Understanding Process Dynamics and Control 1st Edition Costas Kravaris Download PDF

Full download test bank at ebookmeta.
com
Understanding Process Dynamics and Control 1st

Edition Costas Kravaris
For dowload this book click LINK or Button below
https://ebookmeta.com/product/understanding-
process-dynamics-and-control-1st-edition-costas-
kravaris/
OR CLICK BUTTON
DOWLOAD EBOOK
Download More ebooks from https://ebookmeta.com

More products digital (pdf, epub, mobi) instant
download maybe you interests ...
Advanced Chemical Process Control 1st Edition Morten

Hovd
https://ebookmeta.com/product/advanced-chemical-process-
control-1st-edition-morten-hovd/
Basic Process Engineering Control Paul Serban Agachi
https://ebookmeta.com/product/basic-process-engineering-control-
paul-serban-agachi/
Process Operational Safety and Cybersecurity A Feedback

Control Approach Advances in Industrial Control Zhe Wu
https://ebookmeta.com/product/process-operational-safety-and-
cybersecurity-a-feedback-control-approach-advances-in-industrial-
control-zhe-wu/
Theory of Applied Robotics Kinematics Dynamics and

Control Reza N Jazar
https://ebookmeta.com/product/theory-of-applied-robotics-
kinematics-dynamics-and-control-reza-n-jazar/
Dynamics and Advanced Motion Control of Off Road UGVs
1st Edition Yue Ma
https://ebookmeta.com/product/dynamics-and-advanced-motion-
control-of-off-road-ugvs-1st-edition-yue-ma/
Process Control: Modeling, Design, and Simulation 2nd

Edition B. Wayne Bequette
https://ebookmeta.com/product/process-control-modeling-design-
and-simulation-2nd-edition-b-wayne-bequette/
Advanced Dynamics Modeling, Duality and Control of

Robotic Systems 1st Edition Edward Y.L. Gu
https://ebookmeta.com/product/advanced-dynamics-modeling-duality-
and-control-of-robotic-systems-1st-edition-edward-y-l-gu/
Flight Dynamics, Simulation, and Control: For Rigid and

Flexible Aircraft 2nd Edition Ranjan Vepa
https://ebookmeta.com/product/flight-dynamics-simulation-and-
control-for-rigid-and-flexible-aircraft-2nd-edition-ranjan-vepa/
Theory of Applied Robotics Kinematics Dynamics and

Control 3rd Edition Reza N Jazar
https://ebookmeta.com/product/theory-of-applied-robotics-
kinematics-dynamics-and-control-3rd-edition-reza-n-jazar/
Understanding
Process Dynamics
and Control
Costas Kravaris
Ioannis K. Kookos
Understanding Process Dynamics and Control
Understanding Process Dynamics and Control presents a fresh look at process con-
trol, with a state-space approach presented in parallel with the traditional approach
to explain the strategies used in industry today.
Modern time-domain and traditional transform-domain methods are integrated
throughout and the advantages and limitations of each approach are explained; the
fundamental theoretical concepts and methods of process control are applied to
practical problems.
To ensure understanding of the mathematical calculations involved, MATLAB
is included for numeric calculations and Maple for symbolic calculations, with the
math behind every method carefully explained so that students develop a clear
understanding of how and why the software tools work.
Written for a one-semester course with optional advanced-level material, features
include solved examples, cases including a variety of chemical process examples,
chapter summaries, key terms and concepts, as well as over 240 end-of-chapter prob-
lems, including focused computational exercises.
Costas Kravaris is Professor of Chemical Engineering at Texas A&M University,

USA. He has over 35 years of teaching experience in process dynamics and con-
trol classes at both undergraduate and graduate level. He is an active researcher in
nonlinear control, nonlinear state estimation and nonlinear model reduction, with
applications to chemical processes.
Ioannis K. Kookos is Professor of Process Systems Engineering, in the Department

Chemical Engineering at the University of Patras, Greece. He received his BSc in
Chemical Engineering from the National Technical University of Athens, Greece
and his MSc (1994) and PhD (2001) in Process Systems Engineering from Imperial
College, London (Centre for PSE). He then worked as a lecturer at the University of
Manchester, Department of Chemical Engineering.
Cambridge Series in Chemical Engineering
Series Editor
Arvind Varma, Purdue University
Editorial Board
Juan de Pablo, University of Chicago
Michael Doherty, University of California-Santa Barbara
Ignacio Grossman, Carnegie Mellon University
Jim Yang Lee, National University of Singapore
Antonios Mikos, Rice University
Books in the Series

Baldea and Daoutidis, Dynamics and Nonlinear Control of Integrated Process Systems
Chamberlin, Radioactive Aerosols
Chau, Process Control: A First Course with Matlab
Cussler, Diffusion: Mass Transfer in Fluid Systems, Third Edition
Cussler and Moggridge, Chemical Product Design, Second Edition
De Pablo and Schieber, Molecular Engineering Thermodynamics
Deen, Introduction to Chemical Engineering Fluid Mechanics
Denn,Chemical Engineering: An Introduction
Denn, Polymer Melt Processing: Foundations in Fluid Mechanics and Heat Transfer
Dorfman and Daoutidis, Numerical Methods with Chemical Engineering Applications
Duncan and Reimer, Chemical Engineering Design and Analysis: An Introduction, Second
Edition
Fan,Chemical Looping Partial Oxidation Gasification, Reforming, and Chemical Syntheses
Fan and Zhu, Principles of Gas-Solid Flows
Fox, Computational Models for Turbulent Reacting Flows
Franses, Thermodynamics with Chemical Engineering Applications
Leal, Advanced Transport Phenomena: Fluid Mechanics and Convective Transport Processes
Lim and Shin, Fed-Batch Cultures: Principles and Applications of Semi-Batch Bioreactors
Litster, Design and Processing of Particulate Products
Marchisio and Fox, Computational Models for Polydisperse Particulate and Multiphase
Systems
Mewis and Wagner, Colloidal Suspension Rheology
Morbidelli, Gavriilidis, and Varma, Catalyst Design: Optimal Distribution of Catalyst in
Pellets, Reactors, and Membranes
Nicoud,Chromatographic Processes
Noble and Terry, Principles of Chemical Separations with Environmental Applications
Orbey and Sandler, Modeling Vapor-Liquid Equilibria: Cubic Equations of State and their
Mixing Rules
Pfister, Nicoud, and Morbidelli, Continuous Biopharmaceutical Processes: Chromatography,
Bioconjugation, and Protein Stability
Petyluk, Distillation Theory and its Applications to Optimal Design of Separation Units
Ramkrishna and Song,Cybernetic Modeling for Bioreaction Engineering
Rao and Nott, An Introduction to Granular Flow
Russell, Robinson, and Wagner, Mass and Heat Transfer: Analysis of Mass Contactors and
Heat Exchangers
Schobert, Chemistry of Fossil Fuels and Biofuels
Shell,Thermodynamics and Statistical Mechanics
Sirkar, Separation of Molecules, Macromolecules and Particles: Principles, Phenomena and
Processes
Slattery, Advanced Transport Phenomena
Varma, Morbidelli, and Wu, Parametric Sensitivity in Chemical Systems
Wolf, Bielser, and Morbidelli, Perfusion Cell Culture Processes for Biopharmaceuticals
Understanding Process Dynamics
and Control
Costas Kravaris
Texas A & M University
Ioannis K. Kookos
University of Patras, Greece
University Printing House, Cambridge CB2 8BS, United Kingdom
One Liberty Plaza, 20th Floor, New York, NY 10006, USA
477 Williamstown Road, Port Melbourne, VIC 3207, Australia
314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India
79 Anson Road, #06–04/06, Singapore 079906
Cambridge University Press is part of the University of Cambridge.

