of vector t. Have the function
return vectors y and t to the calling program where it is to create plots of
y versus t. Take t = 0 to 20 s, in steps of 0.1 s. Values of m, k, and c for the four
cases are listed below:

Case 1

N N−s
m = 75 kg , k = 85 , c = 200
m m

Case 2

N N−s
m = 80 kg , k = 150 , c = 40
m m
120 MATLAB® Essentials

Case 3
N N−s
m = 50 kg , k = 50 , c = 100
m m

Case 4
N N−s
m = 100 kg , k = 200 , c = 20
m m

P4.8. Mathematician Joseph Fourier is credited with the theorem that any
periodic waveform may be expressed as a summation of pure sines and
cosines [4]. For example, the square wave of Figure P4.1a can be written as a
sum of sine terms:
v

−1

T = 1 ms
(a)

v 1st term

Sum of 1st, 2nd,

1 and 3rd terms

2nd term 3rd term

−1

(b)

FIGURE P4.1
(a) Square wave and (b) using 1st, 2nd, and 3rd terms in the series and their sum.
Self-Written Functions and MATLAB®’s fminbnd Function 121

4 2π t 4 6π t 4 10π t
v(t) = sin + sin + sin +
π T 3π T 5π T
∞
(P4.8)
2π k t
∑
4
= sin
k =1
πk T
k odd

Figure P4.1b shows the first three terms of the series and their summation.

1. Write a MATLAB script that utilizes the self-written function

sqwave(n,T,i) that takes the following input arguments:
n is the number of terms of the Fourier series.
T is the period of the square wave in seconds.
i is the number samples points per period.
2. The function should return two arrays, t and v, each containing i
elements, where:
t is an array of i time points.
v is an array of i computed values of the nth degree approximated
square wave.
3. Run your sqwave(n,T,i) function and plot the results for the
following arguments:
T = 1 ms, i = 1001, n = 3, 10, 100
4. For n = 100, create a plot of v versus t.

P4.9. This project is a modification of Project P4.7. That project involved the
mass motion in a mass-spring-dashpot system. A sketch of such a system is
shown in Figure 2.19. Disturbing the mass position, y, from its equilibrium
position and releasing it with zero velocity, will result in the y position varying
with time, t. In this project we will consider the case k /m > ( c/2m ) , result-
2

ing in the mass motion being damped oscillations. The governing equation
describing the motion is

  2   2  
 c   k  c    k  c  
y = exp  − t   A cos − t + B sin − t  (P4.7a)
 2m   m  2 m    m  2 m  
    
 

The coefficients A and B are determined by initial conditions. We will take

the initial displacement to be 0.5 m and the initial velocity equal to zero. Then
Ac
A = 0.5 m, B =
2 m arg
122 MATLAB® Essentials

where:
2
k  c 
arg = − 
m  2m 

1. Plot y versus t for t = 0 to 20 s, in steps of 0.1 s.

2. Use MATLAB’s fminbnd function to determine the minimum y
position for t = 0 to 5 s.

Take m = 50 kg , k = 100(N/m), c = 20(N − s/m).

P4.10. This Project is a variation of Exercise E2.4. A basketball player shoots

the ball when he is 8 m from the center of the hoop, instead of the 6 m shown
in Figure 2.20. The ball is released at a velocity, Vo, and makes angle ϑo = 45°
with the horizontal. Using Newton’s second law and the initial conditions
and neglecting the drag on the basketball, we can determine the following
equations for the (x, y) position of the ball as a function of time, t:

x = Vo cos(ϑo ) t (2.5)
g 2
y = Vo sin(ϑo ) t + t (2.6)
2
Take the (x, y) position of the center of the hoop to be (xf , yf) = (8.0 m, 3.048 m),
ϑo = 45°, Vo = 9.5169 m, and time of flight, tf = 1.1888 s.

1. Create a plot of y versus t using 0 ≤ t ≤ tf in steps of (t f/10 ).

2. Using MATLAB’s fminbnd function determines the maximum
height reached by the basketball in its flight to the basketball hoop.

-----------------------------------------------------------------------

References
1. Bober, W., Introduction to Numerical and Analytical Methods with MATLAB for
Engineers and Scientist, CRC Press, Boca Raton, FL, 2014.
2. Bober, W., The use of the Swamee-Jain formula in pipe network problems,
Journal of Pipelines, 4, 315–317, 1984.
3. Thomson, W. T., Theory of Vibration with Applications, Prentice Hall, Englewood
Cliffs, NJ, 1972.
4. Bober, W., Stevens, A., Numerical and Analytical Methods with MATLAB for
Electrical Engineers, CRC Press, Boca Raton, FL, 2012.
5
Working with Characters and Strings

5.1 Introduction
There may be a time that you might wish to create a matrix consisting of a
string of characters and to print it out in a report. This chapter shows you
how to do that. In MATLAB®, characters and strings usually need to be
enclosed by single quotation marks.

Example 5.1
% Example_5_1.m
% This program demonstrates how to print out rows of character
% strings. This can be done by declaring a column vector where each
% element in the vector is a character string.
% Note that all row character strings must have the same number
% of columns and be enclosed by single quotation marks.
clear; clc;
% Assign a string column vector.
% Each row in the column vector must have the same number of columns.
parts=['Internal modem '
'Graphics adapter '
'CD drive '
'DVD drive '
'Floppy drive '
'Hard disk drive '];
for i=1:6
fprintf('%16s \n',parts(i,1:16));
end
----------------------------------------------------------------------------------------

Program Results:
Internal modem
Graphics adapter
CD drive
DVD drive
Floppy drive
Hard disk drive
>>
-----------------------------------------------------------------------

123
124 MATLAB® Essentials

Example 5.2
This example is an interactive program. The user has to input whether to print the
string matrix to the screen or to a file.

% Example_5_2.m
% This example is a modification of Example 5.1.
% The program asks the user if he/she wishes to have the
% output go to the screen or to a file.
% This example also illustrates the use of the switch statement.
clear; clc;
parts=['Internal modem '
'External modem '
'Graphics circuit board '
'CD drive '
'Hard disk drive '];
fprintf('Choose whether to send the output to the\n');
fprintf('screen or to a file named output.txt. \n\n');
var=input('Enter S for screen or F for file (without quotes)\n','s');
switch(var)
case 'S'
for i=1:5
fprintf('%22s \n',parts(i,1:22));
end
case 'F'
fo=fopen('output.txt','w');
for i=1:5
fprintf(fo,'%22s \n',parts(i,1:22));
end
fclose(fo);
otherwise
fprintf('you did not enter an S or a F, try again \n');
exit;
end
---------------------------------------------------------------------------------

Program Results:
Choose whether to send the output to the
screen or to a file named 'output.txt'.
Enter S for screen or F for file (without single quotes)
S
Internal modem
External modem
Graphics circuit board
CD drive
Hard disk drive
>>
--------------------------------------------------------------------------------
Working with Characters and Strings 125

Example 5.3
In this interactive example, we illustrate the use of the if-elseif ladder to establish
a letter grade when the user enters a numerical score.

% Example_5_3.m
% This example uses the if-elseif ladder.
% The program determines a letter grade depending on the score the user
% enters from the keyboard.
clear; clc;
gradearray=['A'; 'B'; 'C'; 'D'; 'F'];
score=input('Enter your test score: \n');
fprintf('score is: %i \n',score);
if score > 100
fprintf('error: score is out of range. Rerun program \n');
break;
elseif score >= 90 && score <= 100
grade=gradearray(1);
elseif score >= 80 && score < 90
grade=gradearray(2);
elseif score >= 70 && score < 80
grade=gradearray(3);
elseif score >= 60 && score < 70
grade=gradearray(4);
elseif score < 60
grade=gradearray(5);
end
fprintf('grade is: %c \n',grade);
---------------------------------------------------------------------------------

Program Results:
Enter your test score: 76
score is: 76
grade is: C
>>
--------------------------------------------------------------------------------

Example 5.4
This example is a modification of Example 5.3. In this interactive example a for loop
is used to establish the interval containing the grade.

% Example_5_4.m
% The program determines a letter grade depending on the score the user
% enters from the keyboard.
% This version uses a loop to determine the correct interval of
126 MATLAB® Essentials

% interest. For a large number of intervals, this method is more

% efficient (fewer statements) than the method in Example 5.3
clear; clc;
gradearray=['A'; 'B'; 'C'; 'D'; 'F'];
sarray=[100 90 80 70 60 0];
score=input('Enter your test score: \n');
fprintf('score is: %i \n',score);
% The following 3 statements are needed for the case when score = 100.
if score == 100
grade=gradearray(1);
else
for i=1:5
if score >= sarray(i+1) && score < sarray(i)
grade=gradearray(i);
break;
end
end
end
fprintf('grade is: %c \n',grade);
---------------------------------------------------------------------------

Program Results:
Enter your test score: 82
score is: 82
grade is: B
>>
---------------------------------------------------------------------------

Example 5.5
This example combines the use of a string matrix and the establishment of a grade.

% Example_5_5.m
% This program determines the letter grades of several students.
% Student's names and their test scores are entered in the program.
% Student names are not connected to real people.
% This example uses nested 'for' loops and an 'if' statement
% to determine the correct letter grade for each student.
clear; clc;
gradearray=['A'; 'B'; 'C'; 'D'; 'F'];
sarray=[100 90 80 70 60 0];
Lname=['Smith '
'Lambert '
'Kurtz '
'Jones '
'Hutchinson'
'Blake '];
Fname=['Joe '
'Jane '
'Howard '
'Mary '
'Peter '
'Henry '];
score=[84; 86; 67; 92; 81; 75];
Working with Characters and Strings 127

avg_score=mean(score);
fprintf('The group average numerical grade is:%4.1f \n',avg_score);
% Index j selects the student and index i selects the letter grade.
% The score = 100 is treated separately.
for j=1:6
if score(j) == 100
grade(j)=gradearray(1);
else
for i=1:5
if score(j) >= sarray(i+1) && score(j) < sarray(i)
grade(j)=gradearray(i);
end
end
end
end
fprintf('Last name First name grade \n');
fprintf('--------------------------------------\n');
for j=1:6
fprintf('%12s %10s %c \n',
Lname(j,1:12), Fname(j,1:10), grade(j));
end
--------------------------------------------------------------------------

Program Results:
The group average numerical grade is:80.8
Last name First name grade
--------------------------------------
Smith Joe B
Lambert Jane B
Kurtz Howard D
Jones Mary A
Hutchinson Peter B
Blake Henry C
>>
---------------------------------------------------------------------------

5.2 MATLAB’s textscan Function

There may be occasions when you wish to enter information into a pro-
gram from a data file that contains both numerical and text data. MATLAB’s
textscan function is best suited for this operation.
Syntax:
C = textscan(fo, format)

The function will read data from an open text file identified by fo into a cell
array C. The format is of the form %f, %d, %c, %s, and so on. The number of
format specifiers determines the number of cells in the cell array C. Each cell
will contain the number of lines contained in the data file and be of the type
128 MATLAB® Essentials

specified by the format statement. String specifiers also include %q, which
is a string enclosed by double quotation marks. In the textscan statement,
the format for a string of n characters is %nc, but in the print statement, the
format for a string of n characters is %ns.

NOTE: To reference the contents of a cell, enclose the cell number with {}. See
the following example:

If you wish to read in N lines from the open data file, use

C = textscan('fo', format,N)

Example 5.6
Loads mixed text and numerical data from a file

% Example_5_6.m
% Load the product data from inv4.txt into the arrays 'cat_nu', 'desc',
% 'cost', and print the results to the screen.
clear; clc;
fo=fopen('inv4.txt'); % Note: inv4.txt is defined below.
C = textscan(fo,'%d %14c %f',5);
% Contents of cell block C contains 5 rows and 3 columns
cat_nu = C{1};
desc = C{2};
cost = C{3};
fclose(fo);
fprintf('catalog # description \t\t cost \n');
fprintf('---------------------------------------\n');
for i=1:5
fprintf('%5i \t %14s \t %6.2f \n\n',...
cat_nu(i),desc(i,1:14),cost(i));
end
-----------------------------------------------------------------------
% inv4.txt file (do not include this line in your data file)
1001 hammer 2.58
1002 plier 1.20
1003 screwdriver 1.56
1004 soldering iron 3.70
1005 wrench 2.60
---------------------------------------------------------------------------------

Program Results:
catalog # description cost
---------------------------------------
1001 hammer 2.58
1002 plier 1.20
1003 screwdriver 1.56
1004 soldering iron 3.70
1005 wrench 2.60
>>
---------------------------------------------------------------------------------
Working with Characters and Strings 129

REVIEW 5.1

1. Suppose you wish to assign a column vector consisting of

string elements, what are the conditions that need to be fol-
lowed in setting up this column vector?
2. Suppose that you had a data file that contains both numerical
and text data, what command would you use to read in the data
into your program.
3. When the command used in reading in the data type described
in item 2, what object type does the data go into?
4. To assign variable names to items in the object which of the
following three symbols would you use: (), [], {}?

Projects
P5.1. Create a MATLAB program that

1. Contains a string array that lists the names of five courses that you
have taken recently.
2. Contains a vector of the numerical grades received in each of the five
courses.
3. Prints to the screen the names of the five courses and their grades.
4. Prints to the screen the average grade received in the five courses.

P5.2. You are a student adviser and you wish to have the user (student)
tell you the grades that he/she received on several courses. You are to use
MATLAB’s input command to ask the student to enter the grade that he
received in the course.
Create a MATLAB program that

1. Creates a string matrix of five courses.

2. Use a for loop and MATLAB’s input command asking the student
to enter the grade that he/she received in the course listed on the
screen. List the courses one at a time.
3. Print to the screen a summary of the results.

NO T E : Allowable grades are only letters (A, B, C, D, F, I).

P5.3. Create a MATLAB program that creates a vector listing all the days of
the week, that is, Monday, Tuesday, and so on. Create a second vector that
describes two activities that you wish to do on each day of the week, such as
130 MATLAB® Essentials

Attend calculus class and have dinner with girl or boy friend, and so on. Do this
for each day of the week. Use an input statement to have the user enter the
day of the week. Based on that entry and using a Switch Statement print out
the activities of that day listed in the second vector.

P5.4. Bob’s Hardware Store wishes to create an online program to sell inven-
tory items in its store. You are to create an interactive MATLAB program for
this purpose. The program is to contain a data file, a main program and a
billing function.

Data file:
The data file is to contain a catalog number and a description of the
inventory items for sale, their costs, and the quantity available for
sale. The list should contain at least 10 items. The data file is to be
loaded into the main script.

Main script:
1. The script is to print to the screen the items for sale, including their
catalog numbers and descriptions, their cost, and the quantity
available for purchase.
2. The script is to ask the user if he/she wishes to make a purchase. If
no, exit the program. If yes, the script is to ask the user for his/her
first and last name. It should then ask the user the catalog number of
the item he/she wishes to purchase and the quantity.
3. The script is to continue asking the user if he/she wishes to make
another purchase. If the response is no, call the billing function and
exit the program. If the response is yes, print to the screen the list
of items, their cost, and the updated items available for purchase.
Then, the script is to ask the user the catalog number of the item
he/she wishes to purchase and the quantity.
4. Billing function:
Input arguments:
First and last names of the purchaser, the number of items purchased,
the catalog numbers, their costs, the quantities of the items purchased.
Output arguments:
None, use open and closed brackets.
Use a global statement for the description of the items purchased
(both in the calling program and in the billing function). The bill
should contain the name and address of the store, the customer’s
first and last names, bill headings, the catalog numbers of all the
items purchased, their unit prices, the total price for each item pur-
chased, and finally the total price of all the items purchased.
Clear the screen, then print the bill to the screen.
6
Roots of Algebraic and
Transcendental Equations

6.1 Introduction
In the analysis of various engineering problems, we are often faced with a
need to find roots of equations whose solution is not easily found analyti-
cally. Given a function f(x), the roots of the function are the values of x that
makes f(x) = 0. For example, the equation

f ( x) = ax 2 + bx + c = 0 (6.1)

where a, b, and c are constants, is an equation that we are all familiar with.
The values of x that satisfy the equation are the roots of f(x). We even have a
formula for the roots, which are

−b ± b 2 − 4 a c
x= (6.2)
2a

We see that there are two roots, x1 and x2, where

−b + b 2 − 4 a c −b − b 2 − 4 a c
x1 = , x2 = (6.3)
2a 2a

More complicated examples include nth degree polynomials and tran-

scendental equations containing trigonometric, exponential, or logarithm
functions. In this chapter, we discuss the search method for obtaining a
small interval in which a root lies. We then discuss MATLAB®’s fzero and
roots functions, which may be used to obtain a more accurate value for the
roots of type of equations just stated.

131
132 MATLAB® Essentials

6.2 Search Method

In the search method, we seek a small interval that contains a real root. This
only gives an approximate value for the real root. Once an interval in which
a real root lies has been established, several different methods, including
the Bisection method, Newton–Raphson method, and MATLAB’s fzero and
roots functions, can be used to obtain a more accurate value for the real
root. In this book, we will give a brief discussion of the Bisection method, but
emphasize MATLAB’s fzero and roots functions.
The search method is especially useful if there is more than one real
root. The equation whose roots are to be determined should be put into the
following standard form:
f ( x) = 0 (6.4)

We proceed as follows: first we subdivide the x domain into N equal subdivi-

sions of width ∆x, giving

x1 , x2 , x3 , xN +1 with xi +1 = xi + ∆x

Then, determine where f ( x) changes sign (see Figure 6.1).

This occurs when the signs of two consecutive values of f ( x) are different,
that is,
f ( xi ) f ( xi +1 ) < 0

The sign change usually indicates that a real root has been passed. However,
it may also indicate a discontinuity in the function. (Example: tan x is
discontinuous at x = π 2.)
A brief description of the Bisection method follows:
f (x)

f (x3) f (x4)
x1 x2
x
x3 x4 ... xN xN+1
f (x2)

f (x1)

FIGURE 6.1
The root of f(x) lies between x2 and x3.
Roots of Algebraic and Transcendental Equations 133

6.3 Bisection Method

Suppose it has been established by the search method, that a root lies between
xi and xi+1. The concept in the bisection method is to cut the interval contain-
ing the root in half, determine which half contains the root, cut that interval
in half, determine which half contains the root, and continue the process
until the interval containing the root is sufficiently small, so that any point
within the last interval is a very good approximation for the root. A more
detailed description follows: Let xi+ 1 2 be the midpoint position of the first cut,
then xi + 1 2 = xi + (∆x 2) (see Figure 6.2). Now compute f ( xi ) f ( xi + 1 2 ):

Case 1: If f ( xi ) f ( xi + 1 2 ) < 0, then the root lies between xi and xi+ 1 2

Case 2: If f ( xi ) f ( xi + 1 2 ) > 0, then the root lies between xi+ 1 2 and xi+1
Case 3: If f ( xi ) f ( xi + 1 2 ) = 0, then xi or xi+ 1 2 is a real root

For cases 1 and 2, select the interval containing the root and repeat the pro-
cess. Continue repeating the process, say r times, then (∆x) f = ∆x 2 r , where
∆x is the initial size of the interval containing the root before the start of the
bisection process (∆x = xi +1 − xi ) and (∆x) f is the size of the interval contain-
ing the root after r bisections. If (∆ x) f is sufficiently small, then a very good
approximation for the root is anywhere within the last bisected interval, say
the midpoint of the interval. For example, in 20 bisections,
∆x
(∆x) f = ≈ ∆x × 1.0 × 10 −6
2 20
---------------------------------------------------------------

MATLAB has built-in functions to determine the roots of a function of one

variable, such as a transcendental equation or an nth degree polynomial.
The fzero function is used for transcendental equations and will determine
the real roots of the equations. The roots function is used for polynomial
equations and will return both the real and imaginary roots. First we will
discuss the fzero function.

f (x)

x
xi xi+½ xi+1

FIGURE 6.2
Bisecting the interval containing the root.
134 MATLAB® Essentials

6.4 MATLAB’s fzero Function

The fzero function is for a single variable nonlinear function whose root we
wish to determine. The syntax for the fzero function is
X = fzero(FUN,X0) (6.5)

where FUN is a function handle to the function whose root is to be deter-

mined, X0 is a scalar and represents an initial guess for the root, and X is
the root determined by the fzero function. FUN may be a separate .m file
or an anonymous function. You may also enter the function directly into the
fzero function using the function creator, @, as shown in Example 6.2. To
get a good value for X0, consider plotting the function and noting where
the function crosses the x-axis.

Example 6.1
Given the equation

3.3 0.325
f(v) = − −2 (6.6)
v − 0.03 v(v + 0.03)

Determine the root of f(v).

% Example_6_1.m
% Simple use of fzero function
% Find the root of f(v)=3.3/(v-0.03)-0.325/(v(v+0.03))-2
% Guess that the root lies between 1 and 2.
% Function name is fv and the function is a .m file.
clear; clc;
v=1.0:0.1:2.0;
for j=1:length(v)
f(j)=fv(v(j));
end
plot(v,f), xlabel('v'), ylabel('f'), title('f vs. v'), grid;
root=fzero('fv',1.6);
% Note: we could also have used root=fzero(@fv,1.6)
fprintf('root=%6.4f \n',root);
froot=fv(root);
fprintf('froot=%12.4e \n',froot);
----------------------------------------------------------------------------
% This function works with Example 6.1
function f=fv(v)
f=3.3/(v-0.03)-0.325/(v*(v+0.03))-2;
------------------------------------------------------------------------------

Program Results:
root= 1.5810
froot= 0.0000e+00
>>
See Figure 6.3.
-----------------------------------------------------------------------
Roots of Algebraic and Transcendental Equations 135

f vs. v
1.2

0.8

0.6

0.4
f

0.2

−0.2

−0.4

−0.6
1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2
v

FIGURE 6.3
Plot of f versus v.

Example 6.2
In the next two examples, we will determine the root of Equation 6.5 by placing
an anonymous function directly into the fzero function and also as by placing
an anonymous function in the main script. In both cases no separate .m file is
used.

% Example_6_2.m
% Anonymous function applied directly into the fzero function
% Find the root of f(v)=3.3/(v-0.03)-0.325/(v(v+0.03))-2
% Guess that the root lies between 1 and 2.
clear; clc;
root=fzero(@(v) (3.3/(v-0.03)-0.325/(v*(v+0.03))-2),1.6);
fprintf('root=%6.4f \n',root);
froot=fv(root);
fprintf('froot=%12.4e \n',froot);
-------------------------------------------------------------------------

Program Results:
root= 1.5810
froot= 0.0000e+00
>>
------------------------------------------------------------------------

% Example_6_3.m
% Anonymous function used with the fzero function.
% Find the root of f(v)=3.3/(v-0.03)-0.325/(v(v+0.03))-2
136 MATLAB® Essentials

% Guess that the root lies between 1 and 2.

clear; clc;
fv=@(v) (3.3/(v-0.03)-0.325/(v*(v+0.03))-2);
v=1.0:0.1:2.0;
for j=1:length(v)
f(j)=fv(v(j));
end
plot(v,f), xlabel('v'), ylabel('f'), title('f vs. v'), grid;
% Note: When using an anonymous function for FUN do not use
% either the @ sign or enclose the name of the anonymous function
% with single quotation marks.
root=fzero(fv,1.6);
fprintf('root=%6.4f \n',root);
froot=fv(root);
fprintf('froot=%12.4e \n',froot);
-------------------------------------------------------------------------

Program Results:
The results are exactly the same that was obtained in Example 6.1.
------------------------------------------------------------------------

In some instances, we would like to find the zero of a function of two argu-
ments, say X and P, where P is a parameter and is fixed. In order to solve
with fzero, P must be defined in the calling program. For example, suppose
myfun is defined in a .m file as a function of two arguments:

function f = myfun(X,P)
f = cos(P*X);

The fzero statement would need to be invoked as follows:

P = 1000;
root = fzero(@(X) myfun(X,P),X0)

where root is the zero of function myfun when P=1000. Note that P needs
to be defined before the fzero function is called.
An alternative to adding parameter P as an argument in myfun is to use
MATLAB’s global statement. The parameter P should be defined in the call-
ing program and be listed as a variable in the global statement. The global
statement needs to be used in both the calling program and in the function
myfun and be exactly the same in both scripts. See the following example.

