Experience Seamless Full Ebook Downloads for Every Genre at textbookfull.

com

Computational Techniques for Process Simulation

and Analysis Using MATLAB® 1st Edition Niket S.
Kaisare

https://textbookfull.com/product/computational-techniques-
for-process-simulation-and-analysis-using-matlab-1st-
edition-niket-s-kaisare/

OR CLICK BUTTON

DOWNLOAD NOW

Explore and download more ebook at https://textbookfull.com

 Recommended digital products (PDF, EPUB, MOBI) that
you can download immediately if you are interested.

An Introduction to Reservoir Simulation Using MATLAB GNU

Octave User Guide for the MATLAB Reservoir Simulation
Toolbox MRST Lie
https://textbookfull.com/product/an-introduction-to-reservoir-
simulation-using-matlab-gnu-octave-user-guide-for-the-matlab-
reservoir-simulation-toolbox-mrst-lie/
textboxfull.com

Practical Guide for Biomedical Signals Analysis Using

Machine Learning Techniques A MATLAB Based Approach 1st
Edition Abdulhamit Subasi
https://textbookfull.com/product/practical-guide-for-biomedical-
signals-analysis-using-machine-learning-techniques-a-matlab-based-
approach-1st-edition-abdulhamit-subasi/
textboxfull.com

Numerical Methods for Engineers and Scientists Using

MATLAB® Ramin S. Esfandiari

https://textbookfull.com/product/numerical-methods-for-engineers-and-
scientists-using-matlab-ramin-s-esfandiari/

textboxfull.com

Network Modeling Simulation And Analysis In MATLAB Theory

And Practices Dac-Nhuong Le

https://textbookfull.com/product/network-modeling-simulation-and-
analysis-in-matlab-theory-and-practices-dac-nhuong-le/

textboxfull.com
Chemical thermodynamics for process simulation Gmehling

https://textbookfull.com/product/chemical-thermodynamics-for-process-
simulation-gmehling/

textboxfull.com

Advances in Computational Techniques for Biomedical Image

Analysis: Methods and Applications 1st Edition Deepika
Koundal
https://textbookfull.com/product/advances-in-computational-techniques-
for-biomedical-image-analysis-methods-and-applications-1st-edition-
deepika-koundal/
textboxfull.com

Signals and systems : analysis using transform methods and

MATLAB Third Edition Michael J. Roberts

https://textbookfull.com/product/signals-and-systems-analysis-using-
transform-methods-and-matlab-third-edition-michael-j-roberts/

textboxfull.com

Numerical Methods for Engineering: An introduction using

MATLAB® and computational electromagnetics examples
(Electromagnetic Waves) 2nd Edition Karl F. Warnick
https://textbookfull.com/product/numerical-methods-for-engineering-an-
introduction-using-matlab-and-computational-electromagnetics-examples-
electromagnetic-waves-2nd-edition-karl-f-warnick/
textboxfull.com

Power system transient analysis : theory and practice

using simulation programs (ATP-EMTP) 1st Edition Arai

https://textbookfull.com/product/power-system-transient-analysis-
theory-and-practice-using-simulation-programs-atp-emtp-1st-edition-
arai/
textboxfull.com
 Computational Techniques for
Process Simulation and Analysis
Using MATLAB®
 Computational Techniques for
Process Simulation and Analysis
Using MATLAB®

Niket S. Kaisare
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: 978-1-1387-4608-4 (Paperback)

International Standard Book Number-13: 978-1-4987-6211-3 (Hardback)

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

Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or utilized in any
form by any electronic, mechanical, or other means, now known or hereafter invented, including 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
To Pradnya for her unending love, patience, and support
and
To my parents, Komal and Satish Kaisare
Contents

Preface, xix
Author, xxiii

Chapter 1   ◾    Introduction 1

1.1 OVERVIEW 1
1.1.1 A General Model 1
1.1.2 A Process Example 2
1.1.3 Analysis of Dynamical Systems 3
1.2 STRUCTURE OF A MATLAB® CODE 3
1.2.1 Writing Our First MATLAB® Script 5
1.2.2 MATLAB Functions
® 7
1.2.3 Using Array Operations in MATLAB ® 9
1.2.4 Loops and Execution Control 10
1.2.5 Section Recap 11
1.3 APPROXIMATIONS AND ERRORS IN NUMERICAL METHODS 12
1.3.1 Machine Precision 12
1.3.2 Round-Off Error 14
1.3.3 Taylor’s Series and Truncation Error 15
1.3.4 Trade-Off between Truncation and Round-Off Errors 18
1.4 ERROR ANALYSIS 20
1.4.1 Convergence and Stability 20
1.4.2 Global Truncation Error 21
1.5 OUTLOOK 23

vii
viii   ◾    Contents

Section I  Dynamic Simulations and Linear Analysis

Chapter 2   ◾    Linear Algebra 27

2.1 INTRODUCTION 27
2.1.1 Solving a System of Linear Equations 27
2.1.2 Overview 28
2.2 VECTOR SPACES 30
2.2.1 Definition and Properties 30
2.2.2 Span, Linear Independence, and Subspaces 32
2.2.3 Basis and Coordinate Transformation 34
2.2.3.1 Change of Basis 34
2.2.4 Null (Kernel) and Image Spaces of a Matrix 35
2.2.4.1 Matrix as Linear Operator 35
2.2.4.2 Null and Image Spaces in MATLAB ® 39
2.3 SINGULAR VALUE DECOMPOSITION 41
2.3.1 Orthonormal Vectors 41
2.3.2 Singular Value Decomposition 42
2.3.3 Condition Number 47
2.3.3.1 Singular Values, Rank, and Condition Number 47
2.3.3.2 Sensitivity of Solutions to Linear Equations 47
2.3.4 Directionality 51
2.4 EIGENVALUES AND EIGENVECTORS 54
2.4.1 Orientation for This Section 54
2.4.2 Brief Recap of Definitions 54
2.4.3 Eigenvalue Decomposition 56
2.4.4 Applications 58
2.4.4.1 Similarity Transform 62
2.4.4.2 Linear Differential Equations 63
2.4.4.3 Linear Difference Equations 64
2.5 EPILOGUE 65
EXERCISES 67

Chapter 3   ◾    Ordinary Differential Equations: Explicit Methods 69

3.1 GENERAL SETUP 69
3.1.1 Some Examples 69
3.1.2 Geometric Interpretation 72
 Contents   ◾    ix

3.1.3 Euler’s Explicit Method 74

3.1.4 Euler’s Implicit Method 76
3.1.5 Stability and Step-Size 78
3.1.5.1 Stability of Euler’s Explicit Method 78
3.1.5.2 Error and Stability of Euler’s Implicit Method 79
3.1.6 Multivariable ODE 80
3.1.6.1 Nonlinear Case 81
3.2 SECOND-ORDER METHODS: A JOURNEY THROUGH THE WOODS 82
3.2.1 Some History 82
3.2.2 Runge-Kutta (RK-2) Methods 83
3.2.2.1 Derivation for RK-2 Methods 83
3.2.2.2 Heun’s Method 84
3.2.2.3 Other RK-2 Methods 86
3.2.3 Step-Size Halving: Error Estimate for RK-2 87
3.2.4 Richardson’s Extrapolation 89
3.2.5 Other Second-Order Methods (*) 91
3.2.5.1 Trapezoidal Rule: An Implicit Second-Order Method 91
3.2.5.2 Second-Order Adams-Bashforth Methods 92
3.2.5.3 Predictor-Corrector Methods 92
3.2.5.4 Backward Differentiation Formulae 93
3.3 HIGHER-ORDER RUNGE-KUTTA METHODS 93
3.3.1 Explicit Runge-Kutta Methods: Generalization 93
3.3.2 Error Estimation and Embedded RK Methods 97
3.3.2.1 MATLAB® Solver ode23 100
3.3.3 The Workhorse: Fourth-Order Runge-Kutta 101
3.3.3.1 Classical RK-4 Method(s) 102
3.3.3.2 Kutta’s 3/8th Rule RK-4 Method 103
3.4 MATLAB® ODE45 SOLVER: OPTIONS AND PARAMETERIZATION 103
3.5 CASE STUDIES AND EXAMPLES 105
3.5.1 An Ideal PFR 106
3.5.1.1 Simulation of PFR as ODE-IVP 106
3.5.1.2 Numerical Integration for PFR Design 108
3.5.1.3 Comparison of ODE-IVP with Integration 110
3.5.2 Multiple Steady States: Nonisothermal CSTR 111
3.5.2.1 Model and Problem Setup 111
x   ◾    Contents

3.5.2.2 Simulation of Transient CSTR 113

3.5.2.3 Step Change in Inlet Temperature 115
3.5.3 Hybrid System: Two-Tank with Heater 116
3.5.4 Chemostat: Preview into “Stiff ” System 120
3.6 EPILOGUE 125
EXERCISES 125