It furthers the University’s mission by disseminating knowledge in the pursuit of
education, learning, and research at the highest international levels of excellence.
www.cambridge.org
Information on this title: www.cambridge.org/9781107035584
DOI: 10.1017/9781139565080
© Costas Kravaris and Ioannis K. Kookos 2021
This publication is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 2021
Printed in the United Kingdom by TJ Books Limited, Padstow Cornwall
A catalogue record for this publication is available from the British Library.
ISBN 978-1-107-03558-4 Hardback
Cambridge University Press has no responsibility for the persistence or accuracy of
URLs for external or third-party internet websites referred to in this publication
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.
Dedicated to our families,
Irene, Michael, Evangeline and Cosmas

C. Kravaris
and
Natasa, Kostas and Georgia

I.K. Kookos
Contents
Preface page xvii
1 INTRODUCTION 1
Study Objectives 1
1.1 What is Process Control? 1
1.2 Feedback Control System: Key Ideas, Concepts and Terminology 2
1.3 Process Control Notation and Control Loop Representation 8
1.4 Understanding Process Dynamics is a Prerequisite for Learning
Process Control 9
1.5 Some Historical Notes 11
Learning Summary 15
Terms and Concepts 15
Further Reading 16
Problems 17
2 DYNAMIC MODELS FOR CHEMICAL PROCESS SYSTEMS 18

Study Objectives 18
2.1 Introduction 18
2.2 Conservation Laws 20
2.3 Modeling Examples of Nonreacting Systems 23
2.4 Modeling of Reacting Systems 28
2.5 Modeling of Equilibrium Separation Systems 37
2.6 Modeling of Simple Electrical and Mechanical Systems 39
2.7 Software Tools 43
Learning Summary 45
Further Reading 46
Problems 47
3 FIRST-ORDER SYSTEMS 55
Study Objectives 55
3.1 Examples of First-Order Systems 55
3.2 Deviation Variables 58
3.3 Solution of Linear First-Order Differential Equations with Constant
Coefficients 59
ix
x Contents
3.4 The Choice of Reference Steady State Affects the Mathematical

Form of the Dynamics Problem 62
3.5 Unforced Response: Effect of Initial Condition under Zero Input 63
3.6 Forced Response: Effect of Nonzero Input under Zero Initial Condition 63
3.7 Standard Idealized Input Variations 65
3.8 Response of a First-Order System to a Step Input 68
3.9 Response of a First-Order System to a Pulse Input 73
3.10 Response of a First-Order System to a Ramp Input 75
3.11 Response of a First-Order System to a Sinusoidal Input 77
3.12 Response of a First-Order System to an Arbitrary Input – Time
Discretization of the First-Order System 82
3.13 Another Example of a First-Order System: Liquid Storage Tank 88
3.14 Nonlinear First-Order Systems and their Linearization 94
3.15 Liquid Storage Tank with Input Bypass 97
3.16 General Form of a First-Order System 99
Learning Summary 106
Further Reading 108
Problems 108
4 CONNECTIONS OF FIRST-ORDER SYSTEMS 115

Study Objectives 115
4.1 First-Order Systems Connected in Series 115
4.2 First-Order Systems Connected in Parallel 119
4.3 Interacting First-Order Systems 122
4.4 Response of First-Order Systems Connected in Series or in Parallel 123
Further Reading 136
Problems 137
5 SECOND-ORDER SYSTEMS 144

5.1 A Classical Example of a Second-Order System 145
5.2 A Second-Order System can be Described by Either a Set of Two
First-Order ODEs or a Single Second-Order ODE 147
5.3 Calculating the Response of a Second-Order System – Step Response of a
Second-Order System 148
5.4 Qualitative and Quantitative Characteristics of the Step Response of a
Second-Order System 154
Contents xi
5.5 Frequency Response and Bode Diagrams of Second-Order Systems

with ζ > 0 159
5.6 The General Form of a Linear Second-Order System 161
Further Reading 167
Problems 168
6 LINEAR HIGHER-ORDER SYSTEMS 171

6.1 Representative Examples of Higher-Order Systems – Using Vectors
and Matrices to Describe a Linear System 171
6.2 Steady State of a Linear System – Deviation Variables 175
6.3 Using the Laplace-Transform Method to Solve the Linear Vector
Differential Equation and Calculate the Response – Transfer Function
of a Linear System 177
6.4 The Matrix Exponential Function 179
6.5 Solution of the Linear Vector Differential Equation using the Matrix
Exponential Function 182
6.6 Dynamic Response of a Linear System 187
6.7 Response to an Arbitrary Input – Time Discretization of a Linear System 191
6.8 Calculating the Response of a Second-Order System via the Matrix
Exponential Function 195
6.9 Multi-Input–Multi-Output Linear Systems 197
Further Reading 206
Problems 207
7 EIGENVALUE ANALYSIS – ASYMPTOTIC STABILITY 215

7.1 Introduction 215
7.2 The Role of System Eigenvalues on the Characteristics of the Response of
a Linear System 216
7.3 Asymptotic Stability of Linear Systems 220
7.4 Properties of the Forced Response of Asymptotically Stable Linear Systems 224
7.5 The Role of Eigenvalues in Time Discretization of Linear Systems –
Stability Test on a Discretized Linear System 225
7.6 Nonlinear Systems and their Linearization 228
xii Contents