Example
Calling program:

global P;
P=1000;
Xo=10.0;
root = fzero(@myfun,Xo);
-------------------------------------------------------------------------
Roots of Algebraic and Transcendental Equations 137

% The file myfun.m:

function f = myfun(x)
global P;
f = cos(P*x);
--------------------------------------------------------------------------------

Now, let us consider the case when there is more than one root in the function
under consideration. First, it is best to use the search method to obtain small
intervals in which the roots lie. For each obtained interval define the argu-
ment X0 in the fzero command (see Equation 6.5) as a vector of length two;
that is, X0(1) is the X position at the beginning of one of the found intervals
and X0(2) is the X position at the end of that interval. This should result in
the sign of FUN(X0(1)) to differ from the sign of FUN(X0(2)). If that is
not the case, MATLAB will return an error message. The following example
illustrates this concept:

Example 6.4
The position, y, of a mass in a mass-spring-dashpot system (see Figure 2.19) that is
underdamped is given by

  k 2   k 2 
 c   c    c  
y = exp  − t   A cos  −  t + B sin  −  t  (6.7)
 2m    m  2m    m  2m   
    

Determine the number of roots, their values and y values at the obtained roots for
0 ≤ t ≤ 10 s. Take
m = 25.0 kg, c = 5.0 N-s/m, k = 200.0 N/m, A = 0.2 m, and

c A
B= ×
2m k/m − (c/2m)2

% Example_6_4.m
% This program determines the number of roots and their values
% in the mass-spring-dashpot system in the time span 0<=t<=10 s.
% The governing equation for the displacement, y(t), of the
% under-damped vibration problem is:
% y(t)=exp(-c*t/2/m)(A*cos(arg1*t)+B*sin(arg1*t)), where
% arg1=sqrt(k/m-(c/2/m)^2)
% k=200 N/m, m=25kg, c=5 N-s/m
% A=0.2m, B=c/(2*m)*A/arg1.
% The values for constants A and B represent an initial
% displacement of the mass from its equilibrium position
% at zero velocity.
% A global statement in used to bring the constants, k, m, c,
% A, B and arg1 into the function fun_spring.
clear; clc;
global m k c A B arg1;
m=25; k=200; c=5; A=0.2;
arg1=sqrt(k/m-(c/2/m)^2);
B=c/2/m*A/arg1;
138 MATLAB® Essentials

% ir is the root number

ir=0;
t=0:0.1:10;
for i=1:length(t)
y(i)=fun_spring(t(i));
end
plot(t,y), xlabel('t(s)'), ylabel('y(m)'),
title('mass displacement, y, vs. t'), grid;
for i=1:length(t)-1
if y(i)*y(i+1) <= 0.0
ir=ir+1;
tr(1)=t(i);
tr(2)=t(i+1);
root(ir)=fzero('fun_spring',tr);
y(ir)=fun_spring(root(ir));
end
end
if ir ~= 0
fprintf('root # root(s) y(root) (m) \n');
fprintf('-----------------------------------------\n');
for j=1:ir
fprintf('%3i %10.6f %12.4e \n',j,root(j),y(j));
end
else
fprintf('\n\n No roots lie within 0 <= t <= 20 s');
end
-------------------------------------------------------------------------
% fun_spring.m
% This function is used in Example 6.4
% The function determines the spring position, y(t), as a
% function of time
function y=fun_spring(t)
global m k c A B arg1;
y=exp(-c*t/2/m)*(A*cos(arg1*t)+B*sin(arg1*t));
-------------------------------------------------------------------------
NO T E : We could have avoided the use of the global statement by defining the values
of m, k, c, A, B, and arg1 in the function instead of the main program. We chose to use
the global statement to illustrate its use.

Program Results:
root # root(s) y(root)(m)
-----------------------------------------
1 0.568218 -1.3111e-17
2 1.679634 -1.2099e-16
3 2.791049 -7.2174e-18
4 3.902465 1.0979e-16
5 5.013881 -2.9893e-16
6 6.125296 6.1582e-17
7 7.236712 -4.2906e-17
8 8.348127 2.8230e-17
9 9.459543 -1.5493e-17
>>
See Figure 6.4.
---------------------------------------------------------------------------
Roots of Algebraic and Transcendental Equations 139

mass displacement, y, vs. t

0.2

0.15

0.1

0.05
y(m)

−0.05

−0.1

−0.15

−0.2
0 1 2 3 4 5 6 7 8 9 10
t(s)

FIGURE 6.4
Plot of mass motion in a mass-spring-dashpot system.

6.5 MATLAB’s roots Function

MATLAB has a function named roots that obtains the roots of a polyno-
mial. The function determines both real and imaginary roots of the specified
polynomial.
The syntax for the function is
V = roots(C) where C is a vector specifying the coefficients of the polyno-
mial and V is the roots. If C has n + 1 components, the polynomial is

C(1) x n + C(2) x n −1 + ... C(n) x + C(n + 1) = 0

Thus, to find the roots of the polynomial ax 4 + bx 3 + cx 2 + dx + e = 0, run

roots([a b c d e]). The roots function will give both real and imagi-
nary roots of the polynomial.
Some additional useful MATLAB functions are

poly(V) finds the coefficients of the polynomial whose roots are V.

real(V) gives the real part of V.
imag(V) gives the imaginary part of V.
140 MATLAB® Essentials

Example 6.5
In this example, MATLAB’s roots function is used to find the roots of a polynomial.

% Example_6_5.m
% This program determines the roots of two different polynomials
% using MATLAB’s 'roots' function.
clear; clc;
% The first polynomial is: f=x^3-5.7*x^2-35.1*x+85.176. The
% roots of this polynomial are all real.
% Define coefficients of first polynomial (real roots)
fprintf('The coefficients of the first polynomial are: \n');
C=[1.0 -5.7 -35.1 85.176]
fprintf('The roots are: \n');
V=roots(C)
fprintf('Polynomial coefficients determined from poly(V) are:\n');
C_recalc=poly(V)
fprintf('----------------------------------------\n');
% The second polynomial is: f=x^3-9*x^2+23*x-65. The roots of
% this polynomial are both real and complex. Complex roots must
% be complex conjugates.
% Define the coefficients of second polynomial
fprintf('The coefficients of the second polynomial are \n');
D=[1.0 -9.0 23.0 -65.0]
fprintf('The roots are: \n');
W=roots(D)
fprintf('The real and imaginary parts of the roots are:\n');
re=real(W)
im=imag(W)
fprintf('Polynomial coefficients determined from poly(W) are:\n');
W_recalc = poly(W)
----------------------------------------------------------------------

Program Results:
The coefficients of the first polynomial are:
C =
1.0000 -5.7000 -35.1000 85.1760
The roots are:
V =
8.6247
-4.9285
2.0038
Polynomial coefficients determined from poly(V) are:
C_recalc =
1.0000 -5.7000 -35.1000 85.1760
-------------------------------------------------------------------------
The coefficients of the second polynomial are
D =
1 -9 23 -65
The roots are:
W =
7.0449 + 0.0000i
Roots of Algebraic and Transcendental Equations 141

0.9775 + 2.8759i
0.9775 - 2.8759i
The real and imaginary parts of the roots are:
re =
7.0449
0.9775
0.9775
im =
0
2.8759
-2.8759
Polynomial coefficients determined from poly(W) are:
W_recalc =
1.0000 -9.0000 23.0000 -65.0000
>>
------------------------------------------------------------------------------

REVIEW 6.1

1. What is meant by the term root of function f(x) = 0?

2. What is the objective in the search method for determining a
root of the equation f(x) = 0?
3. What is the name of the MATLAB function for determining the
roots of a transcendental equation of the form f(x) = 0?
4. In MATLAB’s function for determining the roots of a transcen-
dental equation, how does one define the function whose roots
are to be determined?
5. If you suspect that there is more than one real root, what
method should be used in combination with the MATLAB’s
fzero function to obtain the roots?
6. If you are using the search method in combination with the
fzero function, what can you say about the second argument
in the fzero function?
7. What is the purpose of the global statement?
8. If the function f(x) is a polynomial, what MATLAB function
should you use to obtain its roots?
142 MATLAB® Essentials

Projects
P6.1. This project is a variation of Project P2.1. In that project a tennis
player on serve places the tennis ball close to the outside line of the service
box when the ball hits the ground (see Figure P6.1a and b). The horizon-
tal distance from the point where the ball leaves the racket to where the
ball hits the ground is 18.925 m. The horizontal distance from the point
where the ball leaves the racket to where the net is located is 12.509  m.
The vertical distance above the ground when the ball leaves the racket
is yo = 2.438 m. The speed of the ball as it leaves the racket is 58.0 m/s.
Determine the angle, ϑo , that the ball makes with the horizontal on leav-
ing the players racket that would result in the ball hitting the ground at
the position stated. Neglecting drag, the governing equations describing
the motion are

x = Vo cos(ϑo ) t (P6.1a)

g
y = − t 2 − Vo sin(ϑo ) t + yo (P6.1b)
2
Hint: Let ( x f , y f ) be the position and t f be the time when the ball hits
the ground. Substitute these values in Equations P6.1a and P6.2a, then
solve for t f in the  modified Equation P6.1a and substitute the obtained
expression into the modified Equation P6.1b. This gives a transcendental

Ball hits ground

Ball path

Ball leaves
racket
Net
(a)
y
Ball leaves racket

Ball hits ground

Vo ϑo 2.438 m
0.914 m Net
x
12.509 m
18.925 m
(b)

FIGURE P6.1
(a) Tennis court layout. (b) Path of tennis ball after it leaves the racket.
Roots of Algebraic and Transcendental Equations 143

equation for ϑo where ϑo is the only unknown. Note yf = 0. Write a pro-

gram in MATLAB that

1. Plots f (ϑo ) versus ϑo.

2. Uses MATLAB’s fzero function to determine ϑo.
3. Determine whether the obtained value for ϑo will result in the ball
clearing the net.
4. Find the time, tf , it takes for the ball to hit the ground after it leaves
the racket.
5. Using circles, plot the x-y position of the ball for 0 ≤ t ≤ t f in steps of tf/10.
6. Determine the height of the tennis ball when it reaches the x position
of the net. Does the tennis ball clear the net?

Assume that ϑo can range anywhere from 0° to 8° in steps of 0.1°.

P6.2. This Project is a variation of Exercise E2.4. A basketball player shoots

the ball when he is 6 m from the center of the hoop as shown Figure 2.20. The
ball is released at a velocity, Vo = 8.71 m/s, and at an angle ϑo with the hori-
zontal. Using Newton’s second law and the initial conditions and neglecting
the drag on the basketball, we can determine the following equations for the
(x, y) position of the ball as a function of time, t:

x = Vo cos(ϑo ) t (P6.2a)
g 2
y= t + Vo sin(ϑo )t + yo (P6.2b)
2
Take the (x, y) position of the center of the hoop to be (xf , yf ) = (6.0 m, 3.048 m)
and yo = 1.98 m.

1. Determine the angle ϑo that will result in the ball reaching the center
of the hoop at time tf .
2. Determine the time, tf , that it takes for the ball to reach the center
of the hoop. Time, t, equals zero when the ball leaves the player’s
hands.
3. Create a table consisting of t, x, y for 0 ≤ t ≤ t f in steps of t f 10. Carry
variables to four decimal places. Print the table to an output file,
including tf and ϑo.
4. Create a plot of y versus x.

P6.3. The equation of state for a substance is a relationship between pres-

sure, p, temperature, T, and specific volume, v. Many gases at low pressures
and moderate temperatures behave approximately as an ideal gas. The ideal
144 MATLAB® Essentials

TABLE P6.1
Values of a, b and, R for Carbon Dioxide in Redlich–Kwong’s Equation of State

 N- m4 - K1/ 2   m3   Nm 
a  2  b  R  
Gas  kmol   kmol   K- kmol 

Carbon dioxide 65.43 × 105 0.02963 8314

Source: Moran, M. J. and Shapiro, H. N., Fundamentals of Thermodynamics, John Wiley & Sons,
Hoboken, NJ, 2004.

gas equation of state with p in N/m2, v in m3/kmol, T in K, and R in (N-m)/

(K-kmol) is
RT
p= (P6.3a)
v
where R is the universal gas constant and is the same for all gasses. As tem-
perature decreases and pressure increases, gas behavior deviates from ideal
gas behavior. The Redlich–Kwong’s equation of state is often used to approxi-
mate non-ideal gas behavior. Redlich–Kwong’s equation of state is [1]
RT a
p= − (P6.3b)
v − b v ( v + b ) T 1/2
or
RT a
f (v) = − −p = 0 (P6.3c)
v − b v ( v + b)T 1/2
The values for R, a, and b for carbon dioxide is tabulated in Table P6.1.
We wish to determine the % error in the specific volume by using the
ideal gas relationship while assuming that Redlich–Kwong’s equation of
state is the correct equation of state for carbon dioxide. Vary the tempera-
ture from 350–700 K in steps of 50 K, while holding the pressure constant
at 1.0132 × 107 N/m2 (100 atm). Using the specified temperatures and pres-
sure determine the specific volumes, v, by both the ideal gas equation and
the Redlich–Kwong’s equation and determine the % error in the specific
volume resulting from the use of the ideal gas equation. Take the % error in
the specific volume to be
v ideal gas − v Redlich − Kwong
% error = × 100 (P6.3d)
v Redlich − Kwong

Write a MATLAB program utilizing the fzero function to calculate the spe-
cific volume by Redlich–Kwong’s equation. Use the value of v obtained from
the ideal gas law as your guess for the root in MATLAB’s fzero function.
Construct a table as shown in Table P6.2.
Roots of Algebraic and Transcendental Equations 145

TABLE P6.2
v Determined by Redlich–Kwong Equation and by Ideal Gas Law for Carbon Dioxide
Ideal Gas Redlich–Kwong Equation
T(K) v (m3/kmol) v (m3/kmol) % Error in v
350 --------- ---------- ---------
400 --------- ---------- ---------
----- --------- ---------- ---------
----- --------- ---------- ---------
700 --------- ---------- ---------

P6.4. Determine the first root of the voltage, v(t), of the underdamped parallel
RLC circuit described in Project P2.7. The governing equation for v(t) is

  1 2   1 2 
 1   1    1  
v(t) = exp  − t   A cos  −  t + B sin  −   t  (P6.4)
 2RC    LC  2RC    LC  2RC   
    
Use MATLAB’s fzero function to find the first root. Print this value to the
screen. Also plot v versus t. Assume the following parameters:

R = 100 Ω, L = 10 −3 H, C = 10 −6 F, A = 6.0 V , B = −9.0 V

0 ≤ t ≤ 5.0 × 10 −4 s in steps of 1.0 × 10 −5 s.

P6.5. Repeat Project 6.3, but replace the Redlich–Kwong’s Equation with Van
der Waals’ Equation [1]. In addition, do the Project for all three gasses listed
in Table P6.3 by the use of a for loop. In your program, use an if-elseif
ladder within the for loop to select the proper constants for the gas. Van der
Waals’ equation of state is
RT a
p= − (P6.5)
v−b v 2

The constants a and b are tabulated in Table P6.3.

TABLE P6.3
Van der Waals’ Constants
 N m4   m3   Nm 
a  2  b  R 
Gas # Gas  kmol   kmol   K kmol 

1 Air 1.368 × 105 0.0367 8314

2 Oxygen 1.369 × 105 0.0317 8314
3 Carbon dioxide 3.647 × 105 0.0428 8314

Source: Moran, M. J. and Shapiro, H. N., Fundamentals of Thermodynamics, John Wiley & Sons,
Hoboken, NJ, 2004.
146 MATLAB® Essentials

P6.6. The temperature distribution of a thick flat plate, initially at a uniform

temperature, T0 , and which is suddenly immersed in a huge bath at a tem-
perature T∞ , is given by (see Figure P6.2a on the next page)

 x 2 2
∞ sin(δn )cos  δn  e − aδn t/L
T ( x , t) = T∞ + 2(T0 − T∞ ) ∑
n=1
 L 
cos(δn )sin(δn ) + δn
(P6.6a)

where:
L = 1/2 of the plate thickness
a is the the thermal diffusivity of the plate material
δn are the roots of the equation:
hL
F(δ) = tan δ − =0 (P6.6b)
kδ
where:
h is the convective heat transfer coefficient for the bath
k is the thermal conductivity of the plate material

There are an infinite number of roots to Equation P6.5b. This can be seen in
Figure P6.2b. The roots being δ1, δ2 , δ3 , …, δn. Note that δ1 lies between 0 and π/2,
δ2 lies between π and 3π/2, δ3 lies between 2π and 5π/2, and so on. Subtracting
T∞ from Equation P6.6a and dividing by T0 − T∞ , we obtain Equation P6.6c:

x   x  − aδn2t/L2
T  , t  − T∞ ∞ sin(δn )cos  δn e
TRATIO = 
L 
T0 − T∞
=2
n=1
∑  L
cos(δn )sin(δn ) + δn
(P6.6c)

A plot of TRATIO versus time, for several different values of x/L should
appear as shown in Figure P6.2c.
Finally, the heat transfer ratio, Qratio, from the plate to the bath in time t is
given by
∞

∑ δ [sin δ cos δ
Q(t) 2hL sin δn cos δn  2 2

QRATIO = = 1 − e − at δ n /L  (P6.6d)
n + δn ] 
2
Q0 k n =1 n n 

where:
Q(t) is the amount of heat transferred from the plate to the bath in time t.
Q0 is the amount of heat transferred from the plate to the bath in infinite time,
which equals the change in internal energy in infinite time.

1. Write a computer program that will solve for the roots, δ1, δ2, . . . . , δ50
using MATLAB’s fzero function. Print out the δ values in 10 rows and
5 columns. Also print out the functional values at the roots, that is, f(δn).
NO T E : Only 50 δ values were asked to be computed.
Roots of Algebraic and Transcendental Equations 147

Bath at Bath at
temperature T∞ Plate temperature T∞

Heat flow L L Heat flow

from plate from plate
to bath to bath

(a) x

tan δ, hL/kδ

tanδ

hL
kδ

0 π/2 3π/2 5π/2

δ
δ1 δ2 δ3 δ4
π 2π 3π

(b)

TRATIO

1.0
x/L = 0.0

x/L = 0.2
x/L = 0.4
x/L = 0.6
x/L = 1.0
t
(c) 0

FIGURE P6.2
hL
(a) Plate cooling in a bath. (b) Plot of tan δ and versus δ. (c) Plot of TRATIO versus t.
kδ

2. Solve Equation P6.6c for TRATIO for x/L = 0.0, 0.2, 0.4, 0.6, 0.8, 1.0
and t = 0, 10, 20, … 200 s. Print out results in table form as shown
in Table  P6.4. Also use MATLAB to produce a plot similar to
Figure P6.2c.
148 MATLAB® Essentials

TABLE P6.4
Temperature Ratio, TRATIO
X/L
Time(s) 0.0 0.2 0.4 0.6 0.8 1.0

0 1.0 1.0 1.0 1.0 1.0 1.0

10 --- --- --- --- --- ---
20 --- --- --- --- --- ---
--- --- --- --- --- --- ---
200 --- --- --- --- --- ---

3. Construct a table for QRATIO versus t for times 0, 10, 20, 30, . . . . , 200 s.
4. Use MATLAB to produce a plot of QRATIO versus t.

Use the following values for the parameters of the problem:

T0 = 300°C, T∞ = 30°C, h = 45 W/m2-°C,

k = 10.0 W/m-°C, L = 0.03 m, a = 0.279 × 10−5 m2/s

P6.7. In this project we consider a semiinfinite slab (such as thick layer

of ice) having a uniform temperature, Ti , that is suddenly subjected to a
change in air temperature caused by a warm front moving in over the
region of interest (see Figure P6.3). The surface temperature, Ts , of the
slab will be a function time, t. It will also depend on the parameters:
h, Ti , T∞ , k, and α, where h is the convective heat transfer coefficient, Ti
is the initial temperature of the slab and T∞ is the air temperature, k
and α are the thermal conductivity and diffusivity of the slab material
respectively. The governing equation describing the surface tempera-
ture, Ts , as a function of time [2] is:

h 2 αt
  h αt   Ts − Ti
1− e k2
 1 − erf    − =0 (P6.7)

  k   T∞ − Ti

Air at temperature T∞ Heat transfer

from air to ice

x Ice initially at temperature Ti

FIGURE P6.3
Ice slab subjected to warm front.
Roots of Algebraic and Transcendental Equations 149

Given: Ti = −20°C, T∞ = 20˚C, k = 2.22 W/m-C, α = 12.6 × 10−7 m2/s, and h = 100
W/m2-C. We wish to determine the time, t, when the surface temperature of
the slab reaches the following temperatures:

T = [−10 −5 0]°C

Use the search method to find an interval in which the root of Equation P6.7
lies. Then use MATLAB’s fzero function to solve for the time, t, for each
condition and print the results in a table with table headings. Assume that
0 ≤ t ≤ 1000 s with a step size of 10 s.

P6.8. A wood circular cylinder, having a specific gravity, S, floats in water

as shown in Figure P6.4. For a floating body, the weight of the floating body
equals the weight of fluid displaced. An equation that describes the depth, d,
of the submerged part of the floating cylinder is given by

f ( x) = ( x − 1) 2 x − x 2 + sin −1( x − 1) − (S − 0.5)π = 0 (P6.8)

where x = d R and S is the specific gravity of the wood.

For a complete derivation of Equation P6.8, see Project P5.3 in Reference 2.
Use the following parameters: R = 0.5 m and 0.3 ≤ S ≤ 0.5 in steps of 0.05.
Create a MATLAB program that

1. Selects a proper range for x by observing Figure P6.4.

2. Uses the search method to find a small interval containing the root,
use step sizes of 0.05 m.

y
x(y) = ±√R2 − y2

x
d
dy

2x(y)

FIGURE P6.4
Floating wood circular cylinder.
150 MATLAB® Essentials

3. Uses MATLAB’s fzero function to obtain a better value for the root.
4. Create a table consisting of S and d (include table headings) and a
plot of d versus S.

P6.9. The velocity of the piston described in Project 2.8 is

2πωr 2 sin( 2πωt)cos(2πωt)

V(t) = −2πωr sin( 2πωt) − (P6.9)
b 2 − r 2 sin 2 (2πωt)

Take r = 9 cm, ω = 100 revolutions per second, b = 14 cm.

Create a MATLAB program that will determine the time when the velocity
of the piston, described in that project, reaches 4000 cm/s during the time
span 0 ≤ t ≤ 0.01 s. Use 50 subdivisions in the t domain. Print the values to
the screen.

P6.10. In the time span 0 ≤ t ≤ 0.5 ms, determine the number of roots and
their values of the voltage, v(t), of the underdamped parallel RLC circuit
described in Project 2.8. The governing equation for v(t) is

  1 2   1 2 
 1   1    1  
v(t) = exp  − t   A cos  −  t + B sin  −  t  (P6.10)
 2RC    LC  2RC    LC  2RC   
    

1. Use the search method to find a small interval in which each root lies.
2. In each found interval, use MATLAB’s fzero function to find the
root value.
3. Print out to the screen, the root number and the root value.

Assume the following parameters:

R = 100 Ω, L = 10 −3 H, C = 10 −6 F, A = 6.0 V , B = −9.0 V

0 ≤ t ≤ 5.0 × 10 −4 s in steps of 1.0 × 10 −5 s.

P6.11. The current–voltage relationship of a semiconductor PN diode can be

written as follows [3]:

 qv D 
Vin − IS  e kT − 1  RD − vD = 0 (P6.11a)
 
 

where vD is the diode voltage as defined in Figure P6.5, IS is a constant (with

units of amperes), which is determined by the semiconductor doping con-
centrations and the device geometry, q = 1.6 × 10 −19 coulomb is the unit elec-
tric charge, k = 1.38 × 10 −23 J/K is the Boltzmann constant, and T is absolute
temperature (in K). We seek the value of vD that satisfies Equation P6.11a.
Roots of Algebraic and Transcendental Equations 151

iR
+

+ R vR

Vin −

− +
iD vD
−

FIGURE P6.5
Semiconductor for PN diode.

Let
 qv D 
f (vD ) = Vin − IS  e kT − 1  R − vD = 0 (P6.11b)
 
 
1. Create a MATLAB function for f (vD ) and plot for the interval
0 ≤ vD ≤ 0.8 V for 10 mV steps (80 subdivisions on the vD domain).
2. Use the search method to obtain a small interval within which the
root of Equation P6.11b lies.
3. Use MATLAB’s fzero function to obtain a more accurate value for
the root. Use the following parameters:
T = 300 K, I s = 10 −14 A, Vin = 5 V, R = 1000 Ω .