Chapter 4   ◾    Partial Differential Equations in Time 127

4.1 GENERAL SETUP 127
4.1.1 Classification of PDEs 128
4.1.2 Brief History of Second-Order PDEs 128
4.1.3 Classification of Second-Order PDEs and Practical Implications 129
4.1.3.1 Elliptic PDE 129
4.1.3.2 Hyperbolic PDE 130
4.1.3.3 First-Order Hyperbolic PDEs 131
4.1.3.4 Parabolic PDE 132
4.1.4 Initial and Boundary Conditions 132
4.2 A BRIEF OVERVIEW OF NUMERICAL METHODS 133
4.2.1 Finite Difference 133
4.2.2 Method of Lines 134
4.2.3 Finite Volume Methods 134
4.2.4 Finite Element Methods 135
4.3 HYPERBOLIC PDE: CONVECTIVE SYSTEMS 135
4.3.1 Finite Differences in Space and Time 136
4.3.1.1 Upwind Difference in Space 136
4.3.1.2 Forward in Time Central in Space (FTCS) Differencing 138
4.3.1.3 Lax-Friedrichs Scheme 139
4.3.1.4 Higher-Order Methods 139
4.3.2 Crank-Nicolson: Second-Order Implicit Method 140
4.3.2.1 Preview of Numerical Solution 141
4.3.3 Solution Using Method of Lines 141
4.3.3.1 MoL with Central Difference in Space 142
4.3.3.2 MoL with Upwind Difference in Space 145
4.3.4 Numerical Diffusion 149
 Contents   ◾    xi

4.4 PARABOLIC PDE: DIFFUSIVE SYSTEMS 150

4.4.1 Finite Difference in Space and Time 152
4.4.2 Crank-Nicolson Method 153
4.4.3 Method of Lines Using MATLAB ODE Solvers
® 154
4.4.3.1 MoL with Central Difference in Space 154
4.4.4 Methods to Improve Stability 157
4.5 CASE STUDIES AND EXAMPLES 157
4.5.1 Nonisothermal Plug Flow Reactor 157
4.5.2 Packed Bed Reactor with Multiple Reactions 164
4.5.3 Steady Graetz Problem: Parabolic PDE in Two Spatial Dimensions 170
4.5.3.1 Heat Transfer in Fluid Flowing through a Tube 170
4.5.3.2 Effect of Velocity Profile 174
4.5.3.3 Calculation of Nusselt Number 174
4.6 EPILOGUE 176
EXERCISES 177

Chapter 5   ◾    Section Wrap-Up: Simulation and Analysis 179

5.1 BINARY DISTILLATION COLUMN: STAGED ODE MODEL 181
5.1.1 Model Description 181
5.1.2 Model Equations and Simulation 183
5.1.3 Effect of Parameters: Reflux Ratio and Relative Volatility 185
5.2 STABILITY ANALYSIS FOR LINEAR SYSTEMS 186
5.2.1 Motivation: Linear Stability Analysis of a Chemostat 187
5.2.1.1 Phase Portrait at the Steady State 190
5.2.1.2 Trivial Steady State and Analysis 190
5.2.2 Eigenvalues, Stability, and Dynamics 191
5.2.2.1 Dynamics When Eigenvalues Are Real and Distinct 192
5.2.2.2 An Example 197
5.2.2.3 Summary 197
5.2.3 Transient Growth in Stable Linear Systems 198
5.2.3.1 Defining Normal and Nonnormal Matrices 198
5.2.3.2 Analysis of Nonnormal Systems 199
5.3 COMBINED PARABOLIC PDE WITH ODE-IVP: POLYMER CURING 201
xii   ◾    Contents

5.4 TIME-VARYING INLET CONDITIONS AND PROCESS DISTURBANCES 208

5.4.1 Chemostat with Time-Varying Inlet Flowrate 208
5.4.2 Zero-Order Hold Reconstruction in Digital Control 212
5.5 SIMULATING SYSTEM WITH BOUNDARY CONSTRAINTS 215
5.5.1 PFR with Temperature Profile Specified 216
5.6 WRAP-UP 219
EXERCISES 219

Section II  Linear and Nonlinear Equations and Bifurcation

Chapter 6   ◾    Nonlinear Algebraic Equations 225

6.1 GENERAL SETUP 225
6.1.1 A Motivating Example: Equation of State 226
6.2 EQUATIONS IN SINGLE VARIABLE 227
6.2.1 Bisection Method 228
6.2.2 Secant and Related Methods 233
6.2.2.1 Regula-Falsi: Method of False Position 235
6.2.2.2 Brent’s Method 235
6.2.3 Fixed Point Iteration 236
6.2.4 Newton-Raphson in Single Variable 238
6.2.5 Comparison of Numerical Methods 240
6.3 NEWTON-RAPHSON: EXTENSIONS AND MULTIVARIATE 241
6.3.1 Multivariate Newton-Raphson 241
6.3.2 Modified Secant Method 245
6.3.3 Line Search and Other Methods 247
6.4 MATLAB® SOLVERS 249
6.4.1 Single Variable Solver: fzero 249
6.4.2 Multiple Variable Solver: fsolve 250
6.5 CASE STUDIES AND EXAMPLES 253
6.5.1 Recap: Equation of State 253
6.5.2 Two-Phase Vapor-Liquid Equilibrium 253
6.5.2.1 Bubble Temperature Calculation 254
6.5.2.2 Dew Temperature Calculation 254
6.5.2.3 Generating the T–x–y Diagram 255
6.5.3 Steady State Multiplicity in CSTR 257
 Contents   ◾    xiii

6.5.4 Recap: Chemostat 261

6.5.5 Integral Equations: Conversion from a PFR 262
6.5.5.1 First-Order Kinetics 263
6.5.5.2 Complex Kinetics 266
6.6 EPILOGUE 268
EXERCISES 271

Chapter 7   ◾    Special Methods for Linear and Nonlinear Equations 273
7.1 GENERAL SETUP 273
7.1.1 Ordinary Differential Equation–Boundary Value Problems 274
7.1.2 Elliptic PDEs 274
7.1.3 Outlook of This Chapter 275
7.2 TRIDIAGONAL AND BANDED SYSTEMS 275
7.2.1 What Is a Banded System? 275
7.2.1.1 Tridiagonal Matrix 276
7.2.2 Thomas Algorithm a.k.a TDMA 276
7.2.2.1 Heat Conduction Problem 277
7.2.2.2 Thomas Algorithm 281
7.2.3 ODE-BVP with Flux Specified at Boundary 285
7.2.4 Extension to Banded Systems 288
7.2.5 Elliptic PDEs in Two Dimensions 289
7.3 ITERATIVE METHODS 290
7.3.1 Gauss-Siedel Method 291
7.3.2 Iterative Method with Under-Relaxation 295
7.4 NONLINEAR BANDED SYSTEMS 296
7.4.1 Nonlinear ODE-BVP Example 296
7.4.1.1 Heat Conduction with Radiative Heat Loss 297
7.4.2 Modified Successive Linearization–Based Approach 298
7.4.3 Gauss-Siedel with Linearization of Source Term 302
7.4.4 Using fsolve with Sparse Systems 304
7.5 EXAMPLES 304
7.5.1 Heat Conduction with Convective or Radiative Losses 304
7.5.2 Diffusion and Reaction in a Catalyst Pellet 305
xiv   ◾    Contents

7.5.2.1 Linear System and Thiele Modulus 305

7.5.2.2 Langmuir-Hinshelwood Kinetics in a Pellet 308
7.6 EPILOGUE 311
EXERCISES 311

Chapter 8   ◾    Implicit Methods: Differential and Differential Algebraic

Systems 313
8.1 GENERAL SETUP 313
8.1.1 Stiff System of Equation 313
8.1.1.1 Stiff ODE in Single Variable 315
8.1.2 Implicit Methods for Distributed Parameter Systems 316
8.1.3 Differential Algebraic Equations 316
8.2 MULTISTEP METHODS FOR DIFFERENTIAL EQUATIONS 317
8.2.1 Implicit Adams-Moulton Methods 318
8.2.2 Higher-Order Adams-Moulton Method 319
8.2.3 Explicit Adams-Bashforth Method 320
8.2.4 Backward Difference Formula 322
8.2.5 Stability and MATLAB Solvers® 325
8.2.5.1 Explicit Adams-Bashforth Methods 325
8.2.5.2 Implicit Euler and Trapezoidal Methods 325
8.2.5.3 Implicit Adams-Moulton Methods of Higher Order 325
8.2.5.4 BDF/NDF Methods 325
8.2.5.5 MATLAB® Nonstiff Solvers 326
8.2.5.6 MATLAB® Stiff Solvers 326
8.3 IMPLICIT SOLUTIONS FOR DIFFERENTIAL EQUATIONS 327
8.3.1 Trapezoidal Method for Stiff ODE 327
8.3.1.1 Adaptive Step-Sizing 329
8.3.1.2 Multivariable Example 330
8.3.2 Crank-Nicolson Method for Hyperbolic PDEs 331
8.3.2.1 Exploiting Sparse Structure for Efficient Simulation 337
8.4 DIFFERENTIAL ALGEBRAIC EQUATIONS 337
8.4.1 An Introductory Example 338
8.4.1.1 Direct Substitution 338
8.4.1.2 Formulating and Solving a DAE 339
 Contents   ◾    xv