Further Reading 245
Problems 245
8 TRANSFER-FUNCTION ANALYSIS OF THE INPUT–OUTPUT BEHAVIOR 251

8.2 A Transfer Function is a Higher-Order Differential Equation in
Disguise 252
8.3 Proper and Improper Transfer Functions – Relative Order 257
8.4 Poles, Zeros and Static Gain of a Transfer Function 259
8.5 Calculating the Output Response to Common Inputs from the
Transfer Function – the Role of Poles in the Response 261
8.6 Effect of Zeros on the Step Response 268
8.7 Bounded-Input–Bounded-Output (BIBO) Stability 273
8.8 Asymptotic Response of BIBO-Stable Linear Systems 275
Further Reading 288
Problems 288
9 FREQUENCY RESPONSE 297

9.2 Frequency Response and Bode Diagrams 298
9.3 Straight-Line Approximation Method for Sketching Bode Diagrams 303
9.4 Low-Frequency and High-Frequency Response 311
9.5 Nyquist Plots 312
Further Reading 322
Problems 322
10 THE FEEDBACK CONTROL SYSTEM 327

10.1 Heating Tank Process Example 327
10.2 Common Sensors and Final Control Elements 329
10.3 Block-Diagram Representation of the Heating Tank Process Example 332
Contents xiii
10.4 Further Examples of Process Control Loops 335

10.5 Commonly Used Control Laws 338
Further Reading 346
Problems 346
11 BLOCK-DIAGRAM REDUCTION AND TRANSIENT-RESPONSE CALCULATION

IN A FEEDBACK CONTROL SYSTEM 350
11.1 Calculation of the Overall Closed-Loop Transfer Functions in a
Standard Feedback Control Loop 350
11.2 Calculation of Overall Transfer Functions in a Multi-Loop Feedback
Control System 356
11.3 Stirred Tank Heater under Negligible Sensor Dynamics:
Closed-Loop Response Calculation under P or PI Control 359
Further Reading 373
Problems 374
12 STEADY-STATE AND STABILITY ANALYSIS OF THE CLOSED-LOOP SYSTEM 377

12.1 Steady-State Analysis of a Feedback Control System 377
12.2 Closed-Loop Stability, Characteristic Polynomial and
Characteristic Equation 385
12.3 The Routh Criterion 389
12.4 Calculating Stability Limits via the Substitution s = iω 394
12.5 Some Remarks about the Role of Proportional,
Integral and Derivative Actions 395
Further Reading 405
Problems 405
13 STATE-SPACE DESCRIPTION AND ANALYSIS OF THE CLOSED-LOOP SYSTEM 409

13.1 State-Space Description and Analysis of the Heating Tank 409
13.2 State-Space Analysis of Closed-Loop Systems 415
xiv Contents
13.3 Time Discretization of the Closed-Loop System 422

13.4 State-Space Description of Nonlinear Closed-Loop Systems 426
Further Reading 435
Problems 435
14 SYSTEMS WITH DEAD TIME 437

14.2 Approximation of Dead Time by Rational Transfer Functions 446
14.3 Parameter Estimation for FOPDT Systems 456
14.4 Feedback Control of Systems with Dead Time – Closed-Loop
Stability Analysis 460
14.5 Calculation of Closed-Loop Response for Systems involving Dead Time 467
Further Reading 476
Problems 476
15 PARAMETRIC ANALYSIS OF CLOSED-LOOP DYNAMICS – ROOT-LOCUS DIAGRAMS 484

15.1 What is a Root-Locus Diagram? Some Examples 484
15.2 Basic Properties of the Root Locus – Basic Rules for Sketching
Root-Locus Diagrams 502
15.3 Further Properties of the Root Locus – Additional Rules for Sketching
Root-Locus Diagrams 508
15.4 Calculation of the Points of Intersection of the Root Locus with the
Imaginary Axis 524
15.5 Root Locus with Respect to Other Controller Parameters 527
Further Reading 537
Problems 537
16 OPTIMAL SELECTION OF CONTROLLER PARAMETERS 541

16.1 Control Performance Criteria 541
16.2 Analytic Calculation of Quadratic Criteria for a Stable System and a
Step Input 549
Contents xv
16.3 Calculation of Optimal Controller Parameters for Quadratic Criteria 557

Further Reading 571
Problems 572
17 BODE AND NYQUIST STABILITY CRITERIA – GAIN AND PHASE MARGINS 575
17.2 The Bode Stability Criterion 576
17.3 The Nyquist Stability Criterion 594
17.4 Example Applications of the Nyquist Criterion 597
Further Reading 608
Problems 608
18 MULTI-INPUT–MULTI-OUTPUT SYSTEMS 613

18.2 Dynamic Response of MIMO Linear Systems 620
18.3 Feedback Control of MIMO Systems: State-Space versus
Transfer-Function Description of the Closed-Loop System 623
18.4 Interaction in MIMO Systems 627
18.5 Decoupling in MIMO Systems 632
Further Reading 639
Problems 639
19 SYNTHESIS OF MODEL-BASED FEEDBACK CONTROLLERS 641

19.2 Nearly Optimal Model-Based Controller Synthesis 648
19.3 Controller Synthesis for Low-Order Models 650
19.4 The Smith Predictor for Processes with Large Dead Time 657
19.5 Effect of Modeling Error 660
19.6 State-Space Form of the Model-Based Controller 668
xvi Contents
19.7 Model-Based Controller Synthesis for MIMO Systems 674

Further Reading 679
Problems 679
20 CASCADE, RATIO AND FEEDFORWARD CONTROL 683

20.2 Cascade Control 684
20.3 Ratio Control 694
20.4 Feedforward Control 695
20.5. Model-Based Feedforward Control 700
Further Reading 715
Problems 715
APPENDIX A LAPLACE TRANSFORM 719

A.1 Definition of the Laplace Transform 719
A.2 Laplace Transforms of Elementary Functions 720
A.3 Properties of Laplace Transforms 721
A.4 Inverse Laplace Transform 725
A.5 Calculation of the Inverse Laplace Transform of Rational
Functions via Partial Fraction Expansion 725
A.6 Solution of Linear Ordinary Differential Equations using the
Laplace Transform 732
A.7 Software Tools 735
Problems 739
APPENDIX B BASIC MATRIX THEORY 743