4. Print the root value to the screen.

P6.12. We wish to determine the DC transfer characteristic for the diode cir-
cuit of Figure P6.5. We will consider Vin as a parameter, where 5 ≤ Vin ≤ 12 in
steps of 1 V. We wish to find the value of vD for all values of Vin .
Write a MATLAB program that will find the roots of f (vD ) = 0, where

 qv
D

f (vD ) = Vin − IS  e kT − 1  R − vD = 0 (P6.12)
 
 

Your program should

1. Take 0.2 ≤ vD ≤ 0.8 V with 60 subdivisions on the vD domain.

2. Use the search method to find a small interval in which the root of
f (vD ) = 0 lies.
3. Use the MATLAB’s fzero function to obtain a more accurate value
for the root.
152 MATLAB® Essentials

4. Construct a table consisting of all values of Vin and the corresponding

roots of f (vD ) = 0.
5. If you did Project 6.11, confirm that the root value obtained in this
project when Vin = 5 V is the same as that obtained in Project 6.11.

References
1. Moran, M. J. and Shapiro, H. N., Fundamentals of Thermodynamics, John Wiley &
Sons, Hoboken, NJ, 2004.
2. Bober, W., Introduction to Numerical and Analytical Methods with MATLAB for
Engineers and Scientists, CRC Press, Boca Raton, FL, 2014.
3. Bober, W. and Stevens, A., Numerical and Analytical Methods with MATLAB for
Electrical Engineers, CRC Press, Boca Raton, FL, 2013.
7
System of Algebraic, Linear Equations

7.1 Introduction
In engineering we are frequently confronted with dealing with a problem
involving a set of algebraic, linear equations. In this chapter we discuss the
use of MATLAB® ’s inv and Gauss-Elimination functions for solving a
system of algebraic, linear equations.
Before the use of computers, the method of determinants was used to
obtain a solution to a system of algebraic, linear equations. Computationally,
it is only practical for a system involving just a few equations [1]. Since it is
much easier to solve such a system by MATLAB’s inv function or MATLAB’s
Gauss-Elimination function, we will skip the method of determinants.

7.2 System of Algebraic, Linear Equations

Given the set of equations
a1,1x1 + a1, 2 x2 + a1, 3 x3 +  + a1, n xn = c1

a2 ,1x1 + a2 , 2 x2 + a2 , 3 x3 +  + a2 , n xn = c2
(7.1)


an ,1x1 + an , 2 x2 + an , 3 x3 +  + an , n xn = cn

where the a’s and the c’s are known and the x’s are the unknowns.
In matrix algebra we can write the set of Equations 7.1 as follows:

AX = C

153
154 MATLAB® Essentials

where:
 x1   a1,1 a1, 2  a1, n   c1 
x  a a2 , 2  
a2 , n  c 
X=  2
, A= 
2 ,1
, C= 
2
(7.2)
       
     
 xn   an ,1 an , 2  an , n   cn 
The unknown vector X has n rows and 1 column. Similarly, the known vector
C has n rows and 1 column. The known coefficient matrix A has n rows and
n columns.
NO T E : In matrix algebra, the number of columns in A must equal the number
of rows in X.

7.2.1 MATLAB’s inv Function

The solution of the set of Equations 7.1 can be obtained by the use of
MATLAB’s inv function as follows:
X = inv(A)*C (7.3)
where:
X is the solution to the set of Equations 7.1
A is the coefficient matrix shown in Equation 7.2
C is the vector shown in Equation 7.2

The method of solving a system of linear equations by use of MATLAB’s inv

function is more computationally complicated than a method called Gauss
elimination. MATLAB’s method for solving the system of Equations 7.1 by
the Gauss-Elimination method is shown below.

7.2.2 Gauss-Elimination Method

To solve the system of Equations 7.1 by MATLAB’s Gauss-Elimination
method use
X = A\C (7.4)
where X, A, and C have the same meaning as in Equation 7.3.
Note the use of MATLAB’s backslash operator to solve for X by Gauss
Elimination.
You can obtain the size of matrix A by the command size(A). This com-
mand is useful when you run a script and you get an error message like
“Index exceeds matrix dimensions.” Entering the size() command in the
script will help you determine the problem.
System of Algebraic, Linear Equations 155

Example 7.1
The following example solves the third-order system of linear equations shown
below, and writes the results to the screen:

3 x1 + 2 x2 − x3 = 10

− x1 + 3 x2 + 2 x3 = 5
x1 − x2 − x3 = −1

% Example_7_1.m
% This program solves a third order linear system of equations by
% MATLAB's inv function and by MATLAB's Gauss elimination method.
clc; clear;
A=[3 2 -1; -1 3 2; 1 -1 -1];
C=[10 5 -1]'; % Transposing a row vector to a column vector.
% check solution:
X1=inv(A)*C % X1 is the solution using MATLAB’s inv function.
X2=A\C % X2 is the solution using MATLAB’s Gauss elimination method.
% Does X2=X1?
% Use the size() command to determine the number of rows and columns.
% Print size of A.
[A_rows A_cols] = size(A);
% Print matrix A.
A
% Print vector C.
C
% Print A*X1, does it give C2=C? \n');
C2=A*X1
end
------------------------------------------------------------------------------

Program Results:
X1 =

-2.0000
5.0000
-6.0000

X2 =

-2.0000
5.0000
-6.0000

A =

3 2 -1
-1 3 2
1 -1 -1
156 MATLAB® Essentials

C =

10
5
-1

C2 =

10.0000
5.0000
-1.0000
>>
--------------------------------------------------------------------------

We see that the use of MATLAB’s Gauss-Elimination method produces the same
answer as the use of MATLAB’s inv function.

Exercises
E7.1. Solve the following set of linear equations by MATLAB’s inv function:

a. 2x1 – x2 = 12
4x1 + 3x2 = –8

b. 2x1 + 3x2 – x3 = 20
4x1 – x2 + 3x3 = −14
x1 + 5x2 + x3 = 21

c. 4x1 + 8x2 + x3 = 8
−2x1 – 3x2 + 2x3 = 14
x1 + 3x2 + 4x3 = 30

7.3 Treatment of Large Systems of Algebraic, Linear Equations

When there is a large system of algebraic, linear equations, it may not be
obvious how to determine the appropriate matrices. There is a system-
atic approach that can be used. This is demonstrated in the following
example:
System of Algebraic, Linear Equations 157

Example 7.2
Suppose we have the following system of equations (this system of equations was
determined by the analysis of a truss consisting of 13 structural members sub-
jected to external forces). In the following set of equations, F1 , F2 , F3 ,... F13 represent
the unknown internal forces in the structural members and the right-hand side
of the equations represent external forces, P1 , P2 , P3 ,... P13 applied to the truss mem-
bers at the joints. (For more details on the process see Section 4.4 in Reference 2).
Since there are 13 unknown forces, there will be 13 equations and each equation
will have 13 a’s, most of which will be zero. The equations need to be numbered as
shown below. In the following set of equations, the coefficient matrix A is be made
up of elements ai , j where the first index is  the  equation number and the second
index is the same number, j, of the index of unknown force, Fj.

0.6 F1 + F2 = 0 (1) We see that only a1,1 and a1, 2 are non-zero.
0.8 F1 = − 62.5 (2) We see that only a2 ,1 is non-zero.
− 0.6 F1 + F5 = 0 (3) We see that only a3 ,1 and a3 , 5 are non-zero.
− 0.8 F1 − F3 = 0 (4) We see that only a4 ,1 and a4 , 3 are non-zero.
− F2 + F6 + 0.6 F4 = 0 (5) We see that only a5 , 2 , a5 , 6 and a5 , 4 are non-zero.
F3 + 0.8 F4 = 50 (6) We see that only a6 , 3 and a6 , 4 are non-zero.
− 0.6 F4 − F5 + F8 + 0.6 F10 = 0 (7) We see that only a7 , 4 , a7 , 5 , a7 , 8 and a7 ,10 are non-zero.
−0.8 F4 − F7 − 0.8 F10 = 0 (8) We see that only a8 , 4 , a8 , 7 and a8 ,10 are non-zero.
− F6 + F9 = 0 (9) We see that only a9 , 6 and a9 , 9 are non-zero.
F7 = 30 (10) We see that only a10 , 7 are non-zero.
− F9 − 0.6 F10 + F11 = 0 (11) We see that only a11, 9 , a11,10 and a11,11 are non-zero.
0.8 F10 + F12 = 40 (12) We see that only a12 ,10 and a12 ,12 are non-zero.
− F12 − 0.8 F13 = 0 (13) We see that only a13 ,12 and a13 ,13 are non-zero.

Since the coefficient matrix is sparse, it is best to initially set all ai , j = 0, and then over-
write the non-zero ai , j terms as specified in the set of equations. In matrix algebra, the
set of equations is of the form AF = P. The program follows:

% Example_7_2.m
% This program solves a system of algebraic, linear equations.
% The system of equations stems from a truss problem in statics.
% The A matrix elements are initially set to zero. The non-zero
% A elements then override the initial values. The first index in the A
% matrix elements represent the equation number. The second index in
% the A matrix element correspond to the index of the force associated with
% that matrix element.
158 MATLAB® Essentials

% There are 13 equations for 13 unknown internal forces, F(1)-F(13).

clear; clc;
ie=13; je=13;
a=zeros(13);
p=zeros(13,1);
fo=fopen('output.txt','w');
fprintf(fo,'Example 7.2 \n');
fprintf(fo,'Program solves for the internal forces of a truss');
fprintf(fo,'Forces in kN \n');
% Overwrite the non-zero elements of matrix a and matrix p.
a(1,1)=0.6; a(1,2)= 1; p(1)=0; % From Equation 1.
a(2,1)=0.8; p(2)=-62.5; % From Equation 2, etc.
a(3,1)=-0.6; a(3,5)=1; p(3)=0;
a(4,1)=-0.8; a(4,3)=-1; p(4)=0;
a(5,2)=-1.0; a(5,6)=1; a(5,4)=0.6; p(5)=0;
a(6,3)=1.0; a(6,4)=0.8; p(6)=50.0;
a(7,4)=-0.6; a(7,5)=-1.0; a(7,8)=1.0; a(7,10)=0.6; p(7)=0;
a(8,4)=-0.8; a(8,10)=-0.8; a(8,7)=-1; P(8)=0;
a(9,6)=-1; a(9,9)=1.0; p(9)=0;
a(10,7)=1.0; p(10)=30;
a(11,9)=-1.0; a(11,10)=-0.6; a(11,11)=1.0; p(11)=0;
a(12,12)=1.0; a(12,10)=0.8; p(12)=40;
a(13,12)=-1.0; a(13,13)=-0.8; p(13)=0;
fprintf(fo,' A matrix \n\n');
jindex=1:je;
fprintf(fo,' ');
for i=1:ie
fprintf(fo,'%5i',jindex(i));
end
fprintf(fo,'\n');
fprintf(fo,'--------------------------------------------');
fprintf(fo,'----------------\n');
for i=1:ie
fprintf(fo, '%4i' ,i);
for j=1:je
fprintf(fo,'%5.1f',a(i,j));
end
fprintf(fo,'\n');
end
F=a\p;
fprintf(fo,'\n\n');
fprintf(fo,'Internal forces, F(1)-F(13)& external forces p(i) \n\n');
fprintf(fo,'Member No. F(kN) Equation No p(kN) \n');
fprintf(fo,'====================================================\n');
for i=1:ie
fprintf(fo,' %3.0f %9.2f \t %3.0f \t\t %5.1f \n',...
i,F(i),i,p(i));
end
----------------------------------------------------------------------------------
System of Algebraic, Linear Equations 159

Program Results:
Example 7.2
Program solves for the internal forces of a truss. Forces in kN.
A matrix
1 2 3 4 5 6 7 8 9 10 11 12 13
-------------------------------------------------------------------
1 0.6 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
2 0.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
3 -0.6 0.0 0.0 0.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
4 -0.8 0.0 -1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
5 0.0 -1.0 0.0 0.6 0.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
6 0.0 0.0 1.0 0.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
7 0.0 0.0 0.0 -0.6 -1.0 0.0 0.0 1.0 0.0 0.6 0.0 0.0 0.0
8 0.0 0.0 0.0 -0.8 0.0 0.0 -1.0 0.0 0.0 -0.8 0.0 0.0 0.0
9 0.0 0.0 0.0 0.0 0.0 -1.0 0.0 0.0 1.0 0.0 0.0 0.0 0.0
10 0.0 0.0 0.0 0.0 0.0 0.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0
11 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 -1.0 -0.6 1.0 0.0 0.0
12 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.8 0.0 1.0 0.0
13 0.0 0.0 0.0 0.0 0.0 0.0 0.0 -1.0 0.0 0.0 0.0 -1.0 0.6

Internal forces, F(1)-F(13)& external forces p(i)

Member No. F(kN) Equation No p(kN)

==================================================
1 -78.13 1 0.0
2 46.88 2 -62.5
3 62.50 3 0.0
4 -15.63 4 0.0
5 -46.88 5 0.0
6 56.25 6 50.0
7 30.00 7 0.0
8 -43.13 8 0.0
9 56.25 9 0.0
10 -21.88 10 30.0
11 43.13 11 0.0
12 57.50 12 40.0
13 -71.88 13 0.0
-------------------------------------------------------------------------

7.4 A Resistive Circuit Problem

Another example involving a system of linear equations can be found in
problems involving resistive circuits (see Figure 7.1). For more information
on the subject see Section 4.5 in Reference 2. The goal is to solve for the
node voltages v1 , v2 , and v3 as functions of the input voltages V1 and V2 and
the input current  I1 . The equations involve the conductances, G1 , G2 ,..., G5 ,
where Gi = 1 Ri and Ri are the resistant members in the circuit.
160 MATLAB® Essentials

R3

i3

1 R2 R4
v2 3
v1 v3
2
i2 i4
R1 R5
i1 i5

+ +
V1 I1 V2
– –

FIGURE 7.1
A resistive circuit for Example 7.3.

The governing equations for this example are

(G1 + G2 + G3 ) v1 − G2 v2 − G3 v3 = V1G1 (7.5)

− G2 v1 + (G2 + G4 )v2 − G4 v3 = I1 (7.6)

− G3 v1 − G4 v2 + (G3 + G4 + G5 ) v3 = V2 G5 (7.7)

The right-hand side of Equations 7.5 through 7.7 will be represented as

vector C.
This system of equations may be solved by the use of MATLAB’s inv func-
tion or by MATLAB’s Gauss-Elimination method as shown in the following
Example 7.3.
The following script solves for the node voltages, v1, v2, and v3 for the fol-
lowing circuit values:

R1 = 2200 Ω, R2 = 10 kΩ, R3 = 6900 Ω, R4 = 9100 Ω, R5 = 3300 Ω

V1 = 12 V , V2 = 3.3 V , I1 = 2 mA

% Example_7_3.m
% Resistive Circuit Problem
% This program solves for the internal node voltages for the circuit
% shown in Figure 7.1.
% The conductances G are in units of Siemens.
% The node voltage V are in units of volts.
System of Algebraic, Linear Equations 161

% The currents I are in units of amps.

clear; clc;
A=zeros(3);
C=zeros(3,1);
V=zeros(3,1);
g1 = 1/2200; g2 = 1/10000; g3 = 1/6900; g4 = 1/9100; g5 = 1/3300;
V1 = 12; V2 = 3.3; I1 = .002;
% Overwrite the non-zero elements of matrix A and vector C.
A(1,1)=g1+g2+g3; A(1,2)=-g2; A(1,3)=-g3; C(1)=V1*g1;
A(2,1)=-g2; A(2,2)=g2+g4; A(2,3)=-g4; C(2)=I1;
A(3,1)=-g3; A(3,2)=-g4; A(3,3)=g3+g4+g5; C(3)=V2*g5;
v = A \ C ;
% print the results
fprintf('V1=%5.1f V V2=%5.1f V I1=%5.1e A \n',V1,V2,I1);
fprintf('g1=%8.5f S g2=%8.5f S g3=%8.5f S \n',g1,g2,g3);
fprintf('g4=%8.5f S g5=%8.5f S\n',g4,g5);
fprintf('\n\n');
fprintf('Node # v (volts) \n');
fprintf('----------------------------\n');
for n=1:length(C)
fprintf(' %3i %9.1f \n', n,v(n));
end
--------------------------------------------------------------------------------

Program Results:
V1= 12.0 V V2= 3.3 V I1= 2.0e-03 A
g1= 0.00045 S g2= 0.00010 S g3= 0.00014 S
g4= 0.00011 S g5= 0.00030 S

Node # v (volts)
----------------------------
1 12.6
2 20.3
3 9.0
>>
--------------------------------------------------------------------------------

7.5 Gauss Elimination

As previously discussed, the Gauss-Elimination method in solving a alge-
braic, linear set of equations is computationally more efficient than the use of
MATLAB’s inv function. In the Gauss-Elimination method, the original sys-
tem is reduced to an equivalent triangular set that can readily be solved by
back substitution (for a complete description of the method see Section 4.6
in Reference 2). The reduced equivalent set would appear like the following
set of equations:
162 MATLAB® Essentials

a11x1 + a12 x2 + a13 x3 +  + a1n xn = c1

a22 x2 + a23 x3 +  + a2 n xn = c2

a33 x3 +  + a3 n xn = c3

(7.8)
  = 

an−1,n−1xn−1 + an−1,n xn = cn−1

an ,nn xn = cn

where the tilde (~) variables are a new set of coefficients (to be determined),
and where the new coefficient matrix A  is diagonal (i.e., all of the coefficients
left of the main diagonal are zero). Then by back substitution,

xn = cn /an , n

1
xn − 1 = ( cn −1 − an −1, n xn )
an −1, n −1


etc.

7.6 Number of Solutions

Suppose a Gauss-Elimination program is carried out and the following
results are obtained:

a11x1 + a12 x2 + a13 x3 +  + a1n xn = c1

a22 x2 + a23 x3 +  + a2 n xn = c2

a33 x3 +  + a3 n xn = c3

  = 

arr xr +  + arn xn = cr (7.9)

0 = c r +1

0 = cr + 2


0 = cn
System of Algebraic, Linear Equations 163

where r < n and a11 , a22 ,...., arr are not zero. There are two possible cases:
1. No solution exists if any one of the cr+1 through cn is not zero.
2. Infinitely many solutions exits if cr+1 through cn are all zero.
NOTE: If you attempt to solve a system of algebraic, linear equations using
MATLAB’s Gauss-Elimination method, and MATLAB arrives at a set of equa-
tions as shown in Equation 7.9, MATLAB will give you a warning as follows:
“Warning: Matrix is close to singular or badly scaled. Results may be
inaccurate.”
If, in the above set, r = n and a11 , a22 ,...., ann are not zero, then the system
would appear as follows:
a11x1 + a12 x2 + a13 x3 +  + a1n xn = c1

a22 x2 + a23 x3 +  + a2 n xn = c2

a33 x3 +  + a3 n xn = c3 (7.10)
  = 

ann xn = cn
For this case there is only one solution.

Exercise
E7.2. Use MATLAB’s inv function to solve the following set of linear
equations. Note the warning that MATLAB gives with its solution.
2 x1 + 3 x2 + x3 − x4 = 1

5 x1 − 2 x2 + 5 x3 − 4 x4 = 5

x1 − 2 x2 + 3 x3 − 3 x4 = 3

3 x1 + 8 x2 − x3 + x4 = − 1

--------------------------------------------------------

REVIEW 7.1

1. Given a set of algebraic, linear equations in the form AX = C,

where A is the coefficient matrix and X and C are column vec-
tors, what are the two ways for solving for X in MATLAB?
2. Given a large system of algebraic, linear equations of the form
AX = C, describe the recommended approach to solving the
system of linear equations.
164 MATLAB® Essentials

Projects
P7.1. The following set of linear equations came from a problem in Statics.
Use the method described in Example 7.2 to solve the following set of linear
equations. The Fi value represents the internal force in structural member
i in kN.
Take Ax = −9 kN, Ay = 7 kN, cos ϑ = 0.6, sin ϑ = 0.8

cos ϑ F1 + F2 = − Ax (1)

sin ϑ F1 = − Ay ( 2)

− cos ϑ F1 + cos ϑ F5 + F6 =0 ( 3)

− sin ϑ F1 − F3 − sin ϑF5 =0 ( 4)

− F2 + F4 =0 (5)

F3 =2 (6)
− F6 + cos ϑ F9 + F10 =0 (7)

− F7 − sin ϑ F9 =0 (8 )

− F4 − cos ϑ F5 + F8 =0 (9)

sin ϑ F5 + F7 =4 (10)

− F10 + F14 =0 (11)

F11 =0 (12)

− F8 − cos ϑ F9 + F12 + cos ϑ F13 = 0 (13)

sin ϑ F9 + F11 + sin ϑ F13 =6 (14)

− cos ϑ F13 − F14 + F18 =0 (15)

− sin ϑ F13 − F15 =0 (16)

− F12 + F16 + cos ϑ F17 =0 (17)

F15 + sin ϑ F17 =4 (18)

− cos ϑ F17 − F18 + cos ϑ F21 = −9 (19)

− sin ϑ F17 − F19 − sin ϑ F21 =0 (20)

− F16 + F20 =0 (21)

System of Algebraic, Linear Equations 165

P7.2. The following set of linear equations came from a problem in Statics.
Use the method described in Example 7.2 to solve the following set of linear
equations. The Fi value represents the internal force in structural member
i in kN.

0.8 F2 = 30 (1)

F1 + 0.6 F2 = −30 (2)

F3 + 0.8 F5 =0 (3)

− F1 + F4 + 0.6 F5 =0 (4)

−0.8 F2 − F3 =5 (5)

−0.6 F2 + F6 =0 (6)

F7 + 0.8 F9 − F11 =0 (7)

− F4 + 0.6 F9 + F11 =0 (8)

−0.8 F5 − F7 =5 (9)

−0.6 F5 − F6 + F8 =0 (10)

0.8 F9 + F10 = −5 (11)

− F8 − 0.6 F9 =0 (12)

− F10 + 0.8 F13 + F14 = 0 (13)

− F11 − 0.6 F13 =0 (14)

F12 + 0.8 F13 − F16 = 0 (15)

0.6 F13 + F15 = 15 (16)

F16 + 0.8 F17 − F18 = 0 (17)

0.6 F17 = 15 (18)

P7.3. Figure P7.1 shows a resistive circuit known as a ladder network.

Using Ohm’s law and Kirchoff’s current law, we can determine a set of
linear equations for the voltages v1 , v2 , v3 , and v4 (For a complete derivation
of the set of equations, see Project P4.5 in Reference 2). The equations are
written in terms of the conductances, Gnm , instead of the resistances, Rnm ,
where Gnm = 1/Rnm.
166 MATLAB® Essentials

R11 R21 R31 R41

v1 v2 v3 v4

+ R12 R22 R32 R42

V1
−

FIGURE P7.1
Fourth-order ladder network.

The governing equations for this system are

−(G11 + G21 + G12 ) v1 + G21 v2 = − G11Vref (1)

G21 v1 − (G21 + G31 + G22 ) v2 + G31 v3 = 0 ( 2)

G31 v2 − (G31 + G41 + G32 ) v3 + G41 v4 = 0 (3)

G41 v3 − ( G41 + G42 ) v4 = 0 ( 4)

Create a MATLAB program to solve for all circuit voltages. Take V1 = 5V and
the following resistor values:

R11 = 2200 Ω, R12 = 2200 Ω

R21 = 1200 Ω, R22 = 6800 Ω

R31 = 3900 Ω, R32 = 2200 Ω

R41 = 3300 Ω, R42 = 5700 Ω

P7.4. Suppose a manufacturer wishes to purchase a piece of equipment that

costs $40,000. He plans to borrow the money from a bank and pay off the
loan in 10 years in 120 equal payments. The annual interest rate is 6%. Each
monthly payment, M, consists of two parts: one part goes toward paying
off the principal, P, and the other part is the interest charged based on the
unpaid balance of the loan. He wishes to determine what his monthly pay-
ment will be. This problem can be solved by a system of linear equations. Let
xj is the amount in the jth payment that goes toward paying off the principal.
Then the equation describing the jth payment is

 n= j −1

jth payment = M = x j +  P −


∑
n=1
xn  I


(P7.4a)

where:
M is the monthly payment
P is the amount borrowed
I is the monthly interest rate = annual interest rate/12
System of Algebraic, Linear Equations 167

The total number of unknowns is 121 (120x values and M).

Applying Equation P7.4a to each month gives 120 equations. One additional
equation is
n=120
P= ∑x
n=1
n (P7.4b)

Develop a MATLAB program that will

1. Ask the user to enter from the keyboard the amount of the loan (P),
the annual interest rate, I, and the time period, Y, in years.
2. Set up the system of linear equations, using An,m as the coefficient
matrix of the system of linear equations. The n represents the equa-
tion number and m represents the coefficient of xm in that equation.
Set x121 = M.
3. Solve the system of linear equations in MATLAB.
4. Print out a table consisting of four columns. The first column should
be the month number, the second column the monthly payment, the
third column the amount of the monthly payment that goes toward
paying off the principal, and the fourth column the interest payment
for that month.