8.4.2 Index of a DAE and More Examples 340

8.4.2.1 Example 2: Pendulum in Cartesian Coordinate System 341
8.4.2.2 Example 3: Heterogeneous Catalytic Reactor 341
8.4.3 Solution Methodology: Overview 342
8.4.3.1 Solving Algebraic Equation within ODE 343
8.4.3.2 Combined Approach 345
8.4.4 Solving Semiexplicit DAEs Using ode15s in MATLAB® 348
8.5 CASE STUDIES AND EXAMPLES 351
8.5.1 Heterogeneous Catalytic Reactor: Single Complex Reaction 351
8.5.2 Flash Separation/Batch Distillation 353
8.6 EPILOGUE 359
EXERCISES 360

Chapter 9   ◾    Section Wrap-Up: Nonlinear Analysis 363

9.1 NONLINEAR ANALYSIS OF CHEMOSTAT: “TRANSCRITICAL”
BIFURCATION 364
9.1.1 Steady State Multiplicity and Stability 364
9.1.2 Phase-Plane Analysis 365
9.1.3 Bifurcation with Variation in Dilution Rate 366
9.1.4 Transcritical Bifurcation 368
9.2 NONISOTHERMAL CSTR: “TURNING-POINT” BIFURCATION 372
9.2.1 Steady States: Graphical Approach 372
9.2.2 Stability Analysis at Steady States 374
9.2.3 Phase-Plane Analysis 376
9.2.4 Turning-Point Bifurcation 377
9.3 LIMIT CYCLE OSCILLATIONS 379
9.3.1 Oscillations in Linear Systems 379
9.3.2 Limit Cycles: van der Pol Oscillator 381
9.3.2.1 Relaxation vs. Harmonic Oscillations 382
9.3.3 Oscillating Chemical Reactions 383
9.4 SIMULATION OF METHANOL SYNTHESIS IN TUBULAR REACTOR 387
9.4.1 Steady State PFR with Pressure Drop 388
9.4.1.1 Reaction Kinetics 388
9.4.1.2 Input Parameters and Initial Processing 389
9.4.1.3 Steady State PFR Model 391
9.4.2 Transient Model 394
xvi   ◾    Contents

9.5 TRAJECTORY OF A CRICKET BALL 398

9.5.1 Solving the ODE for Trajectory 399
9.5.2 Location Where the Ball Hits the Ground 400
9.5.3 Animation 403
9.6 WRAP-UP 405
EXERCISES 405

Section III  Modeling of Data

Chapter 10   ◾    Regression and Parameter Estimation 409

10.1 GENERAL SETUP 409
10.1.1 Orientation 410
10.1.2 Some Statistics 411
10.1.3 Some Other Considerations in Regression 413
10.2 LINEAR LEAST SQUARES REGRESSION 413
10.2.1 Fitting a Straight Line 413
10.2.2 General Matrix Approach 415
10.2.3 Goodness of Fit 418
10.2.3.1 Maximum Likelihood Solution 418
10.2.3.2 Error and Coefficient of Determination 419
10.3 REGRESSION IN MULTIPLE VARIABLES 423
10.3.1 General Multilinear Regression 423
10.3.2 Polynomial Regression 424
10.3.3 Singularity and SVD 428
10.4 NONLINEAR ESTIMATION 429
10.4.1 Functional Regression by Linearization 429
10.4.2 MATLAB® Solver: Linear Regression 432
10.4.3 Nonlinear Regression Using Optimization Toolbox 434
10.5 CASE STUDIES AND EXAMPLES 437
10.5.1 Specific Heat: Revisited 437
10.5.2 Antoine’s Equation for Vapor Pressure 438
10.5.2.1 Linear Regression for Benzene 439
10.5.2.2 Nonlinear Regression for Ethylbenzene 440
 Contents   ◾    xvii

10.5.3 Complex Langmuir-Hinshelwood Kinetic Model 441

10.5.3.1 Case 1: Experiments Performed at Single
Concentration of B 442
10.5.3.2 Case 2: Experiments Performed at Different Initial
Concentrations of B 444
10.5.4 Reaction Rate: Differential Approach 445
10.6 EPILOGUE 448
10.6.1 Summary 448
10.6.2 Data Tables 449
EXERCISES 450

APPENDIX A: MATLAB® PRIMER, 451

APPENDIX B: NUMERICAL DIFFERENTIATION, 475

APPENDIX C: GAUSS ELIMINATION FOR LINEAR EQUATIONS, 485

APPENDIX D: INTERPOLATION, 499

APPENDIX E: NUMERICAL INTEGRATION, 511

BIBLIOGRAPHY, 527

INDEX, 529
Preface

S tudents today are expected to know one or more of the several computing or simu-
lation tools as part of their curriculum, due to their widespread use in the industry.
MATLAB® has become one of the prominent languages used in research and industry.
MATLAB is a numerical computing environment that is based on a MATLAB scripting
language. MathWorks, the makers of MATLAB, describe it as “the language of technical
computing.” The focus of this book will be to highlight the use of MATLAB in technical
computing or, more specifically, in solving problems in the analysis and simulation of pro-
cesses of interest to engineers.
This is intended to be an intermediate-level book, geared toward postgraduate students,
practicing engineers, and researchers who use MATLAB. It provides advanced treatment
of topics relevant to modeling, simulation, and analysis of dynamical systems. Although
this is not an introductory MATLAB or numerical techniques textbook, it may however
be used as a companion book for introductory courses. For the sake of completeness, a
primer on MATLAB as well as introduction to some numerical techniques is provided in
the Appendices. Since mid-2000s,we have always used MATLAB in electives in IIT Madras.
The popularity of MATLAB among students led us to start a core undergraduate (sopho-
more) and a postgraduate (first-year masters) laboratory. Since 2016, I have started teach-
ing a massive open online course (MOOC) on MATLAB programming on the NPTEL
platform.* The first two years of this course had over 10,000 enrolled students. Needless to
say, MATLAB has become an important tool in teaching and research. The focus of all the
above courses is to introduce students to MATLAB as a numerical methods tool. Some of
the students who complete these courses inquire about the next-level courses that would
help them apply MATLAB skills to solve engineering problems. This book may also be used
for this purpose. In introductory courses, a significant amount of time is spent in develop-
ing the background for numerical methods itself. In our effort to make the treatment gen-
eral and at a beginner’s level, we eschew real-world examples in favor of abstracted ones.
For example, we would often introduce a second-order ODE using a generic formulation,
such as y″ + ay′ + b(y − c) = 0. A sophomore who hasn’t taken a heat transfer course may
not yet appreciate a “heating in a rod” problem. An intermediate-level text means that it
is more valuable to use a real example, such as T″ + r−1T′ + β(T − Ta) = 0. The utility of such

* NPTEL stands for National Programme for Technology Enhanced Learning and is a Government of India−funded initiative
to bring high-quality engineering and science courses on an online (MOOC) platform to enhance students’ learning.

xix
xx   ◾    Preface

an approach cannot be understated, since it allows the freedom to introduce some of the
complexity that engineers, scientists, and researchers face in their work.
The value of using real-world examples was highlighted during my experience in indus-
trial R&D, where we used MATLAB extensively. We needed to interface with cross-func-
tional teams: engineering, implementation, and software development teams. Individuals
came from a wide range of backgrounds. These interactions exposed me to a new experi-
ence: Your work must be understood by people with very different backgrounds, who may
not speak the same technical language. The codes had to bridge the “language barrier”
spoken in different teams, and the codes were to be combined with a reasonably intuitive
interface. I have tried to adopt some of these principles in this book, without moving too far
from the more common pedagogy in creating such a book.
Thus, a practically oriented text that caters to an intermediate-level audience is my objec-
tive in writing this book.

ORIGIN OF THIS BOOK

There are several excellent books on numerical techniques for engineers. Laurene
Forsell’s book, Numerical Methods Using MATLAB, provides a MATLAB-based
approach to learning numerical techniques. The books on numerical techniques by
Chapra and Canale and by S.K. Gupta are excellent undergraduate textbooks, which
introduce undergraduates to this subject for the first time. Thus, their focus is concep-
tual understanding of numerical techniques themselves. While undergraduate teaching
is in good stead, a textbook that covers core requirements for a balanced postgraduate
curriculum is missing. Such a book will also be useful to practicing engineers, scien-
tists, and researchers who use MATLAB.
This book is borne out of my experience in teaching a postgraduate course called
Process Analysis and Simulation, postgraduate lab in Process Simulation, and theory of
computational techniques. They provide the first-year postgraduates the basis to tackle
research problems in their theses. The former course takes a balanced focus on model-
ing, simulation, and analysis of chemical process systems, while the process simulation
lab gives them a numerical methods perspective. Postgraduate-level books, such as the
evergreen Numerical Recipes by Press et al., are rather advanced and focused on numeri-
cal methods. On the other hand, the book by Strogatz on nonlinear dynamics or other
similar books are not general enough for the needs of an audience interested in simula-
tions. A “bridge” book, which assumes some familiarity with undergraduate material, but
still covers the basics, is missing.
Having said this, I do not intend this to be a postgraduate numerical methods text. This
book aims to introduce students and practitioners to simulation and analysis of process
systems in MATLAB. We often find it difficult to connect the numerical tool to the physical
analysis of a system. This book intends to bring in a strong process simulation treatment to
linear stability and nonlinear analysis.
Thus, this book intends to bring a practical approach to expounding theories: both
numerical aspects of stability and convergence, as well as linear and nonlinear analysis of
 Preface   ◾    xxi

systems. The “process” is the focus. Numerical methods are introduced insofar as is essen-
tial to make a judicious choice of algorithms for simulation and analysis.