B.1 Basic Notations and Definitions 743
B.2 Determinant of a Square Matrix 747
B.3 Matrix Inversion 749
B.4 Eigenvalues 750
B.5 The Cayley–Hamilton Theorem and the Resolvent Identity 752
B.6 Differentiation and Integration of Matrices 755
B.7 Software Tools 756
Index 760
Preface
Scope of the Book

When we took undergraduate process dynamics and control in the 1970s and the 1980s,
the entire course was built around the Laplace transform and the transfer function. This
conceptual and methodological approach has been in place in undergraduate chemical engi-
neering education since the 1960s and even the 1950s, and it reflected the development and
widespread use of electronic PID control systems, for which it provided a very adequate
background for the chemical engineering graduates. Today, the vast majority of undergrad-
uate chemical process dynamics and control courses still follow exactly the same conceptual
approach, revolving around the Laplace transform and the transfer function. But control
technology has changed a lot during the past 60 years. Even though PID controllers are still
used, model-predictive control has evolved into an industrial standard for advanced applica-
tions. But model-predictive control is formulated in state space and in discrete time, whereas
the standard control course is in the transform domain and in continuous time. There is a big
conceptual gap between what is taught in the classroom and the industrial state of the art.
This gap is well recognized within the chemical process control community, as is the need to
bridge this gap. It is aim of this book to propose a realistic solution on how to bridge this
gap, so that chemical engineering graduates are better prepared in using modern control
technology. This book has evolved after many years of teaching experimentation at Texas
A&M University and the University of Patras.
The main feature of this book is the introduction of state-space methods at the undergrad-
uate level, not at the end of the book, but from day one. There are two main reasons that
this is feasible. The first is that state-space concepts and methods are easy to grasp and com-
prehend, since they are in the time domain. The second is the availability of powerful com-
putational tools that emerge from the state-space methods and can be implemented through
user-friendly software packages. Once the student is given the key ideas and concepts in the
time domain, he/she can painlessly apply them computationally.
Of course, one should not downplay the significance of manual calculations in developing
an understanding of dynamic behavior in open loop and in closed loop. To this end, Laplace-
transform methods offer a distinct advantage over time-domain methods. Even though
industrial practitioners keep telling us that “there is no Laplace domain in their plant,” there
is no question about its educational value. The concept of the transfer function is also an
xvii
xviii Preface
invaluable educational tool for the student to understand connections of dynamic systems,
including the feedback loop, and also to calculate and appreciate frequency response charac-
teristics. For this reason, Laplace-domain methods are used in this book, and they are used
in parallel with state-space methods. Whenever a quick manual calculation is feasible, the
student should be able to go to the “Laplace planet” and come back, whenever calculations
are very involved or simulation is needed, the student should be able to handle it computa-
tionally using software.
This book offers a strong state-space component, both conceptually and computation-
ally, and this is blended with the traditional analytical framework, in order to maximize
the students’ understanding. But there is also an additional advantage. Because of its state-
space component, this book brings the process dynamics and control course closer to other
chemical engineering courses, such as the chemical reactor course. A chemical reactor course
introduces local asymptotic stability in a state-space setting and tests it through eigenval-
ues, whereas a traditional control course defines stability in an input–output sense and tests
it through the poles of the transfer function. This gap is nonexistent in the present book:
asymptotic stability is defined and explained in a state-space context, input–output stability
is defined and explained in a transfer function or convolution integral context, and the rela-
tionship of the two notions of stability is discussed. Moreover, there are a number of chem-
ical reactor examples throughout the book that link the two courses in a synergistic manner.
A final comment should be made about the word “understanding” in the title of this book.
It is our firm belief that engineers must have a thorough understanding of how their tools
work, when do they work and why they work. If they treat a software package as a magic
black box, without understanding what’s inside the box, they have not learned anything. For
this reason, special care is taken in this book to explain the math that is behind every method
presented, so that the student develops a clear understanding of how, when and why.
Organization of the Book

A general introduction is given in Chapter 1. A review of unsteady state material and energy
balances is given in Chapter 2. Reviews of the Laplace transform and of basic matrix algebra
are separate from the chapters, and are given in Appendices A and B.
Chapters 3–9 and the first half of Chapter 14 cover process dynamics. The approach taken
is to start from the simplest dynamic systems (first-order systems) in Chapter 3, and then
progressively generalize. Both time domain (including discrete time) and transfer function
(including frequency response) start from Chapter 3 and are pursued in parallel in the sub-
sequent chapters. Chapters 4 and 5 are generalizations, studying connections of first-order
systems and inherently second-order systems. Chapters 6–9 cover the dynamic analysis of
higher-order systems in both state space (Chapters 6 and 7) and transform domain (Chapters
8 and 9), including asymptotic stability and input–output stability. Dead time is postponed
to Chapter 14. All the dynamics chapters are to be covered; the only part that is optional is
Preface xix
the second part of Chapter 9 on Nyquist diagrams, which is only needed in the second part
of Chapter 17.
The rest of the chapters are on process control. Chapters 10–14 cover the basic feedback
control concepts and analysis methods. Chapter 10 gives a general introduction to feedback
control, and also defines the PID controller in both state-space and transfer function form.
Chapters 11 and 12 do transfer function analysis of the feedback control loop, whereas in
Chapter 13 the same analysis is done in state space. Chapter 14 discusses systems with dead-
time, both open loop dynamics and feedback control. Deadtime is treated separately because
of its distinct mathematical characteristics. Chapters 10–14 provide an absolute minimum
for the feedback control part of the course. From that point, the instructor can choose what
design methods he/she wants to put emphasis on, root locus (Chapter 15), optimization
(Chapter 16), gain and phase margins (Chapter 17) or model-based (Chapter 19). Also, the
instructor has the choice to discuss issues in multivariable control (Chapter 18) or stay SISO
throughout the course. The last chapter (Chapter 20) discusses cascade, ratio and feedfor-
ward control. These control structures are discussed first at a conceptual level, and then
model-based design for cascade and feedforward control is derived. The conceptual part is,
in a sense, a continuation of Chapter 10 and it is essential to be taught; the model-based part
is a continuation of Chapter 19.
The last section of each chapter is about software tools. The use of software for the appli-
cation of the theory of the chapter is explained through simple examples. Two alternative
software packages are used: MATLAB and its control systems toolbox is chosen because of
its strength in numerical calculations, and Maple and its libraries (LinearAlgebra, inttrans,
etc.) because of its strength in symbolic calculations.
The following table gives a sample syllabus for the process dynamics and control course at
Texas A&M University, as it has been taught in the past three semesters. It reflects the per-
sonal choices of the instructor on (i) the design methods for the control part of the course
(optimization and model-based are emphasized) and (ii) the pace of covering the material
(slower at the beginning, faster at the end). Of course, there are many other options, depend-
ing on instructor priorities and students’ background.
Topic From the book Hours