References
1. Kreyszig, E., Advanced Engineering Mathematics, 8th ed., Wiley, New York, 1999.
2. Bober, W., Introduction to Numerical and Analytical Methods with MATLAB for
Engineers and Scientists, CRC Press, Boca Raton, FL, 2014.
8
Curve Fitting

8.1 Introduction
There are many occasions in engineering that require an experiment to
determine the behavior of a particular phenomenon. The experiment may
produce a set of data points that represents a relationship between the
variables involved in the phenomenon. We may then wish to express this
relationship analytically for further analysis. A mathematical expression
that describes the data is called an approximating function. There are two
approaches to determining an approximating function:

1. The approximating function graphs as a smooth curve. The approxi-

mating curve will generally not pass through all the data points, but
we seek to minimize the resulting error in order to get the best fit.
A plot of the data on linear, semilog or log-log graphic paper can
often suggest an appropriate form for the approximating function.
2. The approximating function passes through all data points (as
described in Section 8.4). However, if there is some scatter in the data
points, this approximating function may not be satisfactory.

8.2  MATLAB’s Curve-Fitting Functions

MATLAB® calls curve fitting with a polynomial by the name Polynomial
Regression. The function polyfit(x, y, m) returns a vector of (m  +  1) coef-
ficients, ai  , that represent the best-fit polynomial of degree m for the (xi  , yi)
set of n data points. The coefficient order corresponds to decreasing powers
of x, that is,

yc = a1x m + a2 x m−1 + a3 x m−2 +  am x + am+1 (8.1)

169
170 MATLAB® Essentials

To obtain yc at the data points ( x1 , x2 , x2 ..., xn ) use the MATLAB function

polyval(a, x), where x = [x1 x2 .... xn ]. MATLAB’s polyval(a, x) function
returns a vector of length n giving yc,i where

yc ,i = a1xim + a2 xi m−1 + a3 xi m−2 +  am xi + am+1 (8.2)

MATLAB measures the precision of the fit with a function named mse, which
is defined as follows:
1 n
mse = ∑ ( y i − y c , i )2 (8.3)
n i =1

where yi are the data point y values and yc,i are the approximating curve y
values at the data points xi , and n is the number of data points.
You may also use polyval(a, x), where x is any set of x values, preferably
between x1 and xn.

Example 8.1
% Example_8_1.m
% This program determines the best fit polynomial approximating
% functions of orders 2 thru 5 for the data set listed below.
% MATLAB’s polyfit and polyval functions are used in the program.
% The sprintf command is used in this program to write formatted data
% in the plot title. The sprintf command is the same as the
% fprintf command except that it returns the data in a MATLAB
% string rather than writing to the screen or to a file.
clear; clc;
% Enter data.
x=-10:2:10;
y=[-980 -620 -70 80 100 90 0 -80 -90 10 220];
mse=zeros(4);
% Determine best fit 2-5 degree polynomials to fit the data.
for m=2:5
fprintf('m= %i \n',m);
coef=zeros(m+1);
coef=polyfit(x,y,m);
% Approximating function at x
yc=polyval(coef,x);
% yc is a vector since x is a vector.
mse(m)=sum((y-yc).^2)/length(x);
% y-yc=[y(1)-yc(1) y(2)-yc(2) ..... y(n)-yc(n)].
fprintf(' x y yc \n');
fprintf('------------------------------\n');
for i=1:length(x)
fprintf('%5.1f %5.1f %8.2f \n',x(i),y(i),yc(i));
Curve Fitting 171

end
fprintf('\n\n');
x2=-10:0.5:10;
% Approximating function at x2
yc2=polyval(coef,x2);
subplot(2,2,m-1),plot(x2,yc2,x,y,'o'), xlabel('x'), ylabel('y'),
grid, axis([ -10 10 -1500 500]),
legend('approx curve','data points');
title(sprintf('Degree %d polynomial fit',m));
end
fprintf(' m mse \n')
fprintf('------------------\n');
for m=2:5
fprintf('%d %8.1f \n',m,mse(m))
end
-------------------------------------------------------------------------------

Program Results:
Output for m = 5 is only displayed here.
m = 5
x y yc
------------------------------
-10.0 -980.0 -999.09
-8.0 -620.0 -545.31
-6.0 -70.0 -156.76
-4.0 80.0 78.39
-2.0 100.0 148.18
0.0 90.0 93.80
2.0 0.0 -13.50
4.0 -80.0 -95.45
6.0 -90.0 -89.91
8.0 10.0 26.15
10.0 220.0 213.50

m mse
------------------
2 32842.4
3 2660.0
4 2342.1
5 1502.9
>>

As expected, the mse decreases as the order of the fitted polynomial is increased.
See Figure 8.1.
--------------------------------------------------------------------------
172 MATLAB® Essentials

Degree 2 polynomial fit Degree 3 polynomial fit

500 500

0 0

−500 −500
y

y
−1000 approx curve −1000 approx curve
data points data points
−1500 −1500
−10 −5 0 5 10 −10 −5 0 5 10
x x
Degree 4 polynomial fit Degree 5 polynomial fit
500 500

0 0

−500 −500
y

−1000 approx curve −1000 approx curve

data points data points
−1500 −1500
−10 −5 0 5 10 −10 −5 0 5 10
x x

FIGURE 8.1
Approximating curves with data points.

Example 8.2
This example involves the sampling with respect to time of an audio signal that is
converted to a voltage by a microphone and an amplifier. The data are in volts (V)
versus time in microseconds (µs).

% Example_8_2.m
% This program determines the best fit polynomial approximating
% functions of orders 2 thru 5 for the data set listed below.
% MATLAB's polyfit and polyval functions are used in the program.
% The data involves the sampling in time, t, of an audio signal
% converted to a voltage, V, by a microphone and an amplifier.
% V is in volts and t is in microseconds.
clear; clc;
% Define original data points for V(t)
t = [ 0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 ];
V = [.7 .9 .9 .7 .3 0 -.3 -.7 -.7 -.3 0 .3 .7 .7 .3 ];
% Determine best fit 2-5 degree polynomials to fit the data.
for m=2:5
fprintf('m= %i \n',m);
coef=zeros(m+1);
coef=polyfit(t,V,m);
% Approximating function at t
Vc=polyval(coef,t);
mse(m)=sum((V-Vc).^2)/length(t);
Curve Fitting 173

fprintf(' t V Vc \n');
fprintf('(micro-sec) (volt) (volt) \n');
fprintf('------------------------------------------\n');
for i=1:length(t)
fprintf(' %5.1f %8.4f %8.4f \n',...
t(i),V(i),Vc(i));
end
fprintf('\n\n');
t2 = 0:60;
% Approximating function at t2
Vc2=polyval(coef,t2);
subplot(2,2,m-1),plot(t2,Vc2,t,V,'o'), xlabel('t(\mus)'),
ylabel('V(V)'), grid, legend('approx curve','data points'),
title(sprintf('Degree %d polynomial fit',m));
end
fprintf('m mse \n')
fprintf('---------------------\n');
for m=2:5
fprintf('%d %8.5f \n',m,mse(m))
end
---------------------------------------------------------------------------------

Program Results:
Only results for m = 5 are displayed.
m = 5
t V Vc
(micro-sec) (volt) (volt)
------------------------------------------
0.0 0.7000 0.6791
4.0 0.9000 0.9377
8.0 0.9000 0.9102
12.0 0.7000 0.6748
16.0 0.3000 0.3211
20.0 0.0000 -0.0588
24.0 -0.3000 -0.3790
28.0 -0.7000 -0.5691
32.0 -0.7000 -0.5833
36.0 -0.3000 -0.4088
40.0 0.0000 -0.0754
44.0 0.3000 0.3361
48.0 0.7000 0.6845
52.0 0.7000 0.7592
56.0 0.3000 0.2717

m mse
---------------------
2 0.10789
3 0.10583
4 0.00803
5 0.00446
>>

See Figure 8.2.

----------------------------------------------------------------------
174 MATLAB® Essentials

Degree 2 polynomial fit Degree 3 polynomial fit

2 2
approx curve approx curve
data points data points
1 1

V(V)
V(V)

0 0

−1 −1
0 20 40 60 0 20 40 60
t(µs) t(µs)
Degree 4 polynomial fit Degree 5 polynomial fit
2 1
approx curve
data points
1 0
V(V)

V(V)

0 −1
approx curve
data points
−1 −2
0 20 40 60 0 20 40 60
t(µs) t(µs)

FIGURE 8.2
Approximating function and data points for audio signal versus time.

8.3 Curve Fitting with the Exponential Function

Many physical systems can be modeled as exponential functions. If your
experimental data appears to fall into this category, it can be fitted with a
function of the form
−β1 x
yc = β2 e (8.4)

where β1 and β2 are real constants.

Let us assume that a set of n measured data points ( x1 , y1 ), ( x2 , y2 ),  , ( xn , yn )
exists. Then, let zi = ln yi and zc = ln yc = lnβ2 − β1x , and also let a1 = −β1 and
a2 = lnβ2 . Then taking the log of both sides of Equation 8.4 and making the
above substitutions, we obtain the linear equation

zc = a1 x + a2 (8.5)
For the data points ( x1 , y1 ), ( x2 , y2 ),  , ( xn , yn ), the new set of data points
becomes ( x1 , z1 ), ( x2 , z2 ),  , ( xn , zn ).
Curve Fitting 175

We can then use MATLAB’s polyfit function to determine a1 and a2 .

a
Then, β2 = e 2 and β1 = −a1.

Example 8.3
Suppose we took an oscilloscope picture of the position, y, of the mass in a mass-
spring-dashpot system as shown in Figure 8.3 and measured the (t, ye) positions of
the envelope. Table 8.1 gives the measured position, ye , as a function of time, t. The
governing equation of the envelope is

c
− t
y e = y0 e 2m
(8.6)

where:
c is the damping constant
m is the mass
ye is the ordinate position of the envelope of the plot of the mass displacement from
the equilibrium position

Comparing Equation 8.6 with Equation 8.4 we see that

y vs. t
2
(t2, y2)
1.5 (t3, y3)
(t4, y4)
1

0.5
y(m)

−0.5

−1

−1.5

−2
0 2 4 6 8 10 12 14 16 18 20
t(s)

FIGURE 8.3
Position of a mass in a mass-spring-dashpot system with its envelope.
176 MATLAB® Essentials

TABLE 8.1
ye versus t
t (s) ye (cm)

0.00 2.00
2.17 1.58
4.31 1.32
6.72 1.04
8.96 0.89
11.21 0.68
13.28 0.55
15.52 0.41
17.93 0.36

c
β2 = y0 and β1 = with t replacing x
2m

Thus, the damping factor, c, for the system is given by

c = 2 m β1

Table 8.1 gives the measured values ye versus t.

As a check that the equation of the envelope is truly an exponential we can plot the
data on semilog paper. It should plot as a straight line on semilog paper. This is done
in the following program.
The mass, m, in the system is 25 kg.

The Program Follows:

% Example_8_3.m
% This program determines the best exponential fit for the envelope
% of the motion of a mass-spring-dashpot system.
clear; clc;
t=[0.00 2.17 4.31 6.72 8.96 11.21 13.28 15.52 17.93];
ye=[2.00 1.58 1.32 1.04 0.89 0.68 0.55 0.41 0.36];
z=[log(2.0) log(1.58) log(1.32) log(1.04) log(0.89) log(0.68) ...
log(0.55) log(0.41) log(0.36)];
% If the relationship is exponential, ye should plot as a
% straight line on semi-log paper.
semilogy(t,ye,'x'), xlabel('t(s)'), ylabel('log(ye)'), grid,
title('log(ye) vs. t');
figure;
a=polyfit(t,z,1);
zc=a(1)*t+a(2);
Curve Fitting 177

fprintf('a(1)=%7.3f a(2)=%6.3f \n',a(1),a(2));

beta(1)=-a(1);
beta(2)=exp(a(2));
fprintf('beta(1)=%7.3f, beta(2)=%6.3f \n',beta(1),beta(2));
for i=1:9
yc(i)=beta(2)*exp(-beta(1)*t(i));
end
plot(t,yc,t,ye,'o'), xlabel('t(s)'), ylabel('yc(cm)'),
title('yc vs. t'), grid, legend('yc','ye');
m=25.0;
c=2*m*beta(1);
fprintf('m=%5.1f(kg), The damping constant=%7.4f(N-s/cm) \n',m,c);
-----------------------------------------------------------------------------

Program Results:
a(1)= -0.097, a(2)= 0.695
beta(1)= 0.097, beta(2)= 2.005
m= 25.0(kg), The damping constant= 4.8718(N-s/cm)
>>

See Figures 8.4 and 8.5.

-----------------------------------------------------------------------------

log(ye) vs. t
2
1.8
1.6
1.4

1.2

1
log(ye)

0.8

0.6

0.4

0 2 4 6 8 10 12 14 16 18
t(s)

FIGURE 8.4
Plot of log(ye) versus t.
178 MATLAB® Essentials

yc vs. t
2.2
yc
2 ye

1.8

1.6

1.4
yc(cm)

1.2

0.8

0.6

0.4

0.2
0 2 4 6 8 10 12 14 16 18
t(s)

FIGURE 8.5
Plot of yc and ye versus t.

REVIEW 8.1

1. Suppose an experiment produced a set of data and we wished

to  create an approximating curve, yc , that is a polynomial
expression that best fits the data. What is the name of the
MATLAB function that will do this?
2. After executing MATLAB’s polyfit function you may wish
to obtain values on the approximating curve, yc at positions
(x1 , x2 , x3 ,..., xn). If so, what MATLAB function would you use?

8.4 Cubic Splines

Suppose that we are given a set of n data points and that we select an mth
degree polynomial-approximating curve that produces curve values that are
not allowed. For example, suppose it is known that a particular property
represented by the approximating curve (such as absolute pressure or abso-
lute temperature) must be positive and the approximating function produces
values that are negative. In this case the approximating curve would not be
satisfactory. The method of cubic splines eliminates this problem.
Curve Fitting 179

Given a set of (n + 1) data points (xi , yi), i = 1, 2,…, (n + 1), the method of
cubic splines develops a set of n cubic functions, such that y(x) is represented
by a different cubic in each of the n intervals and the set of cubics passes
through all (n + 1) data points.

8.4.1 MATLAB’s Cubic Spline Curve-Fitting Function

The syntax for MATLAB’s cubic spline function is

yy = spline(xi , yi , xx)

where (xi , yi) is a given set of data points and yy is the value of y at xx.
The spline function determines the four cubic coefficients for each sec-
tion in the given data and will evaluate yy by the cubic-spline method. The
same result can be obtained by using MATLAB’s interp1 function and
specifying the use of the spline method of interpolation. The syntax for
interpolating by the spline method is

yi = interp1( x , y , xi , 'spline')

Example 8.4
The following example involves a measured increase in air pressure at distances
from a blast. The data specifies the pressure above normal atmospheric pressure and
is designated as overpressure. The program demonstrates the use of the MATLAB’s
spline function as well as MATLAB’s interp1 function with the spline option
to determine the pressure at distances not in the data. We see that the two meth-
ods produce the same results. The program follows:

% Example_8_4.m
% This program uses both MATLAB's spline function and MATLAB's
% interp1 function with the cubic spline option to determine the
% over-pressure resulting from a blast. The program calculates the
% over-pressure at locations between data points. The over-pressure
% is in kPa and the distance from the blast in km.
clear; clc;
dist=0.52:0.3:4.12;
press=[165.5 96.5 69.0 52.4 37.2 27.6 21.4 17.2 13.8 11.7 ...
10.3 9.0 7.2];
d=0.52:0.1:4.12;
p1=spline(dist,press,d);
p2=interp1(dist,press,d,'spline');
fo=fopen('output.txt','w');
fprintf(fo,'Peak over-pressure vs. distance from the blast, \n');
fprintf(fo,'Cubic spline fit \n');
fprintf(fo,'dist(km) over-press(kPa) over-press(kPa) \n');
fprintf(fo,' by spline function by interp1 \n');
fprintf(fo,'------------------------------------------------\n');
180 MATLAB® Essentials

for n=1:length(d)
fprintf(fo,'%5.2f %10.2f %10.2f \n',d(n),p1(n),p2(n));
end
plot(d,p1,d,p2,'o'), xlabel('km from ground zero'),
ylabel('overpressure(kPa)'), grid,
title('Peak over-pressure vs. distance from blast')
fclose(fo);
----------------------------------------------------------------------

Program Results:
Peak over-pressure vs. distance from blast, cubic spline fit
dist(km) over-press(kPa) over-press(kPa)
by spline function by interp1
------------------------------------------------
0.52 165.50 165.50
0.62 135.72 135.72
0.72 113.15 113.15
0.82 96.50 96.50
0.92 84.46 84.46
1.02 75.72 75.72
1.12 69.00 69.00
1.22 63.15 63.15
1.32 57.71 57.71
1.42 52.40 52.40
1.52 47.02 47.02
. . .
. . .
3.12 12.28 12.28
3.22 11.70 11.70
3.32 11.19 11.19
3.42 10.73 10.73
3.52 10.30 10.30
3.62 9.88 9.88
3.72 9.46 9.46
3.82 9.00 9.00
3.92 8.49 8.49
4.02 7.89 7.89
4.12 7.20 7.20

See Figure 8.6.

----------------------------------------------------------------------

Example 8.5
Example 8.2 can also be used as an example for the use of MATLAB’s interp1 func-
tion with the spline option. That example involved the sampling with respect to time
of an audio signal that is converted to a voltage by a microphone and an amplifier.
The data is in volts (V) versus time in microseconds (µs). Variable names in this pro-
gram differ from those in Example 8.2. The program follows:
Curve Fitting 181

Peak over-pressure vs. distance from blast

180
spline curve
160 data

140

120
overpressure(kPa)

100

80

60

40

20

0
0.5 1 1.5 2 2.5 3 3.5 4 4.5
km from ground zero

FIGURE 8.6
Blast overpressure versus distance from blast.

% Example_8_5.m
% This program uses interpolation by cubic splines to upsample
% an audio signal V(t) vs. time in microsec.
clear; clc;
% Define original data points for V(t) (time is in microsec)
orig_t = [ 0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 ];
orig_V = [.7 .9 .9 .7 .3 0 -.3 -.7 -.7 -.3 0 .3 .7 .7 .3 ];
% Define upsampled time points
upsample_t = 0:60;
% Calculate interpolated data points using cubic splines
upsample_V = interp1(orig_t,orig_V,upsample_t,'spline');
% Print output to screen
fprintf('Upsampling via cubic spline fit \n');
fprintf('time (microsec) upsample_V \n');
for i=1:length(upsample_t)
fprintf('%8.2f %10.3f \n',upsample_t(i),upsample_V(i));
end
plot(orig_t,orig_V,'o',upsample_t,upsample_V);
xlabel('t(microsec)'); ylabel('V(volt)'); grid;
title('Upsampling with Cubic Spline Interpolation');
legend('original','upsampled');
----------------------------------------------------------------------

Program Results:
Only the plot (Figure 8.7) is shown here.
182 MATLAB® Essentials

Upsampling with Cubic Spline Interpolation

1
original
0.8 upsampled

0.6

0.4

0.2
V(volt)

−0.2

−0.4

−0.6

−0.8
0 10 20 30 40 50 60
t(microsec)

FIGURE 8.7
Upsampling of audio signal.

Projects
P8.1. A formula describing the fluid level, h, in a tank as a function of time as
the fluid discharges through a small circular orifice (see Figure P8.1) is

2
Cd Ao C A 
h = ho − 2 g ho × t +  d o  × 2 g t2 (P8.1)
AT  2 AT 

where:
Cd is the discharge coefficient
ho is the fluid level in the tank at time, t = 0
Ao is the area of the orifice
AT is the cross-sectional area of the tank

An experiment consisting of a cylindrical tank with a small circular ori-

fice was used to determine Cd for that particular orifice and cylinder. The
tank walls were transparent and a ruler was pasted to the wall allowing
for the determination of the fluid level in the tank. The procedure was to
fill the tank with water while the orifice was plugged. The plug was then
removed and the water was allowed to flow through the orifice. The water
level in the tank, hexp in m, was recorded as a function of time, t, in s. The
experimental data is shown below:
Curve Fitting 183

Water level
h(t)
Ruler

FIGURE P8.1
Water in a tank discharging through an orifice.

hexp = [0.288 0.258 0.234 0.215 0.196 0.178 0.160 0.142 0.125 0.110 …

0.095 0.080 0.065 0.053 0.041 0.031 0.022 0.013 0.006 0.0022 …

0.000]

t = [0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 …

180 190 200]

Diameter of the orifice, do  =  0.0055  m and the diameter of the tank,
DT = 0.146 m. The free surface elevation, ho , at t = 0 is 0.288 m. The gravita-
tional constant, g = 9.81 m/s2.
Use the mse as defined by Equation P8.2 to determine the value for Cd that
best fits the data. Vary Cd from 0.3 to 0.9 in steps of 0.01 and evaluate the mse
for each Cd selected, where
N

∑[ h(t ) − h
1
mse = i exp (ti )]2 (P8.2)
N i =1

where:
N is the number of data points
h(ti) is the water level in the tank at ti as determined by Equation P8.1
hexp(ti) is the water level in the tank at ti as determined by experiment

For the Cd with the lowest mse, create a plot of h versus t (solid line) and
superimpose hexp versus t as little x’s onto the plot of h versus t. Also print out
the value of Cd that gives the lowest mse.

P8.2. This project involves determining the best-fit polynomial approximat-

ing curve to the (H vs. Q) data obtained experimentally. The experimental
(H vs. Q) data are shown in Table P8.1.
184 MATLAB® Essentials

TABLE P8.1
Experimental H versus Q Data
Q H Q H
(m3/h) (m) (m3/h) (m)

3.3 43.3 61.6 40.8

6.9 43.4 68.5 39.6
13.7 43.6 75.3 38.7
20.5 43.6 82.2 37.2
27.4 43.3 89 36.3
34.2 43.0 95.8 34.4
41.1 42.7 102.7 32.6

Try degree polynomials of two through four to determine which

degree polynomial will give the smallest mse. Use MATLAB’s function
polyfit that returns the coefficients for each of the three polynomials.
Then use the following MATLAB’s function polyval to create for each
polynomial:

1. A table containing Q, H, and Hc , where Hc are values from the

approximating curve for H versus Q.
2. A plot containing both Hc versus Q (solid line) and H versus Q (small
circles).

P8.3. Figure P8.2 shows a resistor-diode circuit using a type 1N914 silicon
diode (D1) and a 10  kΩ resistor (R1). Table P8.2 shows a list of laboratory
measurements of v2 for various applied voltage levels of v1 at room tempera-
ture (300 K). An equation that describes the behavior of the diode is given in
Equation P8.3.

1. Use the technique described in Section 8.3 to find the best-fit values
for IS and vT .

R1
+
v1
+
−
D1 v2
−

FIGURE P8.2
Diode-resistor circuit for laboratory measurement of diode I-v curve.
Curve Fitting 185

TABLE P8.2
Laboratory Measurements of
Resistor-Diode Circuit
v1 (volts) v2 (volts)

0.333 0.317
0.393 0.356
0.819 0.464
1.067 0.487
1.289 0.501
1.656 0.518
1.808 0.522
2.442 0.541
3.949 0.566
4.971 0.579
6.005 0.588
6.933 0.595
7.934 0.602
9.014 0.607
10.040 0.613
11.009 0.619
15.045 0.634
19.865 0.647
24.64 0.657
29.79 0.666

v2 − v1 v 
= IS exp  2  (P8.3)
R1  vT 
Plot both the lab data and your fitted curve on the same axes.
v1 − v2
NO T E : The diode current, iD = .
R1
2. vT = kT q is known as the thermal voltage (where k is the Boltzmann
constant and q is the unit electric charge). For your best-fit value for
vT , what is the corresponding temperature value T (in Kelvin)? Take
k = 1.38e–23, q = 1.6e–19.
9
Numerical Integration

9.1 Introduction
In this chapter, we cover Simpson’s rule for approximating the value of defi-
nite integrals as well as MATLAB®’s integral function. Understanding the
concept in Simpson’s rule will help you implement MATLAB’s integral
function for evaluating definite integrals. A discussion of MATLAB’s
integral2 function is also included. Finally, examples demonstrating the
usage of these three methods are given.