PREREQUISITES
Since this is a postgraduate-level text, some familiarity with an undergraduate-level
numerical techniques or an equivalent course is assumed, though we will review all the
relevant concepts at the appropriate stage. So, the students are not expected to remem-
ber the details or nuances of “Newton-Raphson” or “Runge-Kutta” methods, but this
book is not the first time they hear these terms.
Some familiarity with coding (MATLAB, Fortran, C++, Python, or any language)
will be useful, but not a prerequisite. MATLAB primer is provided in the Appendix for
first-time users of MATLAB. Finally, with respect to writing MATLAB code, I focus on
“doing it right the first time” approach—by bringing in good programming practices
that I have learnt over the years. Things like commenting and structuring your code,
scoping of variables, etc., are also covered, not as an afterthought but as an integral part
of the discussion. However, these are dealt with more informally than a “programming
language” course.

HOW THIS BOOK IS LAID OUT

This book derives examples from three different courses I have taught: (i) Numerical
Methods, (ii) Process Analysis and Simulation, and (iii) Computational Programming
Lab. It is structured so that it may be used for any of the three courses. Each chapter
deals with one approach to solving computational problems (e.g., ODEs, PDEs, nonlin-
ear equations, etc.), culminating in case studies that utilize the concepts discussed in the
chapter.
I have split the book into three sections, which are laid out with a “Process Analysis”
viewpoint: Section I covers system dynamics and linear system analysis; Section II covers
solving nonlinear equations, including differential algebraic equations (DAEs); and Section
III covers function approximation and optimization for modeling of data. The following
table summarizes the various chapters in the book:

Basics Chapter 1 Appendix A

Introduction MATLAB® Primer
Appendices Appendix B Appendix C Appendix E
Differentiation Linear Equations Integration
Section I. Chapter 2 Chapter 3 Chapter 4 Chapter 5
Dynamics Linear Algebra ODE-IVP Transient PDEs Simulation
Section II. Chapter 6 Chapter 7 Chapter 8 Chapter 9
Equations Nonlinear Equations Special Methods Implicit Methods Nonlinear Analysis
(ODE-BVP/PDE) (DAEs)
Section III. Appendix D Chapter 10
Data Fitting Interpolation Regression
xxii   ◾    Preface

LAYOUT FOR PROCESS ANALYSIS

The layout of this book is largely based on the postgraduate-level process simulation and
analysis course. The material I cover in this course is chronologically as laid out in the book.
The course starts with an introduction to the role of simulation and analysis in engineering,
and a primer on MATLAB. Thereafter, I introduce concepts in linear algebra (Chapter 2),
ODEs (Chapter 3), and solving hyperbolic and parabolic PDEs (Chapter 4). Problems in
either linear analysis or dynamical simulations (Chapter 5) typically form mid-term proj-
ects for students. The second part of the course also follows a similar structure, with nonlin-
ear equations (Chapter 6), ODE-BVPs and elliptic PDEs (Chapter 7), and DAEs (Chapter 8)
providing the adequate background for end-semester projects involving nonlinear analysis
and bifurcation (Chapter 9). I have added Chapter 10 (Parameter Estimation) for the sake
of completion.

LAYOUT FOR NUMERICAL METHODS

This book may also be used for an advanced numerical methods course. In such a case,
I suggest treating the material column-wise. This course may start with the first row to
cover the basics (Introduction, MATLAB Primer, Differentiation, Integration, and Linear
Equations). Thereafter, Chapter 2 and Chapter 6 may be covered, to equip students to solve
linear and nonlinear equations. Chapters 3 and 7 cover ODE-IVP and ODE-BVP, respec-
tively, followed by Parameter Estimation (Chapter 10). A four-credit course may also cover
PDEs (Chapter 4). Typically, Chapters 5 and 9 will be beyond the scope of such a course.

LAYOUT FOR NUMERICAL DIFFERENTIAL EQUATIONS

The shaded chapters (Chapters 3, 4, 5, 7, 8, and 9), along with appendices on numerical
differentiation and integration, can form a numerical differential equations course.

FOR PRACTICING ENGINEER OR NEW RESEARCHER

A practicing engineer or researcher can embark on a self-guided journey through case
studies and examples covered in this book. This includes not only the case studies analyzed
in Chapters 5 and 9 but also the ones discussed in other chapters (penultimate section in
the other chapters).

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
Author

Dr. Niket S. Kaisare is currently an associate professor in the Department of Chemical

Engineering at IIT Madras. He received his PhD in chemical engineering from Georgia
Institute of Technology, working in the area of model-based advanced process control.
He then joined the Department of Chemical Engineering at the University of Delaware,
where he worked on multiscale modeling of reacting flows in microreactors, as a postdoc-
toral researcher. After this, he joined IIT Madras as assistant professor in 2007. While in IIT
Madras, he taught several courses in process analysis and simulation, computational tech-
niques, process simulation laboratory, and advanced control theory. MATLAB was used
extensively in most of these courses. He has also taught an online course called “MATLAB
Programming for Numerical Computations” as a part of NPTEL (National Programme for
Technology Enhanced Learning). This course was popular, with more than ten thousand
students enrolling in it.
He spent three years, from mid-2011 to 2014, in industrial R&D. During this stint, he
worked on numerous simulation problems related to modeling of vehicle catalytic conver-
tors, cryogenic hydrogen storage, monitoring and control of oil and gas wells, and auto-
mation engineering. As a part of the R&D team, he used MATLAB extensively and spent
a significant part of his time interfacing with engineering and development teams.
He has extensive experience working in MATLAB and FORTRAN as well as simulation
softwares Fluent and Comsol. He also has good working experience with various other
simulation tools, such as Aspen-Plus/Unisim, Gaussian, and Abacus. His current research
program is focused on “multiscale modeling, analysis, and control of catalytic microreac-
tors for energy- and fuel-processing applications.”

xxiii
 CHAPTER 1

Introduction

1.1 OVERVIEW
1.1.1 A General Model
This book is targeted toward postgraduate students, senior undergraduates, researchers, and
practicing engineers to provide them with a practical guide for using MATLAB® for process
simulation and numerical analysis. MATLAB was listed among the top ten programming
languages by the IEEE Spectrum magazine in 2015 (a list that was topped by Java, followed by
C and C++). While the basics of MATLAB can be learnt through various sources, the focus
of this book is on the analysis and simulation of processes of interest to engineers.
The terms “analysis” and “simulation” are generic terms that define a rather broad spectrum
of problems and solution techniques. Engineering is a discipline that deals with the transfor-
mation of raw material, momentum, or energy. Thus, this book will focus on those process
examples where the variables of interest vary with time and/or space, including the relationship
of these state variables with their properties. I will use an example of a reactor-separator process
in Section 1.1.2 to illustrate this. While this is a chemical engineering example, the treatment in
this book is general enough for other engineering and science disciplines to also find it useful.
The problems mentioned above that are considered in this book include ordinary and
partial differential equations (ODEs and PDEs), algebraic equations (either linear or non-
linear), or combinations thereof. The three sections of this book are organized based on the
computational methodology and analysis tools that will be used for the respective problems.
Section I of this book includes Chapters 2 through 5 and deals with ODE-IVPs (initial
value problems) as well as the problems that can be converted into a standard form that can
be solved with ODE-IVP tools. A generic ODE-IVP is of the type
dy
= f ( t ,y ;f
f) (1.1)
dt

where
t is an independent variable
y ∈ R n is a vector of dependent variables
ϕ represents parameters
1
2   ◾    Computational Techniques for Process Simulation and Analysis Using MATLAB®

Examples include simulations of level and temperature in a stirred tank, simulations of a

reactor, mass-spring-damper, pendulum (in cylindrical coordinates), and others.
Section II of this book includes Chapters 6 through 9 and deals with the problems of
the type

0 = g ( x;f
f) (1.2)

where
x ∈ R m is a vector of dependent variables
ϕ represents parameters

Nonlinear algebraic equations, such as Equation 1.2, fall under this category. Moreover,
ODE-BVPs (boundary value problems) and several PDEs are also converted into the form
of Equation 1.2. Section II will not only cover techniques to solve algebraic equations but
also expound methods to convert ODEs/PDEs to this form. A combination of Equations 1.1
and 1.2, called differential algebraic equations (DAEs), is covered in Chapter 8. Chapters 5
and 9 are the concluding chapters of the first two sections. They build on the concepts from
the preceding chapters in the respective sections for the analysis of dynamical systems and
provide an introduction to advanced topics in simulations.
Finally, Chapter 10, included in Section III, deals with the parameter estimation prob-
lem, that is, to compute the parameter vector, ϕ, that best fits the experimental data.