Introduction Chapter 1 1
Review of unsteady-state material and energy Chapter 2 1
balances
Review of the Laplace transform Appendix A 2
First-order systems Chapter 3 5
Connections of first-order systems Chapter 4 2
Second-order systems Chapter 5 2
Higher-order systems Chapter 6 and Appendix B (first half) 4½
Eigenvalue analysis, asymptotic stability Chapter 7 and Appendix B (second half) 2½
Transfer-function analysis Chapter 8 2
Bode diagrams Chapter 9 – Bode part 1
xx Preface
Topic From the book Hours

The feedback control system Chapter 10 1
Block-diagram simplification, closed-loop Chapter 11 2
responses
Steady-state analysis, stability analysis Chapter 12 2½
State-space analysis of the closed-loop system Chapter 13 1½
Optimization of feedback controllers Chapter 16 2
Systems with dead time Chapter 14 2
Bode stability criterion, gain and phase Chapter 17 – Bode part 1
margins
Model-based control Chapter 19, excluding MIMO 2
Cascade, ratio and feedforward control Chapter 20 2
Total lecture hours 39
Costas Kravaris and Ioannis K. Kookos, October 2020

1 Introduction
STUDY OBJECTIVES
After studying this chapter, you should be able to do the following.
• Identify the underlying reasons for the need of control systems.

• Identify the main parts of a feedback control system.
• Identify the main terminology and notation used in process control.
• Discuss recent history of process control.
1.1 What is Process Control?

Automatic control is a discipline which studies the design of man-made systems with the
aim to “shape” purposefully their response. Scientists and engineers who work in this field
of study, depending on their background or their area of interest, may give a more specific
or more abstract definition. Automatic control is an interdisciplinary science and plays a key
role in most engineering disciplines including electrical, mechanical and chemical engineer-
ing. There is a common theoretical basis that can be applied to all these systems, despite the
major differences in their physical characteristics.
Process control is the branch of automatic control concerned with production plants in
the chemical, petrochemical, food and related industries. Process control plays a critical role
in ensuring proper operation of the plant, in terms of safety, product quality and profita-
bility. Even though chemical processes are of different physical nature when compared to
robots, unmanned vehicles and aircrafts, missiles and spacecrafts, the underlying principles
of automatic control are the same.
Automatic control is a part of our everyday life. Cars, refrigerators, washing machines, pub-
lic buildings and homes have numerous automatic control systems installed. What is equally
impressive is that these control systems operate and function so efficiently that we hardly ever
take notice of their existence. They deliver, they are reliable, they make our everyday life better
and safer, and they are really cheap. They are the result of the hard work of numerous ingen-
ious scientists and engineers who devoted their life to make our world a better place to be.
1
2 Introduction
The amount of knowledge that has been generated in the past 100 years in the field of
automatic control is vast. Using this knowledge to design and operate control systems in
practice is vital for maintaining the same pace of development in the years to come. This
book aims at explaining the fundamental principles of process control in a way that makes
it easier for future chemical engineers to comprehend past developments and to develop new
tools that advance engineering practice. We also hope that learning process control methods
and concepts will help future chemical engineers to interact and collaborate with control
engineers of other disciplines.
1.2 Feedback Control System: Key Ideas, Concepts and Terminology

The idea of feedback control will be introduced in the present section, along with some per-
tinent key concepts and terminology.
Consider the primitive control system shown in Figure 1.1. A liquid stream is fed to a
buffer tank ( process) and an operator (controller) tries to keep the liquid level in the tank
(measured and controlled variable) at the desired value (set point) by using a logical proce-
dure (control algorithm) based on his/her training and experience. The means to accomplish
this task is the opening or closing of a valve ( final control element) that adjusts the flowrate
(manipulated variable) of the exit stream.
A number of questions are immediately raised.
• Why does the liquid level vary during everyday operation?

• What is the “desired level” of the liquid in the tank and on what grounds is it
determined?
High
Set Point
Low
Figure 1.1 A “primitive” level control system.

• What is the best sequence of actions that the operator needs to take when the liquid
level deviates from the set point?
• Do we need a human operator to control the level? Why not build an “automated
system”?
• Would an “automated system” be more efficient and reliable than the human in con-
trolling the level of the liquid?
• Are there any other measurements that could be used, in combination with the level of the
liquid in the tank, in order to more effectively maintain the liquid level at the desired value?
Some of these types of questions have an easy answer, but some will need further thinking
and elaboration throughout your process control course and even throughout your profes-
sional career.
Most undergraduate courses in chemical engineering consider processes that operate at
steady state. This is a logical and well-documented simplification that allows chemical engi-
neers to design fairly complex processes in a reasonable amount of time. However, in actu-
ality, processes operate in a dynamic environment. Imagine that you design a heat exchanger
that uses sea water as cooling water to cool down a process stream from 100 °C to 50 °C.
At the design stage you have to make an assumption about the temperature of the cooling
water (a unique value) and suppose that you have selected a temperature of 20 °C. Now think
about the chances of the cooling water temperature being exactly 20 °C. Will the system fail
if the actual temperature of the water is 15 °C or 10 °C? The answer is yes, the system will
fail to keep the temperature at the desired value of 50 °C, unless a valve is installed, which
can appropriately adjust the flowrate of the cooling water. In addition, a temperature sen-
sor needs to be installed, to measure the temperature of the process stream exiting the heat
exchanger. Then, using the measured and recorded temperature, an operator can check if
the temperature is at the proper value, and appropriately adjust the cooling water flowrate
to correct any discrepancies, as shown in Figure 1.2. The sea-water temperature can vary
Sensor
Controller
Process
Final
control
element
Figure 1.2 A “primitive” temperature-control system.