9.2 Numerical Integration and Simpson’s Rule

We can evaluate a definite integral of a single variable using Simpson’s rule.
B
In applying Simpson’s rule for evaluating ∫ A f ( x)dx, the first thing one does is sub-
dividing the x domain into N intervals, where N is an even number, giving
x1 , x2 ,..., xN , xN +1. We then determine the functional values at the xn positions
giving f1 , f2 ,....... fN , fN +1. We then connect three points on the curve f ( x) with
second-degree polynomials (parabolas) and sum the areas under the parabo-
las to obtain the approximate area under the curve (see Figures 9.1 and 9.2).
For a complete derivation see Article 6.3 in Reference 1. The final formula for
the integral by Simpson’s rule is

x N +1
∆x
I=
∫
x1
f ( x)dx =
3
[ f1 + 4 f2 + 2 f3 + 4 f4 + 2 f5 +  + 4 fN + fN +1 ] (9.1)

187
188 MATLAB® Essentials

A2 strips

f (x)
f1 . . . fi − 1 fi fi + 1
...
fN + 1

... Δx Δx
...

x
x1 . . . xi − 1 xi xi + 1 . . . xN + 1

FIGURE 9.1
Area under two adjacent strips.

f2 f3
f1
f
fn − 1
fn
A1 A2 A3 fn + 1

An/2

x
x1 x2 x3 xn + 1

FIGURE 9.2
Summing all two-strip areas in Simpson’s rule.

Example 9.1
Solve by Simpson’s rule:

10

∫
I = ( x 3 + 3.2 x 2 − 3.4 x + 20.2)dx
0
(9.2)

% Example_9_1.m
% This program calculates an integral given in Equation 9.2 by
% Simpson's Rule.
% The integrand is: x^3+3.2*x^2-3.4*x+20.2
% The limits of integration are from 0-10.
clear; clc;
A=0; B=10;
N=100; dx=(B-A)/N;
% Compute values of x and f at each point:
% An arithmetic expression involving vector x produces a vector f.
% Need to use element by element multiplication.
x = A:dx:B;
Numerical Integration 189

f = x.^3+3.2*x.^2-3.4*x+20.2;
% Use two separate loops to sum up the even and odd terms
% of Simpson's Rule. Also, exclude endpoints in the loop, that is
% f(1) and f(N+1).
sum_even=0.0;
for i=2:2:N
sum_even=sum_even+f(i);
end
sum_odd=0.0;
for i=3:2:N-1
sum_odd=sum_odd+f(i);
end
% Calculate integral as per Equation 9.2.
I = dx/3 * (f(1) + 4*sum_even + 2*sum_odd + f(N+1));
% Display results
fprintf('Integrand: x^3+3.2*x^2-3.4*x+20.2 \n');
fprintf('Integration limits: %.1f to %.1f \n',A,B);
fprintf('Simpson rule solution, I = %9.4f \n',I);
% Compare with analytic solution.
% Analytic solution:
I2 = 0.25*10^4+3.2/3*10^3-3.4/2*10^2+20.2*10;
fprintf('Analytic solution, I2 = %9.4f \n',I2);
-----------------------------------------------------------------------------

Program Results:
Integrand: x^3+3.2*x^2-3.4*x+20.2
Integration limits: 0.0 to 10.0
Simpson rule solution, I = 3598.6667
Analytical solution, I2 = 3598.6667
>>
-------------------------------------------------------------------------------

We see that solving the integral of Example 6.1 by Simpson’s rule gives the
same answer as the analytical method up to four decimal places.

Exercises

E9.1. Evaluate the following definite integrals by Simpson’s rule:

3
dx
1. I =
∫ 5e
0
3x
+ 2 e −3 x
π/2
sin x dx
2. I =
∫
− π/2
1 − 4 sin 2 x
π

∫
3. I = (sinh x − cos x) dx
0
190 MATLAB® Essentials

9.3 Improper Integrals

An integral is improper if the integrand approaches infinity at some point
within the limits of integration, including the end points. In many cases, the
integration will still result in a finite solution. An example of an improper
integral follows:
1
log(1 + x)
I=
∫
0
x
dx (9.3)

The above integral is improper since both the numerator and denominator
are zero at the lower limit (x = 0). The exact value of I can be obtained by resi-
due theory in complex variables and in this case the integral, I, evaluates to
(π2/12) = 0.822467. MATLAB’s integral function, which is discussed next,
is able to evaluate some improper integrals, but may give you a warning that
the answer may be inaccurate.

9.4 MATLAB’s integral Function

The MATLAB function for evaluating integrals is the function integral. It
is a replacement for MALAB’s quad function. A description of the function
can be obtained by typing help integral in the Command Window. The
syntax for MATLAB’s integral function is

Q = integral(FUN,A,B)

where FUN is a function handle for the self-written function that describes
the integrand. A and B are the limits of integration and Q is the integral
result. The integral function approximates the integral using global adap-
tive quadrature and default error tolerances. Although the integral func-
tion can treat integrand variables that are complex, we are only interested for
cases where the integrand involves only scalar value functions. The function
will accept limits of integration A or B as inf or –inf. The function Y=FUN(X)
should accept a vector argument X and return a vector result Y. The inte-
grand is evaluated at each element of X. The function FUN can be either as a
separate .m file or as an anonymous function. You may use the latter method
if the integrand can be expressed in a single line.
The integral function is also able to evaluate certain improper integrals.
It does this by selecting limits of integration that are very close to the singular
points, but not on them, thus, removing the singularity.
Numerical Integration 191

Example 9.2
We will now repeat Example 9.1, but this time we will use MATLAB’s integral
function to do the integration. The integral I in Example 9.1 is

10

∫
I = ( x 3 + 3.2 x 2 − 3.4 x + 20.2)dx
0

The program follows:

% Example_9_2.m
% This program evaluates the integral of the function 'f1'
% between A and B by MATLAB's integral function. Since the function 'f1'
% is just a single line, we can use the anonymous form of the function.
clear; clc;
f1=@(x) (x.^3+3.2*x.^2-3.4*x+20.2);
A=0.0; B=10.0;
I = integral(f1,A,B);
% Note f1 is not enclosed by single quotation marks.
fprintf('Integration of f1 over [%.0f,%.0f] ',A,B);
fprintf('by MATLAB''s integral function:\n');
fprintf('f1 = x^3+3.2*x^2-3.4*x+20.2 \n');
fprintf('integral = %10.4f \n',I);
--------------------------------------------------------------------------

Program Results:
Integration of f1 over [0,10] by MATLAB's integral function:
f1 = x^3+3.2*x^2-3.4*x+20.2
integral = 3598.6667
>>
--------------------------------------------------------------------------

We see that the results are the same as those obtained in Example 9.1.

Example 9.3
1
t
EVALUATE: I 2 =
∫t
0
3
+t+1
dt

% Example_9_3.m
% This program evaluates the integral of Example 9.3 by MATLAB's
% integral function. A separate .m file describes the integrand
% to be integrated.
% The integrand is t/(t^3+t+1.0)
clear; clc;
A=0.0; B=1.0;
I2 = integral(@f2_func,A,B);
fprintf('Integration of integrand over [%.0f,%.0f] ',A,B);
fprintf('by MATLAB''s integral function:\n');
fprintf('Integrand = t/(t^3+t+1) \n');
fprintf('integral=%f \n',I2);
192 MATLAB® Essentials

-------------------------------------------------------------------------
% This function works with Example_9_3.m
function f=f2_func(t)
f = t./(t.^3+t+1.0);
-------------------------------------------------------------------------

Program Results:
Integration of integrand over [0,1] by MATLAB’s integral function:
Integrand = t/(t^3+t+1)
integral=0.260069
>>
--------------------------------------------------------------------------

Let us evaluate the improper integral described by Equation 9.3 by

MATLAB’s integral function with limits from 0 to 1. Recall that the func-
tion log(1+ x)/x is undefined at x = 0.
Again, we will use an anonymous function to describe the integrand.

Example 9.4
% Example_9_4.m
% This program evaluates the improper integral log(1+x)/x with
% limits from 0 to 1 using MATLAB's integral function.
clear; clc;
I3=@(x) log(1+x)./x;
A=0; B=1;
fprintf('This program uses MATLAB's integral function to \n');
fprintf('evaluate the improper integral of log(1+x)/x \n');
fprintf('from %2.0f to %2.0f. \n',A,B);
I = integral(I3,A,B);
fprintf('I = %10.6f \n',I);
--------------------------------------------------------------------------------

Program Results:
This program uses MATLAB's integral function to evaluate
the improper integral of log(1+x)/x from 0 to 1.
I = 0.822467
>>
------------------------------------------------------------------------------------

We see that the answer by MATLAB’s integral function is the same as

shown in Section 9.3.
Numerical Integration 193

Exercises

E9.2. Use MATLAB’s integral function to evaluate the following integrals.

Note that integral exercises 4, 5, and 6 are improper integrals.
3
dx
1. I =
∫ 5e
0
3x
+ 2 e −3 x
π/2
sin x dx
2. I =
∫
− π/2
1 − 4 sin 2 x
π

∫
3. I = (sinh x − cos x) dx
0
1
3 e x dx
4. I =
∫
0
1− x2
1
log(1 + x )dx
5. I =
∫
0
( 1− x )
1
log(1 + x) dx
6. I =
∫
0
( 1− x2 )

REVIEW 9.1
1. What is the formula for evaluating the integral, I = ∫ AB f ( x) dx by
the Simpson’s rule?
2. What is the name of MATLAB’s function for integrating a
single variable function?
3. In MATLAB’s function for integrating a single variable function
how does one define the function to be integrated?
4. If the integrand contains nonlinear terms, how must they be
treated?
5. Will MATLAB’s integral function treat improper integrals?
194 MATLAB® Essentials

9.5 MATLAB’s integral2 Function

The MATLAB’s function for numerically evaluating a double integral is
integral2. This function replaces MATLAB’s dblquad function. A descrip-
tion of the function follows:

Q = integral2(FUN, XMIN, XMAX, YMIN, YMAX)

where Q is the result of the double integration, FUN(X,Y) is a function handle

for the two-dimensional self-written integrand function. The  limits of inte-
gration are XMIN, XMAX, YMIN(X), YMAX(X), where XMIN <= X <= XMAX
and YMIN(X) <= Y <= YMAX(X). YMIN and YMAX may be either a scalar value
or a function handle.
The self-written function FUN(X,Y) should accept vectors X and Y and
return a vector Z of values of the integrand. The X and Y input variables to
function FUN comes from MATLAB’s integral2 function and the output
vector Z from function FUN goes to MATLAB’s integral2 function. The
output, Q, from the integral2 function goes to the program that calls the
integral2 function.

Example 9.5
Calculate the volume of a hemisphere of radius, R, by MATLAB’s integral2
function.
To find the volume, we define a differential volume element, dV, as follows:

dV = R 2 − x 2 − y 2 dxdy (as shown in Figure 9.3) and double-integrate over the inter-
vals x = [−R, R] and y = [−sqrt(R 2 − x 2 ), sqrt(R 2 − x 2 )].

% Example_9_5.m
% This program calculates the volume of a hemisphere (with R=1)
% using MATLAB's integral2 function. The solution is compared with the
% known exact solution for the volume of a hemisphere.
clear; clc;
R = 1;
ymin=@(x) -sqrt(R^2-x.^2);
ymax=@(x) sqrt(R^2-x.^2);
funz=@(x,y) sqrt(R^2-x.^2-y.^2);
V = integral2(funz,-R,R,ymin,ymax);
V_exact = 2/3*pi*R^3;
% print results
fprintf('Volume V of a hemisphere of radius %.1f m \n',R);
fprintf('V by intgral2 = %.4f m^3\n',V);
fprintf('V exact = %.4f m^3\n',V_exact);
---------------------------------------------------------------------------
Numerical Integration 195

Differential volume element

dV = R2 − x2 − y2 dxdy

R2 − x2 − y2

y
R
dx
dy

FIGURE 9.3
Infinitesimal volume inside a hemisphere.

Program Results
Volume V of a hemisphere of radius 1.0 m
V by intgral2 = 2.0944 m^3
V exact = 2.0944 m^3
>>
---------------------------------------------------------------------------

Example 9.6
The object shown in Figure 9.4 is enclosed by two curves, one of which is a straight
line and the other is a parabola. The object thickness, ∆z , is 5 cm. Take the object mate-
rial to be steel with a mass density, ρ = 8000 kg/m3. The dimensions in the figure are
also in cm.

1. Use MATLAB’s integral2 function to estimate the mass of the object.

NO T E : m = ρ ∆ z
∫∫ dx dy
A

Print out m, include units.

2. Using 60 subdivisions on the x domain, determine ymin and ymax for the
region, where ymin and ymax are the minimum and maximum y positions
respectively in the region of interest. For every other x position, print out a
table of x, ymin , and ymax, include table headings and units.
196 MATLAB® Essentials

integration region
4

3.5

2.5
yc1 & yc2

1.5

0.5

0
−1 −0.5 0 0.5 1 1.5 2
x

FIGURE 9.4
Object enclosed by two curves.

% Example_9_6.m
% This example determines the mass of an object that is enclosed
% by 2 curves. The lower curve is a parabola, y=x^2, and the upper
% curve is a straight line,y=x+2
% x range is from -1 to 2.
% y range is from 0 to 4.
% m=dz*Integration of (rho dx dy) from ymin to ymax.
% The input variables to FUN of MATLAB's integral2 function is (x,y).
% Since the integrand is 1, we needed to express the integrand as x./x
clear; clc;
rho=8e-3;
fun_9_6= @(x,y) x./x;
ymin=@(x) x.^2;
ymax=@(x) x+2;
dz=5.0;
m=dz*rho*integral2(fun_9_6,-1,2,ymin,ymax);
fprintf('m = %8.4f (kg) \n',m)
fprintf('\n');
x=-1:3/60:2;
ymin=x.^2;
ymax=x+2;
fprintf(' j x(cm) ymin(cm) ymax(cm) \n');
fprintf('-------------------------------------------\n');
for j=1:2:length(x)
fprintf('%2i %8.2f %8.4f %8.4f \n',...
j,x(j),ymin(j),ymax(j));
end
Numerical Integration 197

plot(x,ymin,x,ymax), xlabel('x(cm)'),ylabel('ymin(cm)& ymax(cm)'),

grid, title('ymin & ymax vs. x');
------------------------------------------------------------------------

Program Results:
m = 0.1800 (kg)

j x(cm) ymin(cm) ymax(cm)

-------------------------------------------
1 -1.00 1.0000 1.0000
3 -0.90 0.8100 1.1000
5 -0.80 0.6400 1.2000
7 -0.70 0.4900 1.3000
9 -0.60 0.3600 1.4000
11 -0.50 0.2500 1.5000
13 -0.40 0.1600 1.6000
15 -0.30 0.0900 1.7000
17 -0.20 0.0400 1.8000
19 -0.10 0.0100 1.9000
21 0.00 0.0000 2.0000
. . . .
. . . .
51 1.50 2.2500 3.5000
53 1.60 2.5600 3.6000
55 1.70 2.8900 3.7000
57 1.80 3.2400 3.8000
59 1.90 3.6100 3.9000
61 2.00 4.0000 4.0000
>>
-----------------------------------------------------------------

REVIEW 9.2

1. What is the name of MATLAB’s function for integrating a two-

dimensional function?
2. List the arguments that go into MATLAB’s function for inte-
grating a two-dimensional function.

Projects
P9.1. This exercise is from Thermodynamics. The entropy change of an ideal
gas from state (T1 , p1) to state (T2 , p2) is given by
T2
dT p2
s(T2 , p2 ) − s(T1 , p1 ) =
∫ c (T ) T
T1
p − R ln
p1
(P9.1)

where:
s is the entropy (kJ/kg-K)
cp is the specific heat at constant pressure (kJ/kg-K)
198 MATLAB® Essentials

p is the pressure (kPa)

T is the absolute temperatue (K)
R is the gas constant (kJ/kg-K)

The specific heat, cp(T), can be approximated by a fourth-degree polynomial

[2], that is,

cp (T ) = R( a1 + a2T + a3T 2 + a4T 3 + a5T 4 ) (P9.2)

R
R=
M
where:
R is the Universal gas constant = 8.314 (kJ/kmol-K )
M is the Molal mass (kg/kmol)

For carbon dioxide [2],

a1 = 2.401, a2 = 8.735 × 10 −3 , a3 = −6.607 × 10 −6 , a4 = 2.002 × 10 −9 , a5 = 0.0

M = 44.01 kg/kmol

Use MATLAB’s integral function to calculate the change in entropy,

s(T2 , p2 ) − s(T1 , p1 ) for (T1 , p1) = (400 K, 1.0 atm), (T2 , p2) = (900 K, 10.0 atm). Print
the results to the screen.
NO T E : 1 atm = 1.0132 ×10 5 N/m 2
P9.2. An ice slab, initially at temperature, Ti = −20°C is suddenly subjected to
a change in air temperature, T∞ = 10°C. This results in a heat transfer, q, per
unit surface area from the air to the ice slab. An approximate formula for
q (J/m2) follows:
tf  h2 α t   
  h αt   
∫
 k2 
q = h (T∞ − Ti )  e  
× 1 − erf    dt (P9.3)
  k  
0    
 
where:
k is the thermal conductivity of the slab material
h is the convective heat transfer coefficient
α is the thermal diffusivity of the slab material
T∞ is the air temperature
Ti is the initial slab temperature

NO T E 1 : The error function, erf(x) is written in MATLAB as erf(x).

NO T E 2 : If y = erf(x) and x is a vector, then y will also be a vector.
Assume: k = 2.2 W/m-C, α = 12.6 × 10−7 m2/s, h = 100 W/m2-C, and tf = 792 s.
Numerical Integration 199

Develop a MATLAB program using MATLAB’s integral function to evalu-

ate q. Use a separate .m file to describe the integrand. Print the constants, k, h,
α, Ti , T∞ and the result, q to the screen. Use e format for q. Include dimensions.
P9.3. We wish to determine the x-component of the Electric field at posi-
tion ( xo , yo , zo ) due to a line of point charges extending along the z-axis from
z = −0.01 m to z = +0.01 m (see Figure P9.1). For the derivation of the governing
equations for the electric field see Project P6.3 in [1]. Assume that dQ = λdzp
with λ = 2 × 10 −9 C/m. Here dQ is the strength of the point charge distribu-
tion. The x-component of the electric field at position ( xo , yo , zo ) is given by
0.01
λdzp xo − xp
Ex ( xo , yo , zo ) =
∫
−0..01 (
4πε o ( x − x )2 + ( y − y )2 + ( z − z )2
o p o p o p )
3
2
(P9.4)

z Point charge Q
located at rp
+
Measurement
rp r
location ro
ro
E
y

x
(a)
z

Differential line
…+++++++++++++++++…

charge element λdz

r Measurement
Line charge of density location
λ coul/m along ±z axis rp
ro dE

(b)

FIGURE P9.1
(a) Electric field at (xo , yo , zo) due to point charge at (xp , yp , zp). (b) Electric field at (xo , yo , zo) due
to line charge element λdz located along the z-axis.
200 MATLAB® Essentials

where ( xp , yp , zp ) is the coordinates of the point charge. Use MATLAB’s

integral function to determine Ex (0.005, 0, 0). The units of Ex is V/m. Take
ε o = 8.85 × 10 −12 farad/m.
P9.4. The solution for the displacement, Y ( x , t ), from the horizontal of a
vibrating string (see Figure P9.2) is given by
∞
nπx nπct
Y ( x, t ) = ∑ a sin
n=1
n
L
cos
L
(P9.5)

where

2 L
nπx
an =
L ∫ 0
f ( x )sin
L
dx (P9.6)

and

f ( x ) = Y ( x , 0) (P9.7)

For a compete derivation of the governing equation see Section 13.2 in

Reference 1.
Use MATLAB’s integral function to determine an , for n = 1, 2, …, 10.
Create a table and a plot of an versus n. Take L = 1.0 m and

0.4 x , 0 ≤ x ≤ 0.75 L
f (x ) =  (P9.8)
1.2 − 1.2 x , 0.75 L ≤ x ≤ L

P9.5. The components of the electric field Ex , Ey , and Ez resulting from a line
of point charges with a linear charge density, λ, with units Coulomb/meter,
which is evenly distributed along the z-axis from z = −0.01 m to z = +0.01 m
(see Figure P9.1) is given by
0.01
λdzp xo − xp
Ex ( xo , yo , zo ) =
∫
−0..01 (
4πε o ( x − x )2 + ( y − y )2 + ( z − z )2
o p o p o p )
3
2
(P9.9)

0.01
λdzp yo − yp
Ey ( x o , y o , z o ) =
∫
−0..01 (
4πε o ( x − x )2 + ( y − y )2 + ( z − z )2
o p o p o p )
3
2
(P9.10)

y String

FIGURE P9.2
Vibrating string.
Numerical Integration 201

TABLE P9.1
Table Format for Presenting Ex Values
E x Values
x y
−0.05 −0.03 −0.01 0.01 0.03 0.05
−0.05 — — — — — —
−0.04 — — — — — —
−0.03 — — — — — —
.
.
0.03 — — — — — —
0.04 — — — — — —
0.05 — — — — — —

0.01
λdzp zo − zp
Ez ( xo , yo , zo ) =
∫
−0..01 (
4πε o ( x − x )2 + ( y − y )2 + ( z − z )2
o p o p o p )
3
2
(P9.11)

where:
( xp , yp , zp ) represents the position of the point charges
( xo , yo , zo ) represents a point of interest in the vicinity of the point charges

The units of Ex , Ey , and Ez are Newton/Coulomb (N/C) or Volt/m (V/m).

Take ε o = 8.85 × 10 −12 farad/m.
Create a MATLAB program that will determine the electric field com-
ponent, Ex and Ey in the (x, y) plane for the interval −50 ≤ x ≤ 50 mm and
−50 ≤ y ≤ 50 mm with a step size of 10 mm. Omit the point (x, y) = (0,0). Due to
symmetry, assume that Ez = 0. Print Ex and Ey in separate tables using a table
format as shown in Table P9.1. Print Ex and Ey to one decimal place.
P9.6. This project involves determining the surface area of a hemisphere of
radius one meter.
A differential surface area on the hemisphere is dA =(R sin φ dθ) Rd φ
as shown in Figure P9.3. Create a MATLAB program using MATLAB’s
integral2 function to find the surface area. Take φ = [0, π/2] and θ = [0, 2π].
Compare your answer with the known exact expression for the surface area
of a hemisphere, which is 2πR2.
P9.7. An object is enclosed by two curves, one of which is a straight line
and the other is a parabola. The x range of the object is from –2 cm to +4 cm.
The equation of the parabola is

y = 6 − 1.5 x 2
202 MATLAB® Essentials

dA = (R sinϕ dθ)(Rdϕ)

dϕ
y
R R sinϕ
dθ

FIGURE P9.3
A differential surface area on the hemisphere.

The equation of the straight line is

y = −3 x − 6

The object thickness, ∆z, is 5 cm. Take the object material to be steel with a
mass density, ρ = 8000 kg/m3.

a. Using 60 subdivisions on the x domain, determine ymin and ymax for

the region, where ymin and ymax are the minimum and maximum
y positions respectively in the region of interest. For every other x
position, print out a table of x, ymin and ymax , include table headings
and units.
b. Create a two dimensional plot of the object.
c. Use MATLAB’s integral2 function to estimate the mass of the object.

Print out to the Command Window the mass, m, include units.

NO T E : m = ∆z
∫∫ ρ dxdy.
A
Numerical Integration 203

P9.8. The (xc, yc) position of the center of mass of the object described in
Project P9.7 is given by

mxc = ∆z ρ
∫∫ x dxdy
A
∫∫ y dxdy
myc = ∆z ρ
A
(P9.12)

where:
ρ is the mass density of the material
m is the mass of the object
∆z is the thickness of the object

Create a MATLAB program that will evaluate (xc , yc) using MATLAB’s
integral2 function. Print the results to the screen to four decimal places.
P9.9. Using the infinitesimal volume shown in Figure 9.3 and MATLAB’s
integral2 function, determine the centroid position, zc , of the hemisphere
described in Example 9.5. By symmetry, we can assume that xc = 0 and yc = 0.
Noting that zc for the infinitesimal volume is at the center position, that is,

1
zc dV = R2 − x 2 − y 2 × R2 − x 2 − y 2 dxdy
2

1
∫∫ 2 ( R − x )
2 2
zc V = − y 2 dxdy (P9.13)
A

where:
x = [−R, R] and y = [−sqrt(R2 − x 2 ), sqrt(R2 − x 2 )]
V is the volume of the hemisphere = 2/3πR3

We can also determine the centroid position analytically by taking an infini-

tesimal volume shown in Figure P9.4, then

dV = πr 2 dz (P9.14)

where:

r = (R2 − z 2 ) (P9.15)

Thus,
R R
π π  z2 z4  3
zc =
V ∫
0
(R2 z − z 3 )dz =
2
×  R2 −  = R
πR3  2 4 0 8
(P9.16)
3
204 MATLAB® Essentials

dz
r

z
R

FIGURE P9.4
Infinitesimal region used to determine the centroid position zc analytically.