1.1.2 A Process Example

I will use an example from a typical chemical process to motivate the discussion. Figure 1.1
shows a flow sheet of a typical process with a recycle. A reaction of the type A → B takes
place in the plug flow reactor (PFR); the two species are separated in the distillation column;
product B is obtained as the bottoms product, whereas the lighter species A is recycled back.
The PFR is modeled using the following ODE:

dx A
F = -r ( x A ) , x A V =0 = xin (1.3)
dV

Ffeed Fr, xr P, xp

D, xD Purge
Fin, xin

F, xf

F, xf

B, xB

FIGURE 1.1 A typical process consisting of a reactor and a separator, with a recycle.
 Introduction   ◾    3

The reactor outlet conditions are obtained by solving the ODE-IVP above. ODE-IVP prob-
lems are covered in Chapter 3 of this book. If a dynamic response of the PFR is required,
the resulting model is a PDE, where the state variable of interest varies in both space and
time. Solutions to transient PDEs are covered in Chapter 4. Advanced topics in simulation
are presented in Chapter 5, for example, when the inlet conditions or model parameters
vary with time and/or space.
The distillation column consists of N nonlinear algebraic equations in N unknowns (mole
fractions on each tray). For example, one of the model equations for the kth tray is given by

ax i
0 = ( Li -1 xi -1 - Li xi ) + (Vi +1 yi +1 - Vi yi ) where yi = (1.4)
1 + ( a - 1) xi

Such balance equations are written for each ideal stage of the distillation column, resulting
in N nonlinear algebraic equations that need to be solved simultaneously to obtain N vari-
ables. These are further discussed in Chapter 6.
Axial dispersion is neglected while deriving the model (Equation 1.3). Inclusion of the
axial dispersion term converts this IVP to a BVP, which is covered in Chapter 7. Discretizing
the ODE-BVP results in a set of equations with a special matrix structure. Mass transfer
limitations result in DAEs, which are covered in Chapter 8.

1.1.3 Analysis of Dynamical Systems

There is an equal amount of focus on the analysis of dynamical systems. To this end,
Chapter 2 revisits concepts in linear algebra. I present a more contemporary treatment of
linear algebra. Chapter 5 presents transient analysis of dynamical systems and their linear
stability behavior. In addition to stability and dynamics based on eigenvalues of the linear
dynamical system, the topic of transient growth in nonnormal systems is discussed. Related
concepts of directionality and analysis using eigenvalue and singular value decompositions
are discussed.
Chapter 9, which wraps up Section II of this book, is dedicated to nonlinear analysis
and bifurcation. Well-known examples of stirred reactor, chemostat, mass-spring-damper
system, and van der Pol oscillator will be used in this chapter. Chapter 10 is devoted to an
important topic of parameter estimation.
Furthermore, advanced topics in efficient simulation and analysis are also presented.
These include handling time-varying inputs and boundary constraints (Chapter 3), com-
bination of ODE and PDE (Chapter 5), and a fun example of tracking the trajectory of a
cricket ball (or baseball) with simulation and animation.
Before we get to these examples, I will review good practices and structuring of MATLAB
codes as well as errors in numerical approaches.

1.2 STRUCTURE OF A MATLAB® CODE

A brief introduction to some programming practices specific to MATLAB is covered in
this section. The intention of this section is to introduce the reader to good MATLAB
programming practices, rather than “Introduction to MATLAB.” A basic primer on using
4   ◾    Computational Techniques for Process Simulation and Analysis Using MATLAB®

MATLAB is instead provided in Appendix A. This book follows the principle of “learn it
right the first time.” Good programming hygiene, in writing MATLAB codes, is evange-
lized and implemented right from the first example. The book follows another principle
that the best way to learn programming is through extensive practice. MathWorks, the
parent company that develops MATLAB, has good introductory video tutorials, available
at: http://in.mathworks.com/products/matlab/videos.html.* A beginner may want to start
with their “Getting Started” videos.† I also have an introductory MOOC course on using
MATLAB for numerical computations on National Programme for Technology Enhanced
Learning.‡
Figure 1.2 shows a screenshot of MATLAB window. The main section contains two
­windows: MATLAB editor at the top and MATLAB command window at the bottom. The
MATLAB editor currently shows the MATLAB file firstFlowSheet.m, which is a
“driver script” to simulate the reactor-separator flow sheet described above. Line number
13 shows the following statement:

[F,x,err] = solveFlowSheet(Ffeed,Vpfr,purge,initVal);

FIGURE 1.2 Screenshot of MATLAB® window.

* Last accessed on October 13, 2015.

† http://in.mathworks.com/videos/getting-started-with-matlab-68985.html.

‡ http://nptel.ac.in/courses/103106118/ (Last accessed on June 16, 2017).

Exploring the Variety of Random
Documents with Different Content
 Fig. 158
The use of emblematic hardware involves the use of special
designs and patterns, and thus entails a considerably greater cost
than the use of standard patterns. In Fig. 157 are shown several
standard ornamental designs, while in Fig. 158 are shown pieces of
lock trim ornamented with heraldic, or emblematic, designs.

HARDWARE SPECIFICATIONS
 119. The methods of specifying builders’ hardware differ
considerably. Some architects generalize and use the same form of
specification for buildings of all classes, while others are more
specific, and itemize all the hardware for the entire operation, giving,
besides the description, the catalog number and finish.
The first method is not considered good practice, and, though it
involves little work on the part of the architect, it rarely produces the
best results. Thus, wherever possible, the specification should be
complete and comprehensive, giving a full description of all the
hardware in the building.
120. Formerly, builders’ hardware could only be obtained in a few
styles, and there was not much choice regarding its physical
construction and mechanical operation, so that its selection could be
left to the contractor or builder. Specification, therefore, usually
included little more than a mere statement that the necessary
hardware should be furnished and that it should be of good quality.
In recent years, however, the revolution accomplished in the
designing and making of builders’ hardware has elevated it to an
important place in decorative art, and, simultaneously, the creating of
new and higher mechanical grades has radically changed the
requirements in specifications relating to this subject.
121. A few standard forms of hardware specifications that should
secure good results will now be considered. For convenience of
reference, these several forms are marked I, II, and III. The first two
forms are probably as good general forms of hardware specifications
as can be written, but, while they answer the purpose in some
instances, they should not be used where it is intended to have the
general contract include all the hardware, or where it is possible to
prepare an itemized specification such as that given in Form III. The
numbering of the paragraphs in these forms is continuous with this
Section, but in practice they generally start with 1.

FORM I
(See General Conditions)
 Note.—Under the head of General Conditions preceding all
specifications furnished to contractors by the architect, a series of
binding requirements, reservations, and stipulations are specifically
stated, and it is most essential that the contractor carefully peruse
them, as he is bound by them as well as by the clauses under the
heading of Hardware Specifications.
122. Rough Hardware.—Provide all the rough hardware, such as
nails, screws, sash weights, pulleys, chain or cord, anchors, screw
bolts, and all other material in this line necessary for the completion of
the operation.
123. Finishing, or Builders’, Hardware.—All fastenings and
metal trimmings used on doors, windows, transoms, closets, cabinets,
pantries, etc. will be furnished by the owner and delivered at the
building in the quantities and at the times reasonably needed by the
contractor, he to apply the same under the direction and to the
satisfaction of the architect.
The contractor is to be responsible for all hardware after delivery
and until the completion of the building. He shall hang all doors,
properly fit all locks, etc., and return them to their original packages
until after completion of the painting or finishing, when he shall place
them permanently. All knobs shall be covered with Canton flannel, to
protect them from injury, and all keys are to be cared for until the
building is delivered to the owner. The contractor shall place all keys in
their locks or deliver them to the architect with tags attached,
indicating where they belong.
The contractor shall furnish the manufacturer or dealer furnishing
the hardware with details of woodwork or information that may be
necessary in order to understand the requirements and to harmonize
the hardware with the cabinetwork, and, where interferences are
discovered, to have them adjusted before the hardware is delivered.

FORM II
(See General Conditions)
124. Rough Hardware.—Same as in Form I.
 125. Finishing, or Builders’, Hardware.—The contractor shall
include in his estimate the sum of _____ dollars ($____) for the
purchase of the finishing hardware, which is to be selected by the
architect or owner and charged to the contractor. If the hardware costs
less than the stipulated amount, the difference shall be deducted from
his contract price; if it costs more than herein contemplated, the owner
will pay to the contractor the additional amount.
The above hardware is to include all fastenings and metal
trimmings used on doors, windows, transoms, closets, cabinets,
pantries, etc., and will be delivered at the building in the quantities and
at the times reasonably required by the contractor, he to apply the
same under the direction and to the satisfaction of the architect.