4 Introduction
throughout the day, so the operator will need to perform frequent adjustment of the valve
opening to keep the temperature of the process stream close to the desired temperature. In
addition, the operator can implement changes in the desired temperature, if there are reasons
related to the operation of downstream processes.
The basic elements of the temperature-control system shown in Figure 1.2 are also shown
in the block diagram of Figure 1.3. The blocks are used as a means of representing the com-
ponents of the system and the arrows denote a signal or information flow. The measurement
(the line exiting the sensor and entering the controller) is not, in the case of the control
system of Figure 1.2, an actual signal but an information flow and denotes the reading of
the temperature indication by the operator. The operator/controller is a necessary element
of the loop that processes (using a control algorithm) the information and decides on the
appropriate action to be taken (opening or closing of the valve). The opening or closing of
the valve determines the flowrate of the cooling medium (sea water) and thus the rate of heat
transfer in the heat exchanger. Finally, the temperature of the product stream is measured
by the sensor (operator’s eyes) and the loop is closed. In most cases, the controller is a com-
puter-based system that receives a signal from the sensor, executes the control algorithm and
sends a signal that sets the valve position, as indicated in Figure 1.4. Computers can perform
very complex calculations in a very short time, can handle more than one control system
simultaneously and work continuously and, in most cases, without human intervention.
Controller Valve Process Sensor
Figure 1.3 Elements of a “primitive” temperature-control system.
Controller
Set
Sensor
point
Process
Final
control
element
Figure 1.4 A computer-based temperature-control system.

Let’s now summarize the types of process variables encountered in the primitive control
systems that we have seen so far.
Disturbance variables. Any process is affected by several external influences, and many of
them vary in an uncontrollable and unpredictable manner. These are the disturbances that
cause the operation of the process to deviate from the desired steady state. In the case of the
heat exchanger, potential disturbances are the sea-water temperature, the temperature and
the flowrate of the incoming process stream, or equipment aging known as fouling (which
increases the resistance to heat transfer). Some of the disturbances could be measured in real
time but others are difficult, expensive or even impossible to measure.
Manipulated variables. The manipulated variables are those process variables that are
adjusted by the controller in order to achieve the control objectives. A manipulated variable
is also called control input, to signify that it represents the control action that “feeds” the
process. The most frequent manipulated variable in the chemical process industries is the
flowrate through the installation of a control valve or a pump.
Measured variables. Measured variables are all the variables for which we have installed a
sensor or measuring device that continuously measures and transmits the current value of
the variable. Of course, sensors cost money and need frequent maintenance, therefore their
installation should be well justified. The most common measured variables in the chemical
industries are temperature, pressure, flow and level. Others, such as composition, are more
costly and less frequently used.
A measured variable that the controller is maintaining at a particular desired value is
called a controlled variable. The desired value is called the set point of the controlled variable.
The set point is usually kept constant for a long time, but sometimes a need may arise to
change the set point, and this should be handled by the controller.
When the value of the controlled variable agrees with its set point, it is “in control,” oth-
erwise there is an error. The error is defined as the difference between the set point and the
value of the controlled variable, and the job of the controller is to make it equal to zero.
Figure 1.5 depicts a generic feedback control system. It shows all the types of variables
that come into play, as well as the basic elements of the control system and how they are
connected with each other.
The final control element (usually a control valve) together with the process and the
sensor comprise the physical system or open-loop system. We see from Figure 1.5 that, when
the sensor is connected to the controller, and the controller is acting on the final control
element, the overall system has a circular structure, like a ring or a loop, and it is called
the closed-loop system. It is also called a feedback control system. The idea of feedback
control involves continuous monitoring of the controlled variable and “feeding back”
the information, to make changes and adjustments in the process, through changes in the
manipulated variable. The controller’s action is usually based on the error, i.e. the discrep-
ancy between the set point and the measurement of the controlled output. Depending on
the error (its current value, its history and its trend), the controller takes corrective action.
In simple terms, one can describe the operation of a feedback control system as: monitor,
detect and correct.
6 Introduction
Disturbance
variable
Set Measured/
Manipulated
point controlled
Error variable
variable
ysp e=ysp – ym u Final y
Controller control Process
+ element
–
ym
Sensor
Measurement
Physical system or open-loop system
Closed-loop system
Figure 1.5 Basic elements of a feedback control system and their interconnections.
Sensors play a critical role is the proper operation of a feedback control system. Sensors
use an electrical or mechanical phenomenon in order to determine the temperature, pres-
sure, level or flowrate. Temperature sensors are based on the expansion of a liquid or gas
(thermometers), on the Seebeck or thermoelectric effect, the creation of voltage between two
junctions at different temperatures (thermocouples), the variation of electrical resistance of
several materials with temperature (resistance temperature detectors and thermistors) and
the thermal radiation emitted (pyrometers). Most pressure sensors are based on measuring
the deflection or strain caused by the pressure when applied to an area (strain-gauge, elec-
tromagnetic, piezoelectric, etc.). Pressure sensors are used in conjunction with an orifice or
a Venturi tube to measure flow, as differential pressure across the orifice or between two seg-
ments of a Venturi tube (with different aperture) is strongly related to flow. Pressure sensors
are also used to calculate the level of a liquid in a tank as the pressure difference between the
top and the bottom of a tank is directly proportional to the height of the liquid. The trans-
mitter is used to convert the primary measurement by the sensor to a pneumatic or electrical
signal. The combination of the sensor and the transmitter is called a transducer.
In a chemical plant, there may be hundreds or thousands of feedback control loops like
the one depicted in Figure 1.5. The need to transmit all information and functionality to a
central “control room” (see Figure 1.6) to achieve continuous monitoring and reduce dras-
tically the manpower required was quickly identified and implemented in the 1960s. This
centralization was really effective in improving the operation of the plant. At that time, the
controllers (one controller for each control loop) were behind the control room panels, and
all control signals were transmitted back to plant. Gradually the structure was modified as
all functionalities were assigned to a network of input/output racks with their own control
processors which could be distributed locally in the plant (and could communicate with the
Figure 1.6 A control room of the 1960s and a more recent DCS control room (from
Wikimedia Commons, the free media repository).
control room). The distributed control system or DCS was thus born, in which the control-
lers were placed close to the processing units but transmitted all information to a central
location through a central network to minimize cabling runs. Monitoring, interconnection,
reconfiguration and expansion of plant controls were finally easy. Local control algorithms
could be executed by the central units in the case of system failure and thus reliability was
greatly enhanced. Recent advantages such as wireless technology and Internet of Things as
well as mobile interfaces might have a real impact in the near future.
8 Introduction
1.3 Process Control Notation and Control Loop Representation