Compare zc obtained by the use of MATLAB’s integral2 function with the

exact solution. Take R = 1.

Reference
1. Bober, W., Introduction to Numerical and Analytical Methods with MATLAB for
Engineers and Scientists, CRC Press, Boca Raton, FL, 2014.
10
Numerical Integration of Ordinary
Differential Equations

10.1 Introduction
Many ordinary differential equations (ODE) result from a particular physi-
cal law. The physical law is a mathematical model of some particular physi-
cal phenomenon. Many of the equations that have been used in this book
are based on Newton’s second law of motion. For example, the equations
used to describe the motion of a free falling ball in a gravitational field
(Example 2.7) or the motion of the mass in a mass-spring-dashpot system
(Exercise E2.3-for a complete derivation of the governing equations, see
Project P2.5 in Reference  1), or the velocity and position of the basketball
(Exercise E2.4) are differential equations based on Newton’s second law.
The voltage in a parallel RLC circuit (Exercise E2.6) resulted from several
electrical laws, including Kirchhoff’s current law, which resulted in an ordi-
nary differential equation whose solution is given in Equation 2.13 (for a
complete derivation of the governing equations, see P2.7 in Reference 1).
Ordinary differential equations can be broken up into two categories:
1. Initial value problems are those in which the initial conditions of the
variables are known. All of the examples and exercises mentioned
above fall into this category. Additional examples include launch-
ing a rocket with a known initial position and velocity or the value
of a circuit node voltage (or its slope) at t = 0. In this chapter, we
only cover the initial value problem along with MATLAB ®’s built-in
ode45 function to solve these types of problems.
2. Boundary value problems in which we know variable conditions at
specific coordinates in the problem geometry. For example, deter-
mining the temperature at various positions along a bar when the
end temperatures at the bar ends are known. Other examples include
determining the deflection of a beam along its length when the deflec-
tion at its ends is known, or determining the electric potential along
the length of a conductor when the electric potential at both ends of a
conductor are known. This topic is covered in Chapter 11.

205
206 MATLAB® Essentials

10.2 Initial Value Problem and MATLAB’s Ordinary

Differential Equations Function
MATLAB has several built-in ODE functions that solve a system of first-
order ordinary differential equations, including ode23 and ode45. In this
chapter, we will demonstrate MATLAB’s ode45 function, which is based on
fourth- and fifth-order Runge–Kutta methods. A description of the ode45
function follows (MATLAB’s description of ode45 can be obtained by typing
help ode45 in the Command Window):
The ode45 function solves a system of first-order ordinary differential
equations of the form y′n = f (t , y1 , y2 ,..., yn ) from time T0 to TFINAL with
initial conditions Y0. Here we have assumed that the independent variable is
time, t, and the dependent variables are y1 y2 ... yn.
The syntax for MATLAB’s ode45 function is

[TOUT, YOUT] = ODE45(ODEFUN, TSPAN, Y0)

The ODEFUN argument is a function handle to the function describing

the system of differential equations. TSPAN = [T0 Tfinal] is vector
describing a time interval covered by the system of differential equations.
Y0 is a vector describing the initial conditions. The function ODEFUN must
take two input arguments: a scalar for the independent variable, t, and a
vector for the dependent variables, Y = [y1 y2 ... yn ]. The output of function
ODEFUN must be the system of differential equations as a column vector
of the form

y′( 1) = f1( y(1), y(2),..., y(n)),

y′(2) = f 2 ( y(1), y(2), .... y(n)),
.

y′ (n) = f2 ( y(1), y(2), .... y(n))

The time interval TSPAN is typically a two-element vector containing a

start and end time; ode45 will automatically choose an appropriate time
step (and might even vary the time step within the interval). ode45 will
return two vectors: a list of time points TOUT and the solution YOUT at each
time point. If you want to force ode45 to solve the system at specific time
points, then you can explicitly specify the time points in TSPAN = [T0 T1..
TFINAL]. The output variable TOUT is a column vector equal in size to
tspan and YOUT are column vectors of y(1), y(2),….y(n).
Numerical Integration of Ordinary Differential Equations 207

Example 10.1
Let us consider the ball-bearing problem of Example E2.5. Applying Newton’s
second law to the ball bearing gives the following first-order differential equation:

W dV
= 6πRµ( VT − V ) (10.1)
g dt

where:
V is the ball-bearing velocity
V T terminal velocity of the ball bearing = (W − B) 6πR µ
W is the weight of the ball bearing = ρsteel υg
B is the buoyancy acting on the ball bearing = ρfluid υg
R is the radius of the ball bearing
υ is the volume of the sphere = (4 3) πR 3
ρ is the mass density
µ is the viscosity of the fluid
g is the gravitational constant = 9.81 m/s2

To see the full derivation of Equation 10.1 see Exercise E2.5 in Reference 1.
In the notation of ode45, (dV/dt) = V′
Take µ = 3.85 (N-s)/m2, R = 0.01 m, ρsteel = 7910 kg/m3, ρoil = 899 kg/m3.
We will take V(0) = 0.

The program follows.

% Example_10_1.m
% This program determines the velocity of a ball bearing
% dropped in a vat of fluid. The ball bearing reaches a
% terminal velocity when the unbalanced force acting
% on the object is zero.
% The program compares the velocity determined by the
% by both an analytical solution and MATLAB's ode45 function.
clear; clc;
global R mu g VT W VT
R=0.01; rho_steel=7910; rho_fluid=899; mu=3.85; g=9.81;
vol=4/3*R^3;
W=rho_steel*g*vol;
B=rho_fluid*g*vol;
VT=(W-B)/(6*pi*R*mu);
Vo=0;
tspan=0:0.01:0.2;
[t,V]=ode45('dVdt',tspan,Vo);
% Closed form solution is V2
arg=6*pi*R*mu*g/W;
V2=VT*(1-exp(-arg*t));
fprintf('t(s) V(m/s) V2(m/s) \n');
fprintf('---------------------------------\n');
for i=1:length(t)
fprintf('%4.2f %6.4f %6.4f \n',t(i),V(i),V2(i));
end
208 MATLAB® Essentials

fprintf('Terminal Velocity, VT= %6.4f(m/s) \n', VT);

plot(t,V,t,V2,'x'), xlabel('t(s)'), ylabel('V(m/s)'), grid,
title('V vs. t');
----------------------------------------------------------------------
% dVdt.m
% This function works with Example_10_1.m
function Vprime=dVdt(t,V)
global R mu g VT W VT
Vprime= 6*pi*R*mu*g/W*(VT-V);
----------------------------------------------------------------------

Program Results:
See Figure 10.1.

t(s) V(m/s) V2(m/s)

---------------------------------
0.00 0.0000 0.0000
0.01 0.0629 0.0629
0.02 0.0945 0.0945
0.03 0.1103 0.1103
0.04 0.1183 0.1183
0.05 0.1223 0.1223
0.06 0.1243 0.1243
0.07 0.1253 0.1253
0.08 0.1259 0.1259
0.09 0.1261 0.1261
0.10 0.1262 0.1262
. . .
. . .

V vs. t
0.14

0.12

0.1

0.08
V(m/s)

0.06

0.04

0.02

0
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2
t(s)

FIGURE 10.1
Velocity of ball bearing versus time.
Numerical Integration of Ordinary Differential Equations 209

0.15 0.1264 0.1264

0.16 0.1264 0.1264
0.17 0.1264 0.1264
0.18 0.1264 0.1264
0.19 0.1264 0.1264
0.20 0.1264 0.1264
Terminal Velocity, VT= 0.1264(m/s)
>>
-----------------------------------------------------------------------

Example 10.2
Whenever we deal with a second-order differential equation, we need to reduce the
second-order differential equation to two first-order differential equations. Suppose
we consider the mass motion in a mass-spring-dashpot system of Exercise E2.2. The gov-
erning differential equation is

c k
y′′ + y′ + y= 0 (10.2)
m m

To see the full derivation of Equation 10.2 see Project P2.5 in Reference 1.
To reduce Equation 10.2 to two first-order differential equations, let y′ = V , then

c k
V′ = − V− y
m m (10.3)
y′ = V

We will take y(0) = 0.5 and y’(0) = 0.

We can now use MATLAB’s ode45 to solve the system.
The program follows:

% Example_10_2.m
% This program determines the position and velocity
% of a mass in a mass-spring-dashpot system using
% MATLAB's ode45 function.
% m=25 kg; c=5 N-s/m; k=100 N/m;
% Y(1)=y
% Y(2)=V
% Y(1)_prime=Y(2)
% Y(2)_prime= -c/m*Y(2)-k/m*Y(1)
clear; clc;
initial=[0.5 0.0];
tspan=0.0:0.1:10.0;
[t,Y]=ode45(@dYdt,tspan,initial);
y=Y(:,1);
V=Y(:,2);
fprintf(' t(s) y(s) V(m/s) \n');
fprintf('----------------------------------\n');
for i=1:2:101
fprintf('%5.2f %10.4f %10.4f \n',t(i),y(i),V(i))
end
210 MATLAB® Essentials

plot(t,y), xlabel('t(s)'), ylabel('y(m)'), grid,

title('y vs. t');
figure;
plot(t,V), xlabel('t(s)'), ylabel('V(m/s)'), grid,
title('V vs. t');
----------------------------------------------------------------------
% dYdt.m
% This function works with Example_10_2.m
function Yprime=dYdt(t,Y)
m=25; c=5; k=100;
% Y(1)= y; Y(2)=V
Yprime=zeros(2,1);
Yprime(1)=Y(2);
Yprime(2)=-c/m*Y(2)-k/m*Y(1);
----------------------------------------------------------------------

Program Results:
See Figure 10.2a and b.

t(s) y(m) V(m/s)

----------------------------------
0.00 0.5000 0.0000
0.20 0.4611 -0.3818
0.40 0.3523 -0.6894
0.60 0.1932 -0.8785
0.80 0.0105 -0.9239
1.00 -0.1668 -0.8248
1.20 -0.3111 -0.6016
1.40 -0.4017 -0.2942
1.60 -0.4265 0.0466
1.80 -0.3845 0.3670
2.00 -0.2844 0.6178
. . .
. . .
9.00 0.1231 0.3109
9.20 0.1729 0.1808
9.40 0.1939 0.0278
9.60 0.1841 -0.1235
9.80 0.1462 -0.2497
10.00 0.0871 -0.3323
>>
----------------------------------------------------------------------
Numerical Integration of Ordinary Differential Equations 211

y vs. t
0.5

0.4

0.3

0.2

0.1
y(m)

−0.1

−0.2

−0.3

−0.4

−0.5
0 1 2 3 4 5 6 7 8 9 10
(a) t(s)

V vs. t
0.8

0.6

0.4

0.2

0
V(m/s)

−0.2

−0.4

−0.6

−0.8

−1
0 1 2 3 4 5 6 7 8 9 10
(b) t(s)

FIGURE 10.2
(a) Plot of mass displacement versus time and (b) plot of mass velocity versus time.
212 MATLAB® Essentials

Example 10.3
In this example, we consider the voltage in a parallel RLC circuit described in Project
P2.10 (see Figure 2.22). For a complete derivation of the governing equation, see
Project 2.7 in Reference 1. The governing differential equation for the circuit voltage is

d2v 1 dv 1
2
+ + v=0 (10.4)
dt RC dt LC

To reduce Equation 10.4 to two first-order differential equations, let v′ = u, then

1 1
u′ = − u− v
RC LC (10.5)
v′ = u

dv 6
=
We will take v(0) 6=
and (0)
dt RC
The program follows:

% Example_10_3.m
% This program determines the voltage in a parallel RLC circuit using
% MATLAB's ode45 function.
% R=100 ohm; L=1 mHc; C=1 microfarad;
% v(0)=6 volt; dvdt=v(0)/(R*C)
% Y(1)=v
% Y(2)=dv/dt=u
% Yprime(1)=Y(2)
% Yprime(2)= -1/(R*C)*Y(2)-1/(L*C)*Y(1)
clear; clc;
global R L C;
R=100; L=1e-3; C=1e-6;
initial=[6 6/(R*C)];
tspan=0:5e-6:5e-4;
[t,Y]=ode45(@dvoltdt,tspan,initial);
v=Y(:,1);
u=Y(:,2);
t2=t*1.0e+6;
fprintf(' t v dv/dt \n');
fprintf('(micro-sec) (volt) (volt/sec) \n');
fprintf('----------------------------------------\n');
for i=1:length(t2)
fprintf(' %5.0f %10.2f %10.0f \n',t2(i),v(i),u(i))
end
plot(t,v), xlabel('t(s)'), ylabel('v(volt)'), grid,
title('v vs. t');
figure;
plot(t,u), xlabel('t(s)'), ylabel('dv/dt(volt/s)'), grid,
title('dv/dt vs. t');
----------------------------------------------------------------------
Numerical Integration of Ordinary Differential Equations 213

% This function works with Example_10_3.m

function Yprime=dvoltdt(t,Y)
global R L C;
% Y(1)=v
% Y(2)=dv/dt=u
Yprime=zeros(2,1);
Yprime(1)=Y(2);
Yprime(2)= -1/(R*C)*Y(2)-1/(L*C)*Y(1);
----------------------------------------------------------------------

Program Results:
See Figure 10.3a and b.

t v dv/dt
(micro-sec) (volt) (volt/sec)
----------------------------------------
0 6.00 60000
5 6.22 27209
10 6.27 -4644
15 6.17 -34837
20 5.93 -62711
25 5.55 -87688
30 5.06 -109310
35 4.46 -127231
40 3.79 -141188
45 3.06 -151018
50 2.29 -156678
. . .
. . .
450 0.36 -20617
455 0.25 -21094
460 0.15 -21033
465 0.04 -20461
470 -0.06 -19418
475 -0.15 -17955
480 -0.24 -16127
485 -0.31 -13996
490 -0.38 -11629
495 -0.43 -9094
500 -0.47 -6462
>>
----------------------------------------------------------------------
214 MATLAB® Essentials

v vs. t
8

4
v(volt)

−2

−4
0 1 2 3 4 5 6
(a) t(s) × 10−4

× 105 dv/dt vs. t

1

0.5

0
dv/dt(volt/s)

−0.5

−1

−1.5

−2
0 1 2 3 4 5 6
(b) t(s) × 10−4

FIGURE 10.3
(a) Plot of voltage versus time. (b) Plot of dv/dt versus time.
Numerical Integration of Ordinary Differential Equations 215

Exercises

E10.1. The governing equation for the mass displacement in a mass-spring-

dashpot system subjected to an oscillatory driving force is

c k F
y′′ + y′ + y = o sin ωt (10.6)
m m m

Create a MATLAB program, using MATLAB’s ode45 function to solve

for y and y′ as a function of time, t. The natural frequency of the system,
ωn = k/m . Let us consider two cases:

1. ω = 1.5 ωn
2. ω = ωn

Take m = 25 kg, c = 5 N-s/m; k = 100 N/m; Fo = 50 N, y(0) = 0.5 m, y′(0) = 0.

Create a MATLAB program that uses MATLAB’s ode45 function to solve
for y(t) for cases (1) and (2). Notice that when ω = 1.5 ωn the amplitude of
the oscillation grows with time and is much larger than the amplitude for
the case when ω = ωn. This is the effect of resonance.

E10.2. Solve the following system of three first-order differential equations

using MATLAB’s ode45 function:

y1′ = y2 y3t

y′2 = − y1y3

y′3 = −0.51y1y2

Initial conditions: y1(0) = 0, y2 (0) = 1.0, and y3 (0) = 1.0.

E10.3. Solve the parallel RLC circuit of Example 10.3 for voltage, v, and the
inductor current, iL , by using MATLAB’s ode45 function. The governing
equations are

dv 1 1
=− v − iL
dt RC C

diL 1
= v
dt L

Assume R = 50 Ω, L = 1 µH, C = 10 nF, v(0) = 3.3 V, iL (0) = 0 A.

Plot v on iL on two separate graphs. Take 0 ≤ t ≤ 4 µsec in steps of 0.01e-6 sec.
 216 MATLAB® Essentials

Projects
P10.1. This project involves determining the temperature of a small solid
aluminum sphere dropped into a fluid contained in a vertical circular cylinder
of radius R. The sphere radius is r and the fluid depth is L. Neglecting heat
transfer to the container walls, the governing equations for this problem are

 dT 
mc  = h As ( Tf − Tal ) (P10.1a)
 dt  al

 dT 
mc  = h As ( Tal − Tf ) (P10.1b)
 dt  f

Use the following parameters for the problem:

ρal = 2707 kg/m 3 , ρ f = 880 kg/m 3 , cal = 896 J/kg- o C, c f = 1880 J/kg- o C,

o
= =
Tal (0) 80 20o C, r 0=
C, Tf (0) = .2 m , R 0= .5 m, h 890 W/m 2 - o C
.3 m, L 0=

0 ≤ t ≤ 300 s in steps of 0.5 s.

Create a MATLAB program using MATLAB’s ode45 function to determine

the temperatures of the aluminum sphere and the fluid. Plot Tal and Tf on the
same graph.

P10.2. An airplane flying horizontally at 50 m/s and at an altitude of 300 m

is to drop a food package weighing 2000 N to a group of people stranded

in an inaccessible area resulting from an earthquake. A  drag force, D, acts
on the package in the direction of the free stream, V, as seen from the
package (see Figure P10.1). We wish to determine the position of the pack-
age as a function of time and when the package hits the ground. Take

Vairp

y
θ

Ground

FIGURE P10.1
Path of a food package dropped from an airplane.
Numerical Integration of Ordinary Differential Equations 217

( x , y ) to be the position of the package at time t as seen from the posi-

tion of the airplane at the time of release and
(Vx , Vy ) to be the horizontal and vertical components of the package
velocity respectively.

Governing equations:

dVx C ρV 2A
=− d cos ϑ (P10.2a)
dt 2M

dVy C ρV 2A
= g− d sin ϑ (P10.2b)
dt 2M

dx
= Vx (P10.2c)
dt

dy
= Vy (P10.2d)
dt

Vx Vy
cos ϑ = , and sin ϑ = (P10.2e)
V V

V = Vx2 + Vy2 (P10.2f)

where:
Cd is the drag coefficient
ρ is the air density
M is the mass of package
A is the frontal area of package

Initial conditions:

=
x ( 0 ) 0=
, y (= =
0 ) 0, Vx ( 0 ) 50 m/s , Vy ( 0 ) 0.

Use the following parameters:

Cd = 0.8, ρ = 1.225 kg/m 3 , A = 1.0 m 2

Use MATLAB’s ode45 function to solve for ( t , x , y , Vx , Vy ) at intervals of

0.10 seconds for 0 ≤ t ≤ 10.0 seconds.

1. Create plots of x and y versus t both on the same graph.

2. Create plots of Vx and Vy versus t both on the same graph.
218 MATLAB® Essentials

3. Create a table containing ( t, x, y, Vx , Vy ) at intervals of 0.10 seconds.

Stop printing table the first time y > 300 m.
4. Use MATLAB’s function interp1 to interpolate for the ( t, x, Vx , Vy )
values when the package hits the ground. Print out these values.

P10.3. Figure P10.2 shows a third-order RLCC circuit. In order to run a time-
domain transient analysis, we transform the circuit into three first-order
differential equations which are

dvC1 1
= (−vC1 + vC 2 + RiL + vS ) (P10.3a)
dt RC1

dvC 2 1
= (vC1 − vC 2 − vS) (P10.3b)
dt RC2

diL 1
= (− vC1 + vS) (P10.3c)
dt L

vC1 vC2
+ − 1 − + 2
iL
iC1 iC2
C1 C2 iR

+
+ +
vS
vL L R vR
− 1 2 −
−

(a)

vS(t)

5V

t (μsec)
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
(b)

FIGURE P10.2
(a) A third-order RLCC circuit configuration and (b) pulse input.
Numerical Integration of Ordinary Differential Equations 219

We have chosen the three voltages and currents with derivative terms
(vC1, vC2 , and iL) as the state variables for this problem.
Construct a MATLAB program using MATLAB’s ode45 function to solve
for the variables vC1, vC 2 , and iL . Take C1 = 1 μF, C2 = 0.001 μF, R = 100 kΩ,
L = 0.01 mH. Use a time interval of 0 ≤ t ≤ 5 µs and a step size of 0.01 μs.
Assume vS (t) is a 5V pulse starting at time t = 0 with rise time of 0.1 μs, an on
time of 0.8 μs, and fall time of 0.1 μs (as shown in Figure P10.2). Initial condi-
tions: vC1(0) = 0, vC2 (0) = 0 , and iL (0) = 0.
Plot on separate graphs: vC1(t), vC 2 (t), iL (t), and vS (t) versus time.

P10.4. A small rocket with an initial mass of 350 kg, including a mass of
100 kg of fuel, is fired from a rocket launcher (see Figure P10.3). The rocket
leaves the launcher at velocity Vo and at an angle of θo with the horizontal.
Neglect the fuel consumed inside the rocket launcher. The rocket burns fuel
at the rate of 10 kg/s, and develops a thrust T = 6000 N. The thrust acts axi-
ally along the rocket and lasts for 10 s. Assume that the drag force also acts
axially and is proportional to the square of the rocket velocity. The govern-
ing differential equations describing the position and velocity components
of the rocket are as follows:

dVx VxT Vx K Vx2 + Vy2

= − (P10.4a)
dt m Vx2 + Vy2 m

dVy VyT Vy K Vx2 + Vy2

= − −g (P10.4b)
dt m Vx2 + Vy2 m

dx
= Vx (P10.4c)
dt

y
V
θ (t)

y
Ground
3,000 m

x
15 m
x
Vo θo

FIGURE P10.3
Rocket trajectory.
220 MATLAB® Essentials

dy
= Vy (P10.4d)
dt
V 2 = Vx2 + Vy2 (P10.4e)

where:
m is the mass of the rocket (varies with time)
Vx , Vy are the x and y components of the rocket’s velocity relative to the
ground
K is the drag coefficient
g is the gravitational constant
(x, y) are the position of the rocket relative to the ground
t is the time of rocket flight
The target lies on ground, which has a slope of 5%. The ground elevation
relative to the origin of the coordinate system of the rocket is given by

y g = 15 + 0.05 ( x − 3000) (P10.4f)

Using Equations P10.4a through P10.4d, write a computer program in

MATLAB using the MATLAB’s ode45 function that solves for x, y, Vx , and
Vy for 0 ≤ t ≤ 60 seconds in steps of 0.01 s. Use Equation P10.4f to solve for yg .
Take x(0) = 0, y(0) = 0, Vx(0) = Vocos θo , Vy(0) = Vosin θo , Vo = 150 m/s, θo= 60°,
K = 0.045 N–s2/m2, and g = 9.81 m/s2.

1. Print out a table for x, y, yg , Vx , Vy every 1.0 second.

2. Use MATLAB to plot x, y, and yg versus t on the same graph and Vx ,
Vy , versus t on the same graph.
3. Assume a linear trajectory between the closest two data points where
the rocket hits the ground. The intersection of the two straight lines
gives the (x, y) position of where the rocket hits the ground.

P10.5. We wish to examine the time temperature variation of a fluid, Tf ,

enclosed in a container with a heating element and a thermostat. The walls of
the container are pure copper. The fluid is engine oil, which has a temperature
Tf that varies with time. The thermostat is set to cut-off power from the heat-
ing element when the Tf reaches 65°C and to resume supplying power when
Tf reaches 55°C. The outside room temperature, T∞, remains constant at 15°C.
Wall properties:

k = 386.0 w/m-C, c = 0.3831 kJ/kg-C, ρ = 8954 kg/m 3

Engine oil properties:

k = 0.137 w/m-C, c = 2.219 kJ/kg-C, ρ = 840 kg/m 3

The inside size of the container is (0.5 m × 0.5 m × 0.5 m)

Numerical Integration of Ordinary Differential Equations 221

The wall thickness is 0.01 m. Thus, the

Inside surface area, As,i = 1.5 m2

Outside surface area, As,o = 1.5606 m2
Engine oil volume, Voil = 0.125 m3
Wall volume, Vwall = 0.0153 m3
20000 W when t f < 55°C
The power, Q, of the heating element = 
0 when t f > 65°C
The inside convective heat transfer coefficient, hi = 560 W/m2-C
The outside convective heat transfer coefficient, ho = 110 W/m2-C

Using a lump parameter analysis (assuming that the engine oil is well mixed)
in heat transfer, the governing equations describing the time temperature
variation of both materials are as follows:

dθ f
= − a1(θ f − θw ) + a5 (P10.5a)
dt
dθw
= a2 (θ f − θw ) − a3θw = a2θ f − ( a2 + a3 )θw (P10.5b)
dt
where:
θ f = Tf − T∞ (P10.5c)

θw = Tw − T∞ (P10.5d)

hi As , i hA hA Q
a1 = , a2 = i s , i , a3 = o s , o , a4 = a2 + a3 , a5 = (P10.5e)
mf c f mw cw mw cw mf c f

Initial conditions:

Tf (0) = Tw(0) = 15°C

T∞ = 15°C

Using MATLAB’s ode45 function to construct a simulation of this system.