FORM III
(See General Conditions)
126. Front-Entrance Doors.—Provide for the front-entrance
doors, cylinder, mortise, front-door lock and three 5" × 5" loose-pin,
ball-tip butts for each door; also flush extension bolts 12 inches and
24 inches long.
127. Vestibule Doors.—Provide for the vestibule doors, cylinder,
vestibule latch, keyed similar to the front door, and three 5" × 5" loose-
pin, ball-tip butts; also flush bolts, the same as specified for entrance
doors.
128. Side-Entrance and Rear Doors.—Provide for the side-
entrance and rear doors, locks and butts the same as specified for the
front-entrance doors. Exterior basement and kitchen doors to have 5-
inch, three-bolt or 4¼-inch, three-tumbler lock and mortise bolt; also
three 4½" × 4½" butts for each door.
129. Sliding-Doors.—Provide for all sliding-doors, Coburn trolley
hangers; double doors to have locks with astragal fronts, and single
doors to have flat-front locks.
130. Double-Acting Doors.—Provide for the doors where
indicated double-acting in the plans, Bommer spring hinges or floor
pivots of suitable size; also push plates, door holders, dead lock, and
kick plates.
131. Miscellaneous Doors.—Provide for all other doors on first
floor, 4¼-inch mortise lock with metal knobs and combined
escutcheons; also three 4" × 4" loose-pin, ball-tipped butts for each
door.
132. Doors on Second Floor.—Provide for all the doors on the
second floor, 4¼-inch mortise locks with metal knobs and combined
escutcheons; chamber-entrance and bathroom doors to have in
addition a mortise bolt. Also provide for these doors, three 4" × 4"
loose-pin, ball-tipped butts for each door.
133. Doors in Attic and Basement.—Provide for the doors in attic
and basement, 3½-inch mortise locks, pottery knobs, and
escutcheons; also two 3½" × 3½" butts for each door.
134. Transoms.—All transoms throughout are to be hinged at the
top or the bottom or pivoted (as indicated in the plans), and are to
have ⁵/₁₆-inch transom lifts to extend to within 5 feet of the floor.
135. Double-Hung Windows.—The double-hung windows on the
first floor are to have approved sash locks and two flush lifts; those on
the second floor, in the attic, and in the basement are to have
approved sash locks and two flush or hook lifts to each sash. All
window stops are to be secured with flush adjusting stop-screws
placed not more than 14 inches apart.
136. Casement Windows and Hinged Sash.—Provide for the
casement in the lavatory and on the second floor, 3" × 3" tight-joint
butts and suitable casement fasteners and adjusters. The basement
sash are to be hinged at the top with wrought butts and are to have
bolts and hooks and eyes to hold them open.
137. French Windows.—Provide for the French windows in the
dining room, 4" × 4" butts, three to each window sash. Also provide for
these windows, flush bolts, 12 inches and 24 inches on standing leaf,
and mortise turnbuckles.[1]
138. Bookcases and China Closets.—Provide for the doors of
the bookcases and china closets, 3" × 3" ball-tip butts, two to each
door; also locks and keyplates or mortise latches and knobs, and bolts
for double doors. All drawers to have drop drawer pulls, two to each
drawer over 20 inches wide.
139. Pantries, Dressers, and Linen Closets.—The doors of the
pantries, dressers, and linen closets are to be provided with 3" × 3"
butts, together with elbow catches or surface bolts for double doors.
Also provide rim cupboard catches or mortise latches with knobs. All
drawers over 20 inches wide are to have two drawer pulls. Flour bins
are to be hinged at the side or at the bottom, so as to tilt, and are to
be supplied with 3" × 3" wrought butts and bar pull or with large T-
handled cupboard turn. Place in each closet one (or two) rows of
brass-plated wire coat-and-hat hooks (or, bronze-plated cast coat-and-
hat hooks) to extend around all sides; these hooks are to be placed
not over 12 inches nor closer than 8 inches apart. Furnish rubber-
tipped, wooden-base knobs for all doors.
140. Quality and Finish.—All of the above hardware is to be of
Blank & Co.’s manufacture, except butts for interior doors, which are
to be the Stanley Work product. All the hardware on the first floor,
except kitchen and service portion, is to be plain, cast-bronze metal of
the design known as ____, and ____ finish. The hardware of the
second floor is to be of steel, in plain design and ____ finish. The
hardware of the bathroom is to be of plain bronze metal, nickel-plated,
while the basement, kitchen, and attic hardware is to be plain steel in
____ finish.
141. Miscellaneous Hardware.—The contractor shall furnish and
apply all hardware necessary to complete the building, under the
direction and subject to the approval of the architect.
142. The preceding specification is intended to meet the
requirements of a residence, but may be changed so as to apply to
other buildings by adding to the several items or by omitting some of
them. In following the specifications just given, the specification writer
should exercise great care in selecting the hardware required, using
the best that can be obtained with the money to be expended. In all
instances, the name, design, quality, and finish desired should be
stipulated.
 SELECTION, ESTIMATION, AND
APPLICATION OF HARDWARE

143. No other material entering into the construction of a building

will pay a larger return in satisfaction, comfort, and permanent
economy for the time and care devoted to its selection than the
finishing hardware. The range of choice and quality has become too
great for the selection to be left to the general contractor under an
omnibus specification, and the practice of according to it the benefit of
careful and discriminating selection by the architect or client, or both in
consultation, is rapidly becoming general. Hardware has become a
factor of utmost importance in the interior decorative scheme of the
modern building, and its selection demands at least the care accorded
to other elements of interior decoration.
The method that should be employed in the selection and
purchase of hardware depends primarily on the existing conditions. If
the building is intended for the purpose of sale or renting, or if
cheapness is the dominating factor, then a competitive method may
be expedient; but if the client intends it for personal occupancy or for
permanent ownership, then competitive bidding is sometimes not
productive of the best results.
In all cases, a preliminary examination should be made of the
makes and grades of the commercial hardware available, and the
amount that it is desired to expend on this item should be decided on.
The process of selection is greatly facilitated by the elimination of
articles outside of the grades selected. When the maker’s grade of
goods has been decided on, the next step is to select in detail the
articles, designs, and finishes to be used in each room, floor, or
division of the building, making such notes of the decisions so reached
as will be convenient in framing the hardware specification, or
schedule.
144. Allowance for Finishing Hardware.—Architects are
constantly called on to prepare preliminary estimates on proposed
buildings, to assist clients to determine whether they can afford to
build. For this purpose, the architect possesses an approximate
knowledge of the cost of excavation, masonry, woodwork, etc., so that
the necessary expenditure may be computed with a fair degree of
accuracy. Builders’ hardware, however, does not readily admit of
accurate preliminary estimating, and in many cases the architect
merely specifies the sum to be expended for finishing hardware,
stating that selections are to be made by himself or his client later, as
the work progresses. This system is found in many instances to be
satisfactory to all—architect, client, and builder.
The cost of the hardware usually bears a fairly constant ratio to the
total cost of various types of buildings. The following figures, which are
based on experience, indicate the range in this ratio under ordinary
conditions. This schedule gives the ratio of cost of finishing hardware
to the total cost of the building, land excluded.
Type of Building Per Cent.
Hotels, large 1.00 to 1.5
Hotels, small 1.50 to 2.0
Apartment houses 1.50 to 2.0
Office buildings 1.00 to 2.0
Office buildings, fireproof .50 to 1.5
Public buildings 1.50 to 2.0
Libraries .75 to 1.5
Hospitals .50 to 1.0
Residences, city 1.50 to 3.0
Residences, country 2.00 to 4.0