The standard notation used in process control is also indicated in Figure 1.5. The actual
value of the measured and controlled variable is denoted by y. The measurement is denoted
by ym, and this may not match y in a transient event, as the sensor signal may be lagging
behind in the changes of the physical variables that it measures. The desired or set-point
value of the controlled variable is denoted by ysp. The error signal e = ysp – y is also indicated
in the diagram. (The small circle with the two inward arrows with appropriate signs and one
outward arrow indicates the subtraction operation.) The error signal e drives the controller,
which determines the appropriate adjustments, in order to correct the error and eventually
bring it back to zero. The signal u from the controller sets the value of the manipulated var-
iable of the process, which is actually implemented by the final control element. Finally, the
sensor detects the change in the response of the system and the loop is closed.
Process engineers use standard symbols to denote process units such as vessels, heat
exchangers and towers when constructing the Process Flow Diagram (PFD) of a produc-
tion facility. The same holds true for control and instrumentation engineers. The stand-
ards for documenting the details of control and instrumentation have been defined by the
Instrumentation, Systems, and Automation Society (ISA) and are known as Standard ISA-
S5. There are several publications by the ISA that document in great detail the construction
of Process and Instrumentation Diagrams (P&ID) that are routinely used by process engi-
neers during the construction, commission and operation phases. The reader is referred to
these publications as they are outside the scope of this book.
The notation that will be used in this book is presented in Figure 1.7 through an example.
The standard representation of a control system is shown in this figure. A process stream
Temperature Temperature
Controller Transducer
TC TT
Electric signal
Process stream
Cooling water stream

Standard symbol for a
control valve Standard symbol for a
Flow heat exchanger
FC
Controller
Flow
Transducer FT
Standard symbol for a

control valve
Figure 1.7 Common control loop representation.

1.4 Understanding Process Dynamics is a Prerequisite for Learning Process Control 9
that is flow-controlled, is cooled using a heat exchanger and cooling water. The temperature
of the outgoing process steam is measured by a temperature sensor and the measurement is
indicated by the temperature transducer (TT) symbol in Figure 1.7. Bubble or circle symbols
are used to indicate instrumentation (measurement) or control function. Inside the circle
symbol a two-letter coding system is used to denote the specific functionality of the block.
The first letter in the two-letter naming system refers to the variable controlled or measured
and the letters commonly used are the following.
T: temperature
F: flow, flowrate
L: level
P: pressure
C (or A): composition
The second letter indicates whether this is a measuring device or transducer (T) or a
control device (C). TC is therefore used to indicate a temperature controller while FC indi-
cates a flow controller in Figure 1.7. The two circle symbols denoted as TT (temperature
transducer) and TC (temperature controller) are connected through a dashed line (– – – –)
which indicates an electrical signal (4–20 mA, 1–5 V or 0–10 V). Other common conven-
tions are the following: a pneumatic signal is denoted by —//—//— (normally in the range
of 3–15 psig) and a data-transfer signal is denoted by —o—o— (usually binary signal).
We will not try to indicate explicitly whether a signal is electrical or pneumatic, as it adds a
complexity that is unnecessary within this book, and we will be using a dashed line to indi-
cate exchange of information between a sensor, a controller and a final control element, as
shown in Figure 1.7.
1.4 Understanding Process Dynamics is a Prerequisite for Learning

Process Control
The action of a controller is not static: it is dynamic in nature. As external disturbances vary
with time, the controller must take action, in a continuously changing environment. And the
controller is not isolated: it keeps interacting with the sensor and the final control element,
which in turn interact with the process, and all these interactions are transient in nature. To
be able to understand what is happening inside the feedback control loop, we must first have
a thorough understanding of transient behavior.
The process, the final control element, the sensor and the controller are all dynamic sys-
tems, whose behavior changes with time due to a changing environment (such as varying
feed composition or temperature), changing process specifications (such as changing prod-
uct purity) or equipment aging (such as fouling). The mathematical tools normally used to
describe process dynamics are ordinary and partial differential equations accompanied, in
some cases, by algebraic equations.
10 Introduction
In the first part of this book we will study the dynamics of an isolated system, to try to
understand its transient behavior. We will see different kinds of transient behavior, and we
will explain the behavior and characterize it in a systematic way. We will see how to calcu-
late these transient responses, analytically and numerically. One of the key concepts that we
will discuss is the concept of stability, and we will derive tests to determine if a system has
stable behavior. We will also introduce the necessary software tools to calculate routinely the
dynamic response of common process systems.
Equipped with these concepts and tools, we will study interconnected dynamic systems,
in a feedback control loop. We will see how the dynamic behavior of all the elements of the
loop can be combined, and we will derive the dynamic behavior of the overall system, and
calculate its transient response.
Typical process systems’ dynamic responses are presented in Figure 1.8. The response
can be fast or relatively slow as shown in Figures 1.8a and b, respectively. A characteristic
commonly encountered in process systems is that of delayed response as shown in Figure
1.8c. These three general responses are the ones usually obtained by chemical processes like
distillation and absorption columns, evaporators and heat-transfer equipment. There are
Figure 1.8 Representative (a)

cases of process system y(t)
transients: (a) fast transient,
(b) slow transient, (c) delayed
0
transient, (d) oscillatory 0 t
transient and (e) unstable
(b)
transient. y(t)
0
0 t
(c)
y(t)
0
0 t
(d)
y(t)
0
0 t
(e)
y(t)
0
0 t
processes, like chemical reactors, that can exhibit oscillatory (Figure 1.8d) or even unstable
(Figure 1.8e) response.
1.5 Some Historical Notes

A comprehensive presentation of the history of feedback control has been written by O.
Mayr (The Origins of Feedback Control, The MIT Press, 1970), who traces the control of
mechanisms to antiquity with references to the work of Ktesibios, Philon and Heron as the
main representatives of Hellenistic technology. The interested reader is referred to Mayr’s
book for a detailed discussion of this early work. The modern form of control systems tech-
nology is believed to have started in the middle of the eighteenth century. A famous problem
at that time was the search for a means to control the rotation speed of a shaft, used for
instance in the grinding stone in a wind-driven flour mill. A promising method was based
on the use of a conical pendulum, or flyball governor (also called a centrifugal governor), to
measure the speed of the mill. The adaptation of the flyball governor to the steam engine in
the laboratories of James Watt around 1788 made the flyball governor famous. The action
of the flyball governor is shown in Figure 1.9 and was simple: the balls (which spin around
a central shaft) of the governor generate an angle with the horizontal that is analogous to
the speed of the rotating shaft (and the speed of the engine, which is connected to the shaft).
Increasing the speed increases the angle and decreasing the speed decreases the angle. The
angle formed can therefore be used to sense the speed of the engine and then proper control
action can be achieved (by mechanical means) by manipulating the flowrate of the steam (in
the case of the steam engine). When the engine speed drops (due to an increase in the engine’s
load), the ball angle decreases and more steam will be admitted by mechanical means, restor-
ing most of the lost speed. The reverse action will be achieved in the case of decreased load.
A basic form of feedback control is thus realized. Apart from the famous flyball governor,
Balls are raised when