Run the time for 3600 seconds. Print out values of Tf and Tw versus t at every
100 seconds. Construct plots of Tf and Tw versus t. Use tspan = 0:1:3600.
P10.6. We wish to determine the altitude and velocity of a helium filled spher-
ically shaped balloon as it lifts off from its mooring. We will assume that
atmospheric conditions can be described by the U.S. Standard Atmosphere.
We will assume that there is no change in the balloon’s volume. For a com-
plete derivation of the equations given below, see Project P7.9 in Reference 1.
The governing equations describing the motion of the balloon are
222 MATLAB® Essentials

dz
=V (P10.6a)
dt
dV 1
= ( B −W − sgn * D ) (P10.6b)
dt M
where:
z is the altitude of the centroid of the balloon
V is the vertical velocity of the balloon
t is the time
B is the buoyancy force acting on the balloon (varies with altitude)
M is the total mass of the balloon material, ballast, and the gas
W is the total weight of the balloon material, ballast, and the gas = Mg
D is the drag on the balloon
sgn = +1, if (dz dt) ≥ 0 and sgn = −1, if (dz dt) < 0

The U.S. Standard atmosphere as applied to this balloon problem consists of

the following governing equations:

dp p
=− gV (P10.6c)
dt RT

T = Ti − λz (P10.6d)

p
ρ= (P10.6e)
RT
where:
p is the outside air pressure at the centroid of the balloon
ρ is the outside air density at the centroid of the balloon
g is the gravitational constant that varies with altitude
R is the gas constant for air
T is the outside air temperature at the centroid of the balloon
Ti is the temperature at the earth’s surface = 288.15 (K).
λ is the lapse rate
The buoyancy force, B, is given by

B = ρgυ (P10.6f)

and

 r 
g = g0  e  (P10.6g)
 z + re 

where:
υ is the volume of the balloon = (4/3) rb3
rb is the radius of the balloon
Numerical Integration of Ordinary Differential Equations 223

re is the radius of the earth

g0 is the gravitational constant near the earth’s surface
g is the gravitational constant at an elevation of the centroid of the balloon

For low Reynolds Number, Re, less than 0.1, the drag force, D, is given by
Stokes formula, which is

D = 6πµVrb (P10.6h)

For flow speeds with Re > 0.1, use

ρ 2
D = Cd V A (P10.6i)
2

where:
Cd is the drag coefficient
A is the frontal area of the balloon = πrb2

The drag coefficient, Cd , is given by

24 6
Cd = + + 0.4 (P10.6j)
Re 1.0 + Re

where
2ρVrb
Re = (P10.6k)
µ

and µ is the fluid viscosity

The fluid viscosity, µ, can be determined by the Sutherland formula,
which is
1.5
 T   T +S 
µ = µ0    0  (P10.6l)
 T0   T + S 

For air, S = 110.4 K, µ 0 = 1.71e-5 N-s/m2, T0 = 273 K.

Write a computer program, using MATLAB’s ode45 function that will
determine the balloon’s altitude as a function of time.
NO T E : Equations P10.6a through P10.6c represent a system of three first-
order ordinary differential equations that can be solved by MATLAB’s
ode45 function.
Create plots of z versus t, v versus T, and p versus t. Use the following values:

M = 2200 kg , rb = 7.816 m, υ = 4/3πrb3 m 3 , R = 287 J/(kg-K), Ti = 288.15 K ,

λ = 0.0065 (K/m), g0 = 9.81 m/s 2 , and re = 6371e + 3 m

224 MATLAB® Essentials

C1

i3
R1 R2 v2
vin +
v i4 vout
1 i5 −
i1 i2
C2

FIGURE P10.4
Sallen–key circuit.

Use a tspan = 0.0:0.1:1000 and the following initial conditions:

z (0) = rb , V(0) = 0, p(0) = 1.0132e + 5

P10.7. The Sallen–Key circuit (Figure P10.4) is commonly used to implement

second-order (or higher) filters. The following equations model the circuit
using ordinary differential equations. It is assumed that the op amp is ideal
resulting in v2 = vout and i5 = 0. Applying Kirchhoff’s Current Law at the
nodes labeled v1 and v2 , and the constituent relations for resistors (vR = iR R)
and capacitors (iC = C(dvC dt)), the following equations are obtained:

dvout  −1   1 
=  vout +  R C  v1 (P10.7a)
dt R C
 2 2  2 2

dv1  1 1   1 1 1   1 
= −  vout +  R C − R C − R C  v1 +  R C  vin (P10.7b)
dt  R2C1 R2C2   2 2 1 1 2 1   1 1

1. Solve for vout and v1 using MATLAB’s ode45 function. Assume that
the input to the circuit vin is a step voltage that changes from 0 V to
1 V at time t = 0 +. Assume the following values for the circuit ele-
ments: R1 = 5000 Ω, R2 = 5000 Ω, C1 = 2200 pF, C2 = 1100 pF. Use a
time interval of t = [0, 100 µs] s and assume vout=(0) v=
1 (0 ) 0.
2. Find the impulse response of the circuit by first creating a MATLAB
function pulse(t) that returns the following values:
106 for 0 < t < 10 −6
pulse(t) = 
0 otherwise
Then, solve for vout and v1 using MATLAB’s ode45 function where
vi = pulse(t). Use the same component values, time interval, and
initial conditions as in part 1.
3. Plot the step response (from part 1) and the impulse response
(from part 2) on the same set of axes. What relationship can you see
between the two?
Numerical Integration of Ordinary Differential Equations 225

P10.8. Exercise E2.4 involved a basketball player shooting a basketball toward

the hoop. The basketball was released 6 m from the center of the hoop with a
velocity, Vo, and making an angle of 40° with the horizontal (see Figure 2.20).
Equations describing the motion of the basketball based on Newton’s second
law and neglecting drag were given. The solution obtained in Exercise E2.4
was that if Vo = 8.7098 m/s the basketball would reach the center of the hoop
in 0.8993 s. We now want to include drag in determining the motion of the
basketball. We have assumed that the drag is in the opposite direction of the
ball’s motion. The governing equations become

dVx ρACd Vx Vx2 + Vy2

=− (P10.8a)
dt 2m

dVy ρACd Vy Vx2 + Vy2

= − sgn × −g (P10.8b)
dt 2m
dx
= Vx (P10.8c)
dt
dy
= Vy (P10.8d)
dt

V = Vx2 + Vy2 (P10.8e)

where:
ρ is the density of the air
A is the frontal area of the basketball
Cd is the drag coefficient
m is he mass of the basketball
g is the gravitational constant
(Vx , Vy) are the x and y components of the velocity and (x, y) are the hori-
zontal and vertical positions of the basketball
sgn = −1.0 if Vy < 0 and sgn = 1.0 if Vy > 0

Using MATLAB’s ode45 function, determine the x-position of the basketball

when it reaches the height of the hoop. Use the following parameters:

m = 0.623445 kg , ρ = 1.225 kg/m 3 , A = 0.04476 m 2 , Cd = 0.25, g = 9.81 m/s 2 ,

0 ≤ t ≤ 1 s in steps of 0.01 s, and the coordinates of the center of the hoop

( xh , yh ) = (6, 3.048) m , y( 0) = 1.98 m and x (0) = 0.

Hint: MATLAB’s ode45 function should give you x(i), y(i), Vx(i), Vy(i), for
i = 1:length(t). Determine the first i value when y(i) < 3.048 and Vy(i) < 0. Use
that value of i and the one before it to interpolate the x-position when y = 3.048.
Would the basketball hit the rim of the hoop? The rim radius is 0.2286 m.
226 MATLAB® Essentials

Reference
1. Bober, W., Introduction to Numerical and Analytical Methods with MATLAB for
Engineers and Scientists, CRC Press, Boca Raton, FL, 2014.
11
Boundary Value Problems of
Ordinary Differential Equations

11.1 Introduction
When an ordinary differential equation involves boundary conditions
instead of initial conditions, then a numerical approach is most often used
to solve the problem. In a boundary value problem, we essentially need to fit
a solution into the known boundary conditions as opposed to simply inte-
grating from the initial conditions. An example of this type of problem is
the temperature of a bar subjected to known different temperatures at the
ends as it looses heat along the bar by natural convection. Another example
would be the deflection of a beam due to an applied load along the beam
and where the boundary conditions at both ends of the beam are specified.
Another example of this type of problem is the determination of the electric
field between the plates of a capacitor with a known charge density between
the plates and a fixed voltage across the plates. In these three examples, a
solution is found by numerically solving a second-order, nonhomogeneous
ordinary differential equation using finite difference methods.

11.2 Difference Formulas

To numerically solve a boundary value problem involving an ordinary, lin-
ear, and differential equation, we will need the difference formulas obtained
by Taylor series expansion using just a few terms. The finite difference
method first involves subdividing the independent variable domain into N
subdivisions. The finite difference formulas that will be used are tabulated
in Table 11.1.
In the following table, yi is the value of y at position xi and ∆x = xi +1 − xi .

227
228 MATLAB® Essentials

TABLE 11.1
Summary of Finite Difference Formulas for Boundary Value Problems
yi +1 − yi First-order forward difference formula. Usually used for a y ’ boundary
y′i =
∆x condition at the beginning of domain.
yi − yi − 1 First-order backward difference formula. Usually used for a y ’ boundary
y′i =
∆x condition at end of the domain.
y i + 1 + y i −1 − 2 y i Second-order central difference formula for the second derivative in the
y′′i = 2 interior of the domain.
∆x

Example 11.1
In this example we consider a bar having a circular cross section that is subjected to
known temperatures at the two ends and which looses heat along the bar by natural
convection to its surroundings (see Figure 11.1). The governing equations describing
the temperature along the bar is given by the following formula:

d 2T hP
= ( T − T∞ ) (11.1)
dx 2 Ak
where:
T is the temperature of the bar at position x
h is the convective heat transfer coefficient
k is the thermal conductivity of the bar material
P is the bar perimeter
A is the bar cross-sectional area
T∞ is the temperature of the surrounding air

Applying the finite difference formulas to the problem gives the following set of
equations:

Ti + 1 + Ti −1 − 2 Ti hP
= ( Ti − T∞ ) (11.2)
∆x 2 kA

h,T∞

T1 = 200°C x Tn+1 = 20°C

h,T∞

FIGURE 11.1
Bar subjected to different end temperatures and losing heat to the surroundings.
Boundary Value Problems of Ordinary Differential Equations 229

We will assume that the x domain is divided into N subdivisions and that T1 = 200°C
and TN+1 = 20°C. Then the set of equations become

T1 = 200 (1)

hP ∆x 2 hP ∆x 2
T1 + T3 − 2 T2 − T2 = T∞ (2)
kA kA

hP ∆x 2 hP ∆x 2
T2 + T4 − 2 T3 − T3 = T∞ (3)
kA kA

hP ∆x 2 hP ∆x 2
T3 + T5 − 2 T4 − T4 = T∞ (4)
kA kA
.
.
hP ∆x 2 hP ∆x 2
TN −1 + TN + 1 − 2 TN − TN = T∞ (N)
kA kA

TN +1 = 20 (N+1)

The above set of algebraic, linear equations can be solved by using MATLAB ®’s inv
function or MATLAB’s Gauss-Elimination function. In Example 7.3, we discussed
a systematic method for solving this type of problem. The set of equations can be
expressed as the matrix equation AT = C, where

 T1   a1,1 a1, 2  a1, n   c1 

T  a a2 , 2  a2 , n  c 
T =  , A=  , C =  2
2 2 ,1
(11.3)
       
     
Tn   an ,1 an , 2  an , n   cn 

In the above set of equations, the coefficient matrix A is made up of elements ai , j

where the first index is the equation number and the second index is the same as
the index of the unknown temperature, Tj that the ai , j element is associated with. For
example, in Equation 2,

 hP∆x 2 
a2 ,1 = 1, a2 , 3 = 1 and a2 , 2 = −  2 + .
 kA 

In this expression, A is the cross-sectional area of the bar and not the coefficient
matrix, A. From the pattern of the equation set, we can assign all the ai , j terms
within one for loop. This is done in the following program:
 230 MATLAB® Essentials

We will use the following parameters:

= =
k 386 W/m-C, h 60 W/m
= 2
-C, L 0=
.5 m , N 50 , D = 0.2 cm

The following program creates a table and plot of T versus x.

% Example_11_1.m
% This program determines the temperature in a bar having
% different end temperatures and subjected to convective
% heat transfer.
% Units for k are W/m-C, units for h are W/m^2-C
% units for T are C, units for L are m, units for D are cm.
clear; clc;
k=386; h=60; L=0.5; D=0.2e-2;
P=pi*D; A=pi/4*D^2; N=50; dx=L/N;
T(1)=200; T(N+1)=20; Tinf=20;
C1=h/k*P/A^dx^2;
a(1,1)=1; C(1)=200;
a(N+1,N+1)=1; C(N+1)=Tinf;
x=0:dx:L;
for i=2:N
a(i,i-1)=1; a(i,i+1)=1; a(i,i)=-(2+C1); C(i)=C1*Tinf;
end
T=inv(a)*C';
plot(x,T), xlabel('x(m)'), ylabel('T(C)'), grid, title('T vs. x');
fprintf('x(m) T(C) \n')
fprintf('---------------------\n');
for i=1:2:N+1
fprintf('%4.2f %7.2f \n',x(i),T(i));
end
----------------------------------------------------------------------

Program Results:
x(m) T(C)
---------------------
0.00 200.00
0.02 185.00
0.04 171.03
0.06 158.02
0.08 145.91
0.10 134.64
. .
. .
0.40 35.05
0.42 31.55
0.44 28.31
0.46 25.32
0.48 22.55
0.50 20.00
>>

See Figure 11.2.

----------------------------------------------------------------------
Boundary Value Problems of Ordinary Differential Equations 231

T vs. x
200

180

160

140

120
T(C)

100

80

60

40

20
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
x(m)

FIGURE 11.2
Plot of temperature, T versus position x.

Exercises

E11.1. Repeat Example 11.1, but this time replace the boundary condition at
x = L, with
dT
−k (L) = h(T (L) − T∞ ) giving
dx
TN +1 − TN
k + hTN +1 = h T∞
∆x

E11.2. In this exercise, we consider the deflection of a beam subjected

to a uniform load, of weight, w/m (for more details on this subject, see
Example 8.1 in Reference 1. Consider the beam shown in Figure 11.3. The
governing equation for the deflection of a beam is

d 2 y M( x )
= (11.4)
dx 2 EI ( x)
where:
y is the deflection of beam
M is the internal bending moment
E is the modulus of elasticity of beam material
I is the moment of inertia of an area
232 MATLAB® Essentials

w (N/m)

y
x

FIGURE 11.3
Deflection of a uniformly loaded beam.

To obtain the finite difference form of the governing equation, subdivide the
x-axis into N subdivisions, giving x1, x2, x3, …, xN+1.
Let the deflections at these points be: y1, y2, y3, …, yN+1.
The finite difference formula for d 2 y/dx 2 , as shown in Table 11.1 is

d2 y y + y − 2 yn
2
( xn ) = n +1 n −21 (11.5)
dt ∆x

Thus, the governing differential equation becomes

yn+1 + yn−1 − 2 yn Mn
=
∆x 2 EI
or

1 1 M ∆x 2
yn − 1 + yn + 1 − yn = n , for n = 2,3,4,..., N (11.6)
2 2 2EI

The boundary conditions are

y1 = 0 (11.7)

y N +1 = 0 (11.8)

The bending moment Mn for this problem is

wL w xn2
Mn = xn − (11.9)
2 2

Determine the deflection, yi for i = 1:N + 1. Create a plot y versus x and print a
table consisting of yi and xi. Also print out the obtained maximum deflection.
Use the following parameters:

L = 3 m , w = 40 kN/m, EI = 1.5 × 10 3 kN-m 2 , N = 30

Boundary Value Problems of Ordinary Differential Equations 233

Conducting plates

+
Vo D
x
−

Material between
plates with dielectric ε
and charge density ρ(x)

FIGURE 11.4
Parallel plate capacitor with constant applied voltage.

E11.3. Figure 11.4 shows a parallel plate capacitor with constant applied voltage
vo and a fixed charge density ρ between the plates. For cases with planar sym-
metry such as the parallel plate capacitor where the charge density only changes
in the x-direction (i.e., there is no y or z dependency), then Poisson’s equation
describing the electric potential Φ reduces to an ordinary differential equation:

d 2Φ( x) ρ( x) (11.10)
2
=−
dx ε
where:
Φ( x) is the electric potential (in volts)
ρ( x) is the x-dependent charge density (in coul/m3)
ε is the dielectric constant for the material between the plates

We wish to solve for Φ( x) between the plates of the capacitor shown in

Figure 11.4 with a plate separation of D meters, ρ( x) = ρo ( x − D)2 and bound-
ary conditions Φ(0) = 0 and Φ(D) = vo .
Substituting the expression for ρ( x) into Equation 11.10, we obtain

d 2Φ ρ ρ
= − o ( x − D)2 = − o ( x 2 − 2Dx + D2 ) (11.11)
dx 2 ε ε

Equation 11.11 can readily be solved analytically and the solution [2] is

ρo  1 4 D 3 D2 2  εvo D3  
Φ( x) = −  x − x + x − +  x (11.12)
ε  12 3 2  Dρo 4  

Solve Equation 11.11 numerically and compare the numerical solution with
the exact solution. Create a plot of Φ versus x and a table consisting of x, Φ
234 MATLAB® Essentials

obtained numerically, and Φ obtained by Equation 11.12. Use the following

parameters:
N = 40, D = .0004 m , rho = 1 × 10 4 coulomb/m3 ,
epsilon = 1.04e-12, vo = 5 V

Projects
P11.1. For the beam shown in Figure P11.1, determine the beam deflection,
y(x), by the finite difference method utilizing MATLAB’s inverse matrix
function or the MATLAB’s Gauss-Elimination Method. Print the results in a
table format. Also determine the approximate maximum deflection. Use the
following parameters:

P = 10 kN , EI = 1.5e + 3 kN-m 2 , L = 10 m , a = 6 m ,
0 ≤ x ≤ 10.0 m in steps of 0.1 m

For this configuration:

 P ( L − a)
 xi for 0 ≤ xi ≤ a
 L
Mi = 
 P ( L − a ) xi − P ( xi − a) for a < xi ≤ L
 L

P11.2. For the beam shown in Figure P11.2, determine the beam deflection,
y(x), by the finite difference method utilizing MATLAB’s inverse matrix
function or the MATLAB’s Gauss-Elimination method. Print the results in a

P
a

y
x

FIGURE P11.1
Deflection of a beam subjected to a concentrated load.
Boundary Value Problems of Ordinary Differential Equations 235

W2

W1

y
x

FIGURE P11.2
Deflection of a beam subjected to a linear increasing load.

table format. Also determine the approximate maximum deflection. Use the
following parameters:

w1 = 10 kN/m, w2 = 20 kN/m, EI = 1.5e + 3 kN-m2, L = 10 m,

0 ≤ x ≤ 10.0 m in steps of 0.1 m.

Hint: The load can be considered as the sum of a uniform load and a trian-
gular load. For the triangular load, the resultant force equals ( w2 − w1 )L/2
located 2L/3 from the apex of the triangle. This results in the following
equation for M(x):

wL L w x2 x3
M( x) =  1 + (w2 − w1 )  x − 1 − (w2 − w1 )
 2 6 2 6L

References
1. Bober, W., Introduction to Numerical and Analytical Methods with MATLAB for
Engineers and Scientists, CRC Press, Boca Raton, FL, 2014.
2. Bober, W., Stevens, A., Numerical and Analytical Methods with MATLAB for
Electrical Engineers, CRC Press, Boca Raton, FL, 2012.
Appendix: Greek Letters and Special
Characters in MATLAB® Plots

MATLAB® allows the use of Greek and special characters in its plot
headings and labels. The method for doing this is based on the TeX format-
ting language [1] and is summarized in this appendix.
MATLAB provides the functions title, xlabel, ylabel, and text for
adding labels to plots. These labels can include Greek and special characters
by applying the character sequences as shown in Table A.1. These sequences
all begin with the backslash character (\) and can be embedded in any
text string argument to title, xlabel, ylabel, and text. Subscripts
and superscripts may also be applied by using the _ and ^ operators. For
example, the sequence Vo is written as V_o and 106 is written as 10^6. If
the subscripts or superscripts are multiple characters, then use curly braces
to delimit the string to be subscripted, for example, Vout is generated with
V_{out}.

Example A.1
The following MATLAB script shows how to use special characters in a plot:

% Example_A_1.m
% This script shows example usage of special characters in MATLAB plots.
% Plot a 1MHz sine wave over the interval 0<t<2 microsec
t = 0:2e-8:2e-6;
fo = 1e6;
xout = sin(2*pi*fo*t);
plot(t*1e6,xout);
title('Plot of sin(2\pif_{o}t) for f_{o}=10^6 Hz');
xlabel('time (\museconds)');
ylabel('x_{out}(t)');
text(1.5,0.3,'\omega = 2\pi \times f_{o}');
----------------------------------------------------------------------------------------------

The resulting plot is shown in Figure A.1.

237
238 Appendix

TABLE A.1
Greek Letters
Greek Letter Greek Symbol

\alpha α
\beta β
\gamma γ
\delta δ
\epsilon ε
\zeta ζ
\eta η
\theta θ
\vartheta ϑ
\lambda λ
\mu μ
\nu ν
\rho ρ
\sigma σ
\tau τ
\phi ϕ
\omega ω
\Gamma Γ
\Delta Δ
\Phi Φ
\Omega Ω
Appendix 239

Plot of sin(2πfot) for fo = 106 Hz

1

0.8

0.6

0.4
ω = 2π × fo
0.2
xout(t)

−0.2

−0.4

−0.6

−0.8

−1
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
time (μseconds)

FIGURE A.1
Example usage of Greek letters, special characters, subscripts and superscripts in a
MATLAB plot.

Reference
1. Knuth, D.E., The TeXbook, Addison Wesley, New York, 1984.
Review Answers

Review 1.1
1. List several ways engineers use the computer.
a. To solve mathematical models of physical phenomenon.
b. Storing and reducing experimental data.
c. Controlling machines.
d. Communicating with other engineers on a particular project.
2. List several areas of interest for engineers.
a. Designing new products.
b. Improve performance of existing products.
c. Improving manufacturing efficiency.
d. Minimizing costs of production.
e. Minimizing power consumption.
f. Research on developing new products.
3. List several methods that can be used in the design of new products.
a. Full-scale experiments.
b. Small-scale model experiments.
c. A mathematical model describing the phenomenon of interest.
4. Which method mentioned in item 3 is the least expensive?
The mathematical model is the least expensive.
5. List several components of a typical desktop/laptop computer system.
a. Input devices: keyboard, mouse, microphone.
b. Central processing unit consisting of a control unit, an arithmetic
logic unit and registers.
c. Memory and storage unit consisting of main memory which is
used for temporary storage of programs and data.
d. Secondary memory consisting of a hard drive, an optical drive
(a CD or a DVD), and a flash drive.
e. Output devices: monitor, printer, speakers.
f. Operating systems: Windows 10, MacOS, Linux.
6. Name several computer languages used today and in the past by
Engineers.
Fortran, C/C++, MATLAB, Pascal.

241
242 Review Answers

7. What is the lowest level computer language and what numbering

system does it use?
Machine language. It uses the binary system of numbers.
8. For engineers what is the principle advantage of MATLAB® over
some of the other computer programming platforms?
MATLAB has built-in functions that solve many different types of
mathematical problems that other computer platforms do not have.
9. List several items that are recommended in developing a computer
program.
a. List the algebraic equations involved in the project.
b. Create a flow chart or a program outline.
c. Write the program using the list of algebraic equations and the
program outline or flow chart.
d. Run the program and correct any syntax errors.
10. List several items that can be considered building blocks available
in developing a computer program in MATLAB.
a. Assignments (Arithmetic statements).
b. Input/output statements.
c. Loop statements (for loop and while loop).
d. Conditional operatives/alternative path statements (if-else,
if-elseif-else statements).
e. Functions (built-in and self-written).