145. Taking Off Hardware.—The compilation of the estimate

schedule of finishing hardware for a building must be made from the
architect’s plans and specifications. Therefore, where possible, this
work should be done by a hardware expert or salesman, or by some
person who has the ability to read drawings easily and accurately and
who is capable of exercising thoroughness in every detail of the work.
The first step should be to study the specifications relating to the
general work, and especially that portion relating to hardware and
carpenter work, to see whether the latter embodies facts that affect
the hardware. If such is the case, notes should be made of the fact,
and in many cases, by following this rule, miscellaneous information
can be gathered that will be invaluable to the contractor and will
prevent much doubt as to requirements. The drawings should then be
examined, to gain familiarity with the building, the arrangement of the
interior, and other details. Having thus acquired a general
understanding of the whole subject, the compilation of the quantities
may begin.
146. The list of hardware should be compiled in a systematic
manner, beginning at a definite point in the building and progressing
through the several rooms and floors in a definite order. This will
insure the inclusion of every part of the building and the careful
consideration of every opening or other place where hardware is
required. For example, in the case of a residence, it is customary to
commence at the front entrance, including the front and vestibule
doors, passing thence to the hall, taking each room on the first floor in
due order, passing to the second floor and taking each room on it in
like order, and so on until each floor is covered. The attic and
basement are usually left until last, because they require a simpler
class of goods. The same general system can be followed in the case
of hotels, office buildings, apartment houses, etc., the essential point
being that an orderly method be followed. The rooms should always
be taken in natural sequence, and all openings of similar character
totaled and again counted on the drawing, so that the chances of
omission are minimized.
147. In the case of doors, it is necessary to make note of the size
and thickness, the hand, and the bevel, or rabbet, as required, using
in all cases three butts on all exterior doors, and also three butts on all
interior doors 7 feet or more high. The size of the butts should in all
instances be sufficient to cause the door to clear the trim. The width of
stiles and the general construction of the doors should be noted,
paying particular attention to the front-entrance doors, as architects
sometimes design these regardless of the hardware that is to be used
on them. Frequently, a careful scaling and scrutinizing of elaborately
designed doors will show a shelf effect returning back to the lock stile,
covering nearly half its width, just about where the lock should be
placed. Such defects can be remedied easily, if observed in time, by
directing the architect’s attention to them.
In the case of sliding-doors, it should be noted whether they are
single or double, what is to be the character of the hanger, or rail, and
the length of the run, and whether the door has a flat or a half-round
astragal. The width of the stile and the size of the friction strip should
also be noted, so that the hardware will not conflict with the woodwork,
molding, etc., as, where necessary, special backsets can be procured
on sliding-door locks to overcome this trouble.
Where double-acting doors are shown, always take note of the
thickness, width, and height, also of the width of the lock stile and the
height of the bottom rail, so that, if push plates or kick plates are
required, the suitable dimensions for them can readily be estimated.
The dead locks for these doors should be ordered with oval fronts for
single doors and oval fronts and strikes for double doors, to allow for
neat fitting to the rounded edge of the doors. Where kick plates are
required for double doors of this character, flush bolts should be
mortised into the edge of the door, so that the lower bolt will not
interfere with the kick plates.
148. In stating the information for windows, full details are needed
as to whether they are single- or double-hung, French casement,
stationary, or ordinary casement. Double-hung windows 20 inches or
less in width require only one sash lift. Extra-wide windows, such as
those over 36 inches, should have either bar lifts or two extra-heavy
hook sash lifts. The window stops should be studied in detail, so as to
estimate the number of screws that will be required for each window.
In the case of French windows, it is well to determine the height, the
width of the stile, whether the joints are rabbeted, beveled, or flat, the
hand, and whether the windows swing in or out; detail cross-sections
through head-jambs, casing, and lock stiles should also be procured if
possible. In casement sash, the pivoting should be noted, so as to
determine whether it is vertical or horizontal; the swing should also be
observed, and if the sash are to be hinged, determine whether the
hinges should be placed at the top, the bottom, or the sides. As in the
consideration of doors, the thicknesses and sizes of all casements
should always be noted. Sections through the jambs, trim, and stiles
should be taken at the points where the hardware is to be applied.
149. In order to determine the kind and length of the lifters to be
used on the transoms, it is necessary to note the distance from the
floor to the center of each transom, also whether they are to be
pivoted or hinged from the bottom or the top, and how they are to
swing. The cross-section through side jambs and casings should
always be ascertained, in order to see that the hardware will be
suitable to meet the conditions of the trim.
150. It is necessary to learn all details of the pantry, including the
thickness and height of doors, whether they are double or single, and
the number of each kind. Note all drawers; those under 20 inches in
width require only one pull. The details of the flour bins should be
examined to determine whether they are to be hinged at the bottom
and are to tilt, or whether they are to be curved on plan and are to be
hinged at the side, as each condition requires different hardware.
151. Where unavoidable interferences are discovered, or where
the plans and specifications are obscure or defective, a note of the
facts should be made, and when all such matters have been collected,
they should be submitted to the architect for his decision as a basis for
final determination.
152. Application of Hardware.—Owing to the fact that hardware
has become more intricate in its manufacture and therefore more
complicated in its construction, it is found that the proper amount of
intelligence is not always exercised in its application. Frequently it is
applied so carelessly that the hardware practically loses its most
essential features. In the case of mortise-lock sets, all of the better
grades are made with easy springs, so that the door will close gently
and surely. However, this one feature alone is sometimes obliterated
by careless application. The trouble in this direction is principally
caused by the carpenter fitting the lock into the door untrue, and then
screwing the escutcheons to the doors out of plumb and alinement.
This causes the knob shanks, when rotated, to bind, or stick, in the
sleeve, or collar, of the escutcheon, thereby holding back the latch bolt
after the latch is withdrawn into the lock case. It is also found that
proper adjustment is seldom given to the knobs and spindles passing
through the escutcheons. In many cases when the hardware is
applied carelessly, the owner, not knowing the real cause of the
trouble, condemns the best hardware as being cheap and
unsatisfactory.
In the application of cylinder locks for either front doors or office
doors, it is found that a great amount of inexperience is shown by the
average mechanic. In some cases, the locks are inverted, with the
cylinder below the knobs, when it is designed to be above; also,
sometimes, in reversing a front-door lock with night work, good
judgment is not always used.
153. Such troubles as those just described can be easily
overcome, and if the methods of applying hardware that follow are
strictly adhered to, satisfactory operation will be assured.
Reversing the ordinary mortise, or rim, lock requires no special
ability on the part of the workman, as the operation consists of merely
taking off the cap and turning over the latch bolt. Where, however, a
mortise lock is to be applied, care must be exercised so that it will
work properly. First fit the lock into the door at the proper height from
the floor, making the mortise large enough for the lock to enter easily.
Bore the holes for the knob and the key, and insert the screws in the
lock face and drive them home. Now take the escutcheons, or roses,
and knobs and apply them to the door, carefully adjusting the knobs
and spindles so that there is only slight play in the length. Then true
up both escutcheons, by holding one with the thumb and the other
with the fingers. When this much is accomplished, turn the knobs right
and left to see whether they work properly; if they bind in the collar,
adjust the escutcheons a trifle, vertically or horizontally, until the knobs
rotate back to the original position in both directions without binding.
By using a scratch awl and punch carefully, the exact center of each
screw hole may be located, and by holding the awl perfectly true the
screws will be started true. The escutcheons may now be released
and the screws driven home. If this method is carefully followed,
workmen will find that the latch bolt will work properly in almost every
case. Always try the knob, when the application is completed, by
rotating both to the right and to the left, and if found to bind, always
make the proper adjustment before leaving the work.
154. In fitting strikes to the jambs, it is suggested that the workman
locate them as low as possible, to conform to the locks, leaving the
margin in the strike below the lock bolts. Doors in new buildings are
liable to sag slightly, and if proper allowance is not made, the strikes
may have to be readjusted, sometimes even before the contract is
completed.
 It is also necessary for the carpenter or superintendent to caution
the painters not to paint or varnish any hardware, especially the lock
faces, as this will also retard the operation of the latch bolt. As
previously stated, the hardware should be fitted to the work and then
replaced in the original packages, so as to protect it from abuse and
damage, applying it permanently only after the painting is completed.
155. In applying cylinder, or front-door, locks, the conditions are
still more difficult to overcome. When reversing cylinder locks that
have swivel spindles, unscrew the cap and reverse the latch bolt; also,
reverse the hubs, that is, the parts of the lock through which the
spindles pass. Cylinder locks should always be mortised into the door
at the proper height from the floor, and the holes for the cylinders and
knob spindles then made. Insert the lock into the mortise, but do not
put screws into the face; then apply the escutcheons and adjust the
knob, taking care that the swivel of the spindle centers at the joint in
the split hub. Now proceed to screw in the cylinder to its full extent, so
that the cylinder collar is held rigidly between the cylinder and the
escutcheon, and the cylinder is perfectly vertical. Then set the cylinder
adjusting screw, and insert the wooden screws in lock face and drive
them home with the screwdriver. Hold and adjust the escutcheon so
that the knob will rotate back and forth automatically without friction,
starting the screw holes with an awl. If the doors are made of
hardwood, an automatic drill will have to be used to bore the screw
holes. In such a case, follow the directions just given regarding
escutcheons; then start the holes with a scratch awl while still holding
the escutcheons, and enlarge the holes with the drill. Screws that are
off center, when driven home into the countersinking, will surely draw
the escutcheon so that the knobs will bind and thus destroy the most
desirable feature of the lock.
156. Mechanics will find that the better the grade of the hardware,
the more care must be taken in its application. Hardware for fine
buildings is generally furnished with high, or bracket-bearing, collars
on escutcheons and screwless knobs for adjustment without washers.
These are fitted and turned carefully, so that there is very little play
between knob shanks and collars, and for this reason they bind more
readily when improperly applied. The mechanic should try the lock in
every possible way before leaving the work. He should set the stop in
the lock face, to be sure that swivel spindles are adjusted properly,
and also insert the key in the cylinder, to see whether it operates
smoothly while operating both bolts.

SCHEDULES AND DRAWINGS FOR

THE HARDWARE CONTRACTOR

157. Hardware Schedules.—The contractor for the hardware

usually takes off his own bill of material from the general drawings and
the specifications furnished by the architect. However, the architect
will sometimes supply the hardware schedule and obtain prices for the
same direct from the dealer, thus saving the owner the general
contractor’s profit and insuring bids based on material of uniform
character and quality. It is well, therefore, to give some thought to the
preparation of the schedule, so as to insure a good classification and
to avoid omissions.
In drafting a schedule of hardware required for a building, it must
be compiled from a copy of the hardware specifications and a list of all
openings, cabinets, etc. that has been taken from the drawings. The
schedule should be subdivided primarily into building sections, such
as first floor, second floor, etc. as headings. Under these headings,
each opening or group of openings in each room or division should be
listed. Under the name of each opening or series of openings should
be grouped all the hardware required, itemizing each and every article
necessary and stating the quantity, numbers, sizes, or dimensions,
design, and finish.
Schedules thus prepared are readily priced for estimating, and
such classification eliminates the possibilities of omissions. Besides,
the order can be executed by the manufacturer in exact conformity
with the instructions, the hardware for each opening can be combined
in a separate package, and each package clearly labeled to indicate
its contents and the room or opening to which it belongs. The following
shows a typical form of schedule:

First Floor
One Pair of Entrance Doors, 2' 8" × 7' 6" × 2¼" Rt.-Hd. rabbeted.
 1 cylinder, front door, Set No. _____. Plain bronze metal.
1 push button, No. _______. Plain bronze metal.
2 extension flush bolts, No. ______, ¹/₁₂", ¹/₁₈". Plain bronze metal.
3 pair of butts, No. ______, 5" × 5". Plain bronze metal.
One Pair of Sliding-Doors, Hall to Parlor,
2" astragal face, 1½" friction strip.
1 set of locks, No. ______, 3" backset. Plain bronze metal.
3 dozen countersunk screws and washers. Plain bronze metal.
2 sets of anti-friction, noiseless, parlor-door
hangers and track, complete, No. _____.
Pantry Cupboard.
2 pair of doors, 1⅛"; 2 single doors, 1⅛".
1 tilt flour bin; 4 drawers (2 narrow).
6 pair of butts, No. _____, 2½" × 2½". Plain bronze plated.
1 pair of butts, No. _____, 3" × 2½". Plain bronze plated.
2 elbow catches, No. _____. Japanned.
4 cupboard turns, No._____. Plain bronze plated.
7 draw pulls, No._____. Plain bronze plated.

Second Floor
Ten Doors, 2' 8" × 7" × 1¾", 6 R. H., 4 L. H. transom, top-hung.
10 inside lock sets, No._____. Plain bronze metal.
15 pair of butts, No. ______, 4½" × 4½". Plain bronze plated.
10 pair of butts, No._____, 3" × 2½". Plain bronze plated.
10 transom lifts, No.______, ⁵/₁₆" × 4". Plain bronze plated.
10 wooden-base knobs, 2½" diameter.

Double-Hung Windows
10 sash fasts, No. ____. Plain bronze metal.
20 sash lifts, No. ____. Plain bronze metal.
10 sash sockets, No. ____. Plain bronze metal.
4 sash hooks, No. ____. Plain bronze metal.
10 dozen adjusting screws and washers. Plain bronze metal.

The foregoing schedule is merely intended as an example of a

classified, brief, and at the same time comprehensive hardware
schedule. A schedule of this kind shows, almost at a glance, the
nature of the hardware to be supplied, and if prepared by the
architect, should tend to procure bids based on uniform requirements.
158. Detail Drawings for the Hardware Contractor.—In some
architects’ offices, it is the practice to provide, on a single detail sheet,
cross-sections of every type of door stile in the building. The time
required to make such a drawing is slight, while the results
accomplished are most useful. Copies of this drawing are furnished to
the contractors for the cabinet trim and to the hardware contractor, so
that each of them will have identical information and that the work will
assemble properly when the hardware is put in place.
Such drawings should show the dimensions of the transverse
sections of the vertical stile, or lock stile for each door, also the
overlapping, if any, of panel moldings, together with the shape of
bevels, rabbets, astragals, and any other details affecting the size and
location of locks, hinges, butts, etc. The hand of the doors is usually,
and better, indicated on the floor plans. A further argument is that the
character and dimensions are liable to be overlooked when arranging
the paneling of doors; whereas, both should be considered. The use
of narrow stiles, special rabbets, and astragals, shelf effects, friction
strips, etc. is resorted to without sufficiently considering the
disadvantages that result from the contracted space in which the lock
and its trim must be placed. Where the items just mentioned are
proposed, the architect should ascertain what locks are available and
should provide space for those selected. Special locks are always
expensive, while a cramped space precludes the best construction.

GLASS AND GLAZING

159. Single- and Double-Thick Common Glass.—In the
manufacture of the ordinary window glass used in building
construction, the glass is first blown in cylindrical form and then cut
and flattened, after which it is reduced to salable sizes for shipment.
Owing to this method of manufacture, the percentage of first-quality
glass is small, the larger proportion of the product being second or
third quality. All glass of this kind is wavy or uneven and also has a
slight bow, or curve, to its length that is impossible to overcome when
making. The only difference in the qualities of common glass is in the
amount of the imperfections that it contains. These imperfections
consist of bull’s-eyes, blisters, scratches, and streaks that distort
objects seen through the glass.
The better grade of common glass is manufactured in what are
called continuous tanks, which are heated by natural gas. This glass is
put up in boxes containing 50 square feet. Single-thick glass runs
from thirteen to fifteen lights to 1 inch of thickness, and double thick
from eight to ten lights to the inch. Single-thick glass above 20 in. × 24
in. in size should not be used except for work of a speculative nature.
The ability to distinguish single and double thick and the various
qualities of glass can only be acquired by observation. In judging
glass, it is well to remember that small lights free from defects or
uneven surfaces may be obtained much more easily than large ones.
160. Ground, or Sand-Blast Glass.—Common or plate glass, the
surface of which is made obscure by the process of grinding, is called
ground glass. This treatment, however, does not injure the diffusing
qualities of the glass.
The sand-blast process of treating glass produces results similar
to grinding. As the name implies, sand is blown against the glass
under high pressure and thus makes numerous small cuttings on its
surface, producing the appearance of ground glass.
 Fig. 159
161. Chipped Glass.—The glass illustrated in Fig. 159 is called
chipped glass, and is manufactured in the following manner: The
glass is first given a coat of glue and is then placed in a heated
receptacle; this process tends to contract and curl the glue, so that
when it dislodges itself from the glass, the cohesion will draw or chip
off particles of glass, leaving the surface uneven and showing a
design similar to that on a frosted window light. Chipped glass can be
procured in either the single or the double process (i. e., with a single
or a double chipping); and these processes are adaptable to either
sheet or plate glass, white or colored.
 Fig. 160
162. Figured Rolled Glass.—Glass known generally as figured
rolled glass is manufactured for use in offices, public buildings, and
private residences where a glass is required to intercept the vision and
still diffuse the light. This glass has practically supplanted the
varicolored cathedral glass previously used, because of its obscurity
without reducing the quantity of light. Figured rolled glass is made in
various artistic designs, several of which are illustrated in Fig. 160,
and is ⅛ and ³/₁₆ inch in thickness. This glass is inexpensive; the
prices range from 15 to 25 cents per square foot.
163. Plate Glass.—The process of manufacturing plate glass is
radically different from that of common sheet glass. The latter is
blown, while the former is cast in large sheets and placed in annealing
ovens to cool. When taken from the ovens the glass is rough and
opaque, but it is afterwards ground and polished to make it
transparent, the polishing being the most delicate process in its
manufacture. Plate glass glistens like a mirror and reflects like one;
objects seen through it are sharp and clean-cut; it has no
imperfections or wavy effects, and does not distort and deform objects
as does sheet glass.
164. Beveled Plate Glass.—The term beveled plate is applied to
plate glass the edges of which are ground and polished to form a
bevel, or border, around the glass. Plate glass finished in this manner
is much used for glazing entrance doors and for ornamental work. The
additional cost for beveling is slight compared with the effect resulting
therefrom.
165. Floor and Skylight Glass.—Floor glass is made only from
rough rolled or hammered glass in ½-, ¾-, and 1-inch thicknesses.
The prices range from 30 cents to $1 per foot, according to the
thickness.
Skylight glass is similar to floor glass, but can be procured in ⅛-,
³/₁₆-, ¼-, and ⅜-inch thicknesses, and in either ribbed or roughened
surfaces. This glass is inexpensive, the prices ranging from 8 to 15
cents per square foot. Large quantities of skylight glass are used for
mills, skylights, and various other purposes, but it is not used as a fire-
retardant, as wire glass has entirely superseded it for this purpose.
166. Wire Glass.—Glass with wire embedded therein, as
illustrated in Fig. 161, is made either ribbed, rough-rolled, “maze”
design, or clear, polished plate. The wire netting is embedded in its
center during the process of manufacture, thus producing a very
strong glass that is a good fire-retardant. The temperature at which
the wire is embedded in the molten glass insures cohesion between
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

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

Let us accompany you on the journey of exploring knowledge and

personal growth!

textbookfull.com