Review 2.1
1. What are the two alternative ways to start the MATLAB program?
If available, start the MATLAB program by double-clicking on the
MATLAB icon on the Window’s desktop. If not available,
a. For Windows versions earlier than Windows 10, go to the
Window’s Start menu, click on All Programs, find the MATLAB
program among the list of available programs and click on it.
This will open up the MATLAB desktop.
b. For Windows version 10, click on the Windows icon on the left
bottom of the screen and search for the MATLAB program and
click on it. This will open the MATLAB desktop.
2. What are the windows in the MATLAB’s default desktop?
The main windows are the Command Window, Current Folder, and
Workspace.
Review Answers 243

3. It is best to write a MATLAB script (program) in the Editor Window.

From MATLAB’s default desktop how does one open the script window?
To Open the Editor Window, click on the New Script icon in the
Toolstrip in MATLAB’s desktop.
4. After you have completed writing a script in the appropriate win-
dow, what is the next step?
Click on the save icon in the Toolstrip. In the window that opens,
select the folder in which the script is to be saved and in the File Name
box type in the name of the script with the .m extension.
5. Name two ways to execute a script.
a. In the Editor Window, click on arrow located just above the Run
icon in the Toolstrip. In the Editor Window the arrow is green.
b. In the Command Window after the MATLAB prompt (>>) sign,
type in the script name (without the .m extension).
6. What happens if you attempted to execute a script and the script is
not in the folder listed in the current folder Toolbar?
A dialog box will open giving you the choice of changing the folder
listed in the path box or adding the folder containing your script to
the MATLAB path.
7. In MATLAB, what is the file name extension for saved scripts?
The file name extension is .m.
8. How does one establish a comment line in a script?
By placing a % sign in front of a statement makes it a comment line.

Review 2.2
1. List at least two conditions in selecting a name for a variable.
a. Variable name must start with a letter.
b. Variable names can only contain letters, numbers, and the under-
score character.
2. Finish the following statement. An arithmetic statement may
involve … constants, variables, arithmetic operators, and elementary
MATLAB functions and self-written functions.
3. What can be said about the variables that appear on the right side of
an arithmetic statement?
All variables on the right side of an arithmetic statement must be
previously defined (given a value) in the program.
244 Review Answers

4. List the arithmetic operators in MATLAB.

The arithmetic operators in MATLAB are
+ addition
− subtraction
* multiplication
/ division
^ exponentiation
5. What is the order in which an arithmetic statement will be carried out?
First all operations enclosed within parentheses will be carried out
in the following order: exponentiation, multiplication, and divi-
sion, then addition and subtraction. Then proceeding from left to
right, the operations will be carried out in the same order as listed
above.
6. What is MATLAB’s command for
a. π. pi
b. e. exp
c. ln. log
d. sine function in radians. sin()
e. sine function in degrees. sind()
f. sin−1 function. asin()
g. The number of elements in a vector. lenth()
h. The size of a matrix (the number of rows and columns). size()
i. The sum of the elements in a vector. sum()
j. The maximum element in a vector. max()
k. Preallocating the size of a 3 × 3 matrix. zeros(3)
7. What is the purpose of placing a semicolon at the end of a variable
assignment?
Placing a semicolon at the end of a variable assignment avoids the
variable assignment to be echoed to the screen.
8. What is the command that will clear the Command Window?
The command clc clears the Command Window.
9. What is the basic data structure in MATLAB?
The basic data structure in MATLAB is a matrix.
10. Name two functions of the colon operator.
a. The colon operator may be used to create a new matrix from an
existing matrix.
b. The colon operator can also be used to generate a series of numbers.
Review Answers 245

Review 2.3
1. Name two commands that will result in printing to the screen.
a. fprintf()
b. display()
2. What is the command that will move the cursor to the next line?
\n.
3. What is the format that will print a floating point variable to 10
spaces and to three decimal points?
% 10.3f
4. What is the format that will print a floating point variable in scien-
tific notation to 12 spaces and to four decimal points?
% 12.4e
5. What are the commands necessary to print to a file?
a. fo=fopen('file_name','w');
b. fprintf(fo,'format \n',variables);
6. What is the command to create a linear plot of y versus x and what
type of variable must x and y be?
The command is plot(x,y);
Variables x and y must be vectors.
7. What are the commands that will label the x- and y-axis and provide
a title to a plot?
xlabel('x'), ylabel('y'), title('y vs. x')

Review 2.4
1. What is the objective in using a for loop?
The objective of a for loop is to repeat a series of statements with
just a few lines of code.
2. What is the syntax of a for loop?
for index variable = starting value: step size:
final value
3. Should table headings that are not to be repeated be inside a for
loop?
No.
246 Review Answers

4. If the index of a for loop is used to select an element of a vector or

a matrix, what variable type should the for loop index be?
It should be an integer.
5. What other statement type can be used to create a loop?
The while loop.
6. What is the major difference between a for loop and a while loop?
The syntax of the for loop generates its own index. If a program
requires an index, the program containing the while loop must gen-
erate an index.

Review 2.5
1. Name four commands that can be used in a script to input data into
the workspace. Also list where the data are located.
The commands that can be used to enter data into the work space are
load, fscanf, dlmread, and input. In the load, fscanf, and
dlmread commands, the data are located in a separate file, usually
a .txt file. For the input command, the data are entered from the
keyboard by the user.
2. Which of the four commands makes the program interactive?
The program becomes interactive with the input command.
The user is asked to enter values from the keyboard.

Review 2.6
1. When there is more than one function plotted on a graph, what are
the ways to identify which curve goes with which function?
Each curve can be given a different color, or a different line type.
In  each case you can use the legend command to identify which
curve goes with which function. You can also use the text command
to label each of the curves.
2. What is the name of the function that will allow you to plot several
graphs on one page?
The name of the function that will allow you to plot several graphs on
one page is the subplot command. The subplot command is not a plot
command. It is used to position the several different plots on the page.
Review Answers 247

3. How does one enter Greek symbols into a plot?

The Greek symbol is entered with \name of symbol.
4. What are the commands that will allow you to enter text onto a plot
once the plot has been created?
In the plot window, click on the Insert option in the menu bar.
A dropdown menu will appear that contains the following options:
X Label, Y Label, Title, TextBox, and others. Click on the item
that you wish to enter on the plot. If you select the TextBox option,
a crosshair will appear and you can drag it to the location where you
wish to start the text, then type in the text that you want to enter into
your plot.

Review 3.1
1. What statement is frequently used to establish two conditional paths?
The if-else statements.
2. What series of statements is used to establish several conditional paths?
The if-elseif-else statements.
3. List the various types of logic statements that can be used with the
if–else and the if-elseif-else ladder.
a < b, a > b, a ≤ b, a ≥ b, a == b, a ~= b.
4. Is the else statement required with the if-else and with the
if-elseif-else ladder?
No.
5. What statement group and a MATLAB’s function are alternatives to
the if-elseif-else ladder?
The switch statement and MATLAB’s menu function.

Review 3.2
1. If y = 3.0 * A and A is a vector, what can you say about y?
If A = [a1 a2 a3 .... an ], then y = [ 3 × a1 3 × a2 3 × a3 .... 3 × an ].
2. If y = 3.0*sin(x) and x is a vector, what can you say about y.
If x = [x1 x2 x3 ..., xn ], then
y = [3 × sin( x1 ) 3 × sin( x2 ) 3 × sin( x3 ) ..., 3 × sin( xn )]
248 Review Answers

3. If vector C = A + B, what must be true about vectors A and B.

Vectors A and B must be of the same length.
Each element of C will be the addition of the corresponding elements
of A and B.
4. What is the result of the multiplication of two vectors of the same
length, say A and B, and how it should be programmed?
If A = [a1 a2 a3 .... an ] and B = [b1 b2 b3 ... bn ], then the multiplication of
A and B is [ a1 × b1 a2 × b2 a3 × b3 ... an × bn ]
In MATLAB, the multiplication has to be programmed as A.* B.
5. What is the name of MATLAB’s function that does interpolation?
The function name is interp1.
6. What are the inputs to MATLAB’s interpolation function?
MATLAB’s interp1 function has three arguments, say (X,Y,Xi),
where (X,Y) are a set of known (x, y) data points and Xi is the set
of x values at which the set of y values, Yi, are to be determined by
interpolation. Arrays X and Y must be of the same length. If Xi is a
vector, than Yi will be a vector.
7. What are the outputs from MATLAB’s interpolation function?
In the variables described above the output from MATLAB’s interp1
function are the interpolated values Yi.

Review 4.1
1. When does it seem appropriate to write a self-written function?
If you have a complicated program and you wish to break it down into
smaller sections, it is appropriate to write a self-written function. Also,
if you have a program that requires a series of statements to be repeated
several time, it is convenient to place the series of statements in a self-
written function. Finally, many MATLAB functions require the user to
write a self-written function to describe the problem of interest.
2. In writing a self-written function what must be the first word in the
first executable statement?
The first word in the first executable statement in the function must
be function.
3. A self-written function usually has both an input and an output.
Where does the input come from? Where does the output go to?
The input comes from the calling program.
The output from the function goes to the calling program.
Review Answers 249

4. If a self-written function has more than one output, how must the
output be presented?
If a function has more than one output, the output must be in brackets.
5. How does a self-written function communicate with the calling
program?
The self-written function only communicates with the calling pro-
gram through the input and output variables. The exception is when
a global statement is contained in both the calling program and the
function.
6. What can be said about variables in the self-written function that are
not in the input or output arguments of the function and there are no
global statements?
If a variable in the function is not in the input or output arguments
of the function, then that variable is completely independent of vari-
ables in the calling program.
7. Do the variable names in the input and output arguments between
the calling program and the function have to be the same?
No, they only need to be in the same order.
8. If a programmer wishes to write a self-written function, but does
not wish to create an additional .m file, what can the programmer do
and what is the constraint?
The programmer can write an anonymous function, which is included
in the main program and not as a separate .m file. The constraint is
that it needs to be a single statement.

Review 5.1
1. Suppose you wish to assign a column vector consisting of string ele-
ments, what are the conditions that need to be followed in setting up
this column vector?
The conditions are (a) each string row needs to be enclosed by
single quotation marks, (b) each string row must have the same
number of columns, and (c) the entire matrix must be enclosed by
brackets.
2. Suppose that you had a data file that contains both numerical and
text data, what command would you use to read in the data into
your main program?
The command used to read in the data is the textscan command.
250 Review Answers

3. When the command used in reading in the data type described in

item 2, what object type does the data go into?
The data goes into a cell array.
4. To assign variable names to items in the object, which of the follow-
ing three symbols would you use: (), [], {}.
You should use {} symbol.

Review 6.1
1. What is meant by the term root of function f(x)?
The root of a function is the value of x that makes f(x) = 0
2. What is the objective in the search method for determining a root of
the equation f(x) = 0?
The objective of the search method is to find small intervals contain-
ing the roots.
3. What is the name of the MATLAB function for determining the roots
of a transcendental equation of the form f(x) = 0?
The name of MATLAB’s function to obtain the roots of a transcen-
dental equation is fzero.
4. In MATLAB’s function for determining the roots of a transcendental
equation, how does one define the function whose roots are to be
determined?
A self-written function should describe the function whose roots
are to be obtained. The name of this self-written function should be
entered as the first argument in MATLAB’s fzero function.
5. If you suspect that there is more than one real root, what method should
be used in combination with the MATLAB’s function to obtain the roots?
If you suspect that there is more than one root, you should use the
search method, in combination with MATLAB’s fzero function. The
search method is used to obtain a small interval in which a root lies and
MATLAB’s fzero function determines the root that lies in that interval.
6. If you are using the search method in combination with the fzero
function, what can you say about the second argument in the fzero
function?
The second argument to be entered in the fzero function should
be a vector of length 2 specifying the endpoints of the intervals that
contain the roots.
The functional values at the beginning and end of this interval
should differ in sign.
Review Answers 251

7. What is the purpose of the global statement?

Variables listed in the global statement will be common to both the
calling program and the called function. Therefore, variables defined in
the calling program will be available in the called function, despite the
fact that these variables are not input or output arguments in the called
function. Of course, the reverse is also true. An item determined in the
called function would also become available in the calling program.
8. If the function f(x) is a polynomial, what MATLAB function should
you use to obtain its roots?
You should use MATLAB’s roots function. If the polynomial has
complex roots, MATLAB’s roots function will give the complex roots,
whereas MATLAB’s fzero function will only give the real roots.

Review 7.1
1. Given a set of algebraic, linear equations in the form AX = C, where
A is the coefficient matrix and X and C are column vectors, what are
the two ways for solving for X in MATLAB?
a. X = inv(A)*C.
b. X = A\C.
2. Given a large system of algebraic, linear equations of the form
AX = C, describe the recommended approach to solving the system
of linear equations.
First we need to number each equation in the system. We then need
to determine the coefficients, ai , j , in each equation and the ci , where
the i represents the equation number and the j represents the num-
ber of the x variable associated with the coefficient.
Example:
Suppose we had a system of 10 equations requiring a coefficient
matrix, a(10, 10) and a c(10) matrix to solve the problem. Suppose the
7th equation was as follows:

− 0.6 x4 − x5 + x8 + 0.6 x10 = 0 (7)

a(7 , 1) = a(7 , 2) = a(7 , 3) = a(7 , 6) = a(7 , 7) = a(7 , 9) = 0

and a(7 , 4) = −0.6, a(7 , 5) = −1, a(7 , 8) = 1, a(7 , 10) = 0.6, and c(7) = 0

After establishing all a(10, 10) and c(10) values use MATLAB’s inv
or MATLAB’s Gauss-Elimination function to solve the problem, that
is, X = inv(A)*C or X = A\C.
252 Review Answers

Review 8.1
1. Suppose an experiment produced a set of data and we wished to
create an approximating curve, yc , that is a polynomial expression
that best fits the data. What is the name of the MATLAB function
that will do this?
The name of MATLAB’s function that will do this is polyfit(x, y, m),
where (x, y) is the experimental data set and m is the degree of the
polynomial. The polyfit function returns the coefficients of the
polynomial, a1 , a2 ,..., am+1 where

yc = a1 x m + a2 x m−1 + ... am x + am+1

2. After executing MATLAB’s polyfit function, and you wish

to obtain values on the approximating curve, yc , at positions
( x1 , x2 , x3 ,... xn ) what MATLAB function would you use?
You would use MATLAB’s polyval(a,X) function, where vector a
is the coefficients of the approximating function and X is the vector
of the x position values.

Review 9.1
B
1. What is the formula for evaluating the integral, I = ∫A f ( x) dx by the
Simpson’s rule?
First sub-divide the x domain into N equal intervals, where N is an
even number giving x1 , x2 , x3 ,....., xN+1. Then determine the functional
values at the x positions giving f1 , f2 , f3 ,......., fN+1. The formula for the
integral by Simpson's rule is:
xN +1
∆x
I=
∫ f (x)dx ≈
x1
3
[ f1 + 4 f2 + 2 f3 + 4 f4 + 2 f5 +  + 4 fN + fN +1 ]

2. What is the name of MATLAB’s function for integrating a single

variable function?
MATLAB’s function for integrating a single variable function is
integral.
3. In MATLAB’s function for integrating a single variable function how
does one define the function to be integrated?
One needs to write a self-written function that describes the integrand.
Review Answers 253

The name of this function should be entered as the first argument in

the integral function.
4. If the integrand contains nonlinear terms, how must they be treated?
Nonlinear terms need to be entered as element-by-element multipli-
cation or division. Terms involving exponents also need to be treated
as an element-by-element operation.
5. Will MATLAB’s integral function treat improper integrals?
Yes.

Review 9.2
1. What is the name of MATLAB’s function for integrating a two-
dimensional function?
MATLAB’s function for integrating a two-dimensional function is
integral2.
2. List the arguments that go into MATLAB’s function for integrating a
two-dimensional function.
The arguments that go into MATLAB’s function for integrating a
two-dimensional function are
a. FUN(X,Y) which is a function handle for the function that
describes the two-dimensional function to be integrated.
b. XMIN, XMAX, YMIN, YMAX, where
XMIN ≤ X ≤ XMAX and YMIN ≤ Y ≤ YMAX
XMIN and XMAX are scalars and YMIN and YMAX may be scalars
or function handles.
Index

Note: Page numbers followed by f and t refer to figures and tables, respectively.

A difference formulas, 227–235, 228t

overview, 227
Algebraic and transcendental
temperature, plot, 231f
equations, 131–152
break command, 85–89
bisection method, 133, 133f
Built-in functions, 7, 19, 133
MATLAB’s
MATLAB features, commands,
fzero function, 134–139
special items, 21–30
roots function, 139–141
special values, 26–30
overview, 131
trigonometric and other
search method, 132, 132f
functions, 21–26
Algebraic, linear equations, 153–167
with vector arguments, 92–93
gauss-elimination, 161–162
Bytes, 4
method, 154–156
MATLAB’s inv function, 154
number of solutions, 162–163 C
resistive circuit problem, 159–161 Central Processing Unit (CPU), 3
treatment of large systems, 156–159 Characters and strings, 123–128
Algorithm, 3 Colon operator (:), 28–29, 52
ALU (Arithmetic Logic Unit), 3, 19–20 Command Window, 10, 18
Anonymous functions, 110–112 MATLAB, 20
Application Software, 5 pi in, 23
Approximating curve, 178 undock, 23f
with data points, 172f Compound logical expressions, 82
Approximating function, 169 Computer programming, 3
and data points, 174f Conditional operators, 81–92
Arithmetic Logic Unit (ALU), 3, 19–20 break command, 85–89
Arithmetic operators, 20–21 if command, 81–83
in MATLAB, 244 if-elseif-else command, 83–85
Arithmetic procedure, 3 menu function, 90–92
Arithmetic statement, 19–20 switch command, 89–90
Assignment operator, 19–21 Control flow directives, 7
Control Unit, 3
CPU (Central Processing Unit), 3
B
Cubic splines, 178–182
Backslash character (\), 68, 237 Current folder toolbar
Back substitution, 161–162 and window, 10–11
Bar charts, 63–65, 65f Curve fitting, 169–185
Bisection method, 132–133, 133f cubic splines, 178–182
Bits, 4 MATLAB’s, 179–182
Boundary value problems, 205, 227 with exponential function, 174–178
finite difference formulas, 228t function, MATLAB’s, 169–174
ODE, 227–235 overview, 169

255
256 Index

D global function, 136

Greek letters and mathematical
Data types, 6
symbols, 68
dblquad function, MATLAB, 194
Desktop/laptop computer system,
components, 3–4 H
Dialog box, 14–15, 14f
changing folder/path, 16f Hard drive, 4
disp command, 31 help integral, command
dlmread command, 52 window, 190
Dynamic Random Access Memory Help window, 16f, 17f
(DRAM) devices, 3 hold on command, 59–61

E I

Editor window, 11, 12f, 13f, 38f if command, 81–83

Equivalent triangular set, 161 if-elseif-else command, 83–85
Exponential, square root, and error if-elseif ladder illustration, 125
functions, 25 Improper integrals, 190
Initial value problems, 205
and MATLAB’s ODE, 206–214
F Input and output arguments, 106
figure command, 55–57 input command, 54–55, 107
Finite difference Input devices, 3
formulas Input, MATLAB, 50–55
boundary value problems, 228t dlmread command, 52
different end temperatures and fscanf command, 52–53
losses heat, 228, 228f input command, 54–55
methods, 227 load command, 50–51
Flash drive, 4 Input/output (“I/O”)
fminbnd funtion, 105, 113–114 commands, 7
fopen function, 32, 53 integral2 function, 194–204
for loop, 36–43, 88 self-written integrand
syntax, 36 function, 194
fprintf command, 31–32 integral function, 190–193
fscanf command, 52–53 interp1 function, 93–95
FUN argument, 113 Inverse trigonometric functions,
Function handle, 110 24–25
FUN function, 134, 190, 194
fzero function, 133–139
L
MATLAB, 145, 149–151
statement, 136 legend command, 58
syntax, 134 Linear interpolation, 85–86, 93
Linear plot, 34–36
load command, 50–51
G
Logical expressions, 82
Gauss-Elimination, 154, 161–162 Loops, MATLAB, 36–50
method, 154–156 for loop, 36–43
program, 162 while loop, 43–50
Index 257

M template, 106f
textscan, 127–130
Machine language, 5
fundamentals, 9–80
Main Memory, 3
assignment operator, 19–21
MATLAB, 9
input, see Input, MATLAB
arithmetic operators, 244
loops, 36–50; see also Loops,
backslash operator, 154
MATLAB
colon operator (:), 28–29
overview, 9–10
command window, 20
simple plot commands, 34–36
computer programming, 1–8
variable names and types, 18–19
building blocks, 7
graphics, 55–69
computer programming, 3
bar charts, 63–65, 65f
computer usage, 1–2
figure command, 55–57
desktop/laptop computer system,
greek letters and mathematical
components, 3–4
symbols, 68
mathematical model, 2
hold on command, 59–61
methodologies, 6
interactively annotating plots, 69
needs, 5–6
multiple plots, 57–59
curve-fitting function, 169–174
pie charts, 65–68, 66f
cubic spline, 179–182
plotyy command, 62
desktop, 10–13
saving plots, 69
PLOTS tab, 34
subplot command, 63
windows, 11f, 12f
help window, 17f
features, commands, special items,
input command, 50
and built-in functions, 21–30
matrix, preallocation, 29
special values, 26–30
menu push button, 82f
trigonometric and other
ode45 function, 206, 215, 217
functions, 21–26
computer program, 220, 223
function(s), 26–27
syntax, 206
additional, 139
ODE, initial value problem, 206–214
built-in, 19, 133
“=” operator, 44
built-in ode45, 205
output, 30–34
cubic spline, 179
disp command, 31
dblquad, 194
fprintf command, 31–32
fminbnd, self-written function
printing to file, 32–34, 33f, 34f
and, 105–122
plots, greek letters and special
fzero, 131, 134–139, 145
characters in, 237–239,
Gauss-Elimination, 229, 234
238t, 239f
global, 136
program, 143, 149, 166–167
integral, 190–193
programming languages, 6–7
integral2, 194–204
script, 13–17, 14f, 15f, 237
interp1, 93–95, 179
version R2016A and R2016B, 54
inv, 154, 156, 229, 234
Matrix algebra, 153–154, 157
menu, 90–92
Memory cells, 4
myfun, 136
Memory/Storage Unit, 3
polyfit, 175, 184
menu function, 90–92
polyval, 170, 184
Moler, Cleve, Dr., 5
pulse(t), 224
mse function, 170
roots, 131, 139–141
258 Index

N Q
Network interface, 4 quad function, 190
Newton–Raphson method, 132
Numerical integration, 187–204
R
improper integrals, 190
MATLAB’s Registers, 3
integral2 function, 194–204 Resistive circuit problem, 159–161
integral function, 190–193 roots function, 133, 139–141
of ODE, 205–225 syntax, 139
initial value problem and
MATLAB’s, 206–214
S
overview, 205
and Simpson’s rule, 187–189 Scalar and vector operations, 96–99
Numerical methods, 2 addition, 96
element-by-element
operations, 96–98
O
scalar times vector multiplication, 96
ODE, see Ordinary differential two vector
equations (ODE) addition/subtraction, 96
ode45 function, 206, 219 functions operations, 98–99
ODEFUN function, 206 Search method, 132, 132f
Operating system (OS), 4 Secondary Memory, 4
Operators, 7 Self-written function, 105–110
Optical drive, 4 and MATLAB®’s fminbnd
Ordinary differential equations function, 105–122
(ODE), 205, 227 anonymous functions, 110–112
boundary value problems, 227–235 Self-written integrand function, 194
difference formulas, 227–235, 228t Simple plot commands, 34–36
overview, 227 Simpson’s rule, numerical integration
temperature, plot, 231f and, 187–189, 188f
numerical integration, 205–225 size() command, 154
initial value problem and spline function/method, 179
MATLAB’s, 206–214 sprintf command, 67
overview, 205 String specifiers, 128
OS (operating system), 4 subplot command, 63
Output devices, 4 Subscripts and superscripts, 237
switch command, 89–90
System software, 5
P
Pie charts, 65–68, 66f
T
plotyy command, 62
polyfit function, 169, 184 TeX formatting language, 237
MATLAB’s, 175 textscan function, 127–128
Polynomial regression, 169 Toolstrip, 10, 13–14
Programming languages, 26 open icon, 33
MATLAB, 6–7 print command, 32
overview, 5 save and run icon, 15f
Index 259

Transcendental equations, exponential, square root,

algebraic and, 131–152 and error functions, 25
bisection method, 133, 133f functions, 24
MATLAB’s inverse, 24–25
fzero function, 134–139 special values, 23
roots function, 139–141 TSPAN, time interval, 206
overview, 131
search method, 132, 132f
W
Trigonometric and other
functions, 21–26 while loop, 43–50
complex number, 26 Workspace window, 11