rotational speed of
shaft increases
The flowrate of steam

decreases as the plug is
lowered blocking the
Balls opening
Central shaft
(connected to the Steam to engine
engine)
Figure 1.9 Basic operation of the flyball governor (engine speed control system).
12 Introduction
slow progress took place in the field in the nineteenth century. Things, however, made a dra-
matic change in the beginning of the twentieth century.
Things were very different in the chemical industry just 100 years ago. We present the fol-
lowing extract from one of the most significant journals in chemical engineering (the Journal
of Industrial & Engineering Chemistry, published by the American Chemical Society or ACS)
that was published in 1918 (I&ECR, p. 133, February 1918, by R. P. Brown from The Brown
Instrument Company, Philadelphia):
Probably no employee causes the average works manager so many sleepless nights as does the
furnace man, on whose shoulders rests the responsibility for the accurate heat treatment of the
steel and the uniformity of the product. This is not only true of a steel plant, but is also equally
true in the chemical industry, where the temperature of numerous processes must be accurately
controlled; … The old furnace man, through years of practice, will endeavor to gauge the tem-
perature of the furnace with his eye. Providing he has not been up all the previous night and his
eye is clear, he will probably judge the temperature fairly accurately. …
But we can pardon the works manager or director for asking, “Suppose John dies, gets sick or
quits his job, how am I to handle the output of these furnaces?” He would like to have an under-
study for the old furnace man, but the latter does not like the idea. So he wonders why someone
does not develop a device to automatically control the temperature of the furnaces, so that he
can cease worrying about them.
This is one reason why a great amount of study has been given, not only to perfection of
pyrometers, but also to the automatic control of temperature. It has, however, been only recently
that real results have been accomplished in automatic temperature control.
Some primitive control systems are described only a few years later. One of these is shown
in Figure 1.10. The so-called vapor-tension system consisted of a metallic bulb partly filled
with a liquid, located in the container under
Valve control, and a metallic capillary leading from
the bulb to an expansion or capsular chamber
Heating in the regulator case. Any change in the tem-
fluid Heating perature of the space surrounding the metallic
in Plug fluid
out
bulb will cause a change in the vapor pressure
Stem of the enclosed liquid and therefore in the pres-
Diaphragm sure below the diaphragm. As the pressure in
Capillary tube the capillary tube and diaphragm changes, this
causes the stem and the plug to move up (when
the temperature increases) or down (when the
temperature decreases). If the plug is used in a
valve to restrict or release the opening through
Figure 1.10 A primitive control system (adapted with

modifications from I&ECR, 14(11), 1016, November
1922).
which a thermal fluid is supplied to the system then a rudimentary control action can be
achieved. When the temperature increases the vapor pressure of the enclosed liquid also
increases, raising the stem and the plug and blocking the flow of the heating fluid. This
causes less heating fluid to be supplied to the system and the temperature drops. This is a
continuous control system that was chosen here purposefully despite the fact that most early
control systems were simple on–off devices.
A number of interesting descriptions of early, basic control systems appeared in the lit-
erature in the late 1920s (see, for instance, I. Ginsberg, Automatic control in the chemical
industries, I&ECR, 21(5), 410, 1929). Real progress was achieved in the 1930s and mainly in
the early 1940s with the publication of the seminal papers by Ziegler and Nichols (Ziegler,
J. G. and Nichols, N. B., Trans. Am. Soc. Mech. Eng., 64, 759–68, 1942, and 65, 433–44,
1943). These papers presented a systematic method for selecting the parameters of process
controllers with minimal process upsetting and were based solely on the response of the
open-loop system to a step change in the process input. The controller parameters were then
determined as a function of the characteristics of the step response. The method proved
to be particularly successful and gained widespread acceptance from the community. The
Ziegler–Nichols method is still selected by many researchers, 80 years later, as a basis for
establishing the advantages of their controller tuning techniques.
The numerous military applications of automatic control, developed and perfected dur-
ing the Second World War, created a strong interest in the chemical process industries. Just
before the beginning of the Second World War, a significant change started to materialize in
the chemical plants as most of the large enterprises, such as E. I. du Pont de Nemours and
Company, realized the benefits of continuous operation as compared with batch operation.
However, it all had to wait until the introduction of microcomputers in the late 1950s and
it was during the 1960s that the process control became an integral part of the operation of
the chemical plants.
In the 1950s, process control theory witnessed significant progress as particular atten-
tion was placed in the asymptotic response of process systems to sinusoidal input vari-
ation. This analysis technique, known as frequency response, was particularly mature at
that time in other areas such as telecommunication systems. At this time the first book in
process control was published by N. Ceaglske (Automatic Process Control for Chemical
Engineers, New York: Wiley, 1956) from the Department of Chemical Engineering,
University of Minnesota, Minneapolis. Significant progress was also reported in using
electrical analogs to simulate the response of common process units such as distilla-
tion columns and heat exchangers (see, for instance, work reported in I&ECR, p. 1035,
June 1956, by J. M. Mozley from E. I. du Pont de Nemours and Company). Based on a
technique that used electrical components to simulate the response of process systems
using electrical analog circuits, this work offered the opportunity to simulate relatively
complex dynamics efficiently when computers were expensive and relatively awkward to
use (judged from today’s point of view). The side effect of this development was that
time-domain analysis techniques were studied by chemical engineers in parallel with fre-
quency-response techniques.
14 Introduction
The theory of chemical process control reached maturity in the 1960s and early 1970s with
numerous applications and theoretical developments and the publication of several books
that facilitated wider understanding of the field:
• P. S. Buckley, Techniques of Process Control, New York: John Wiley & Sons, 1964
• P. Harriott, Process Control, New York: McGraw-Hill, 1964
• D. R. Coughanowr and L. B. Koppel, Process Systems Analysis and Control, New York:
McGraw-Hill Company, 1965
• P. W. Murrill, Automatic Control of Processes, Scranton, PA: Intext Educational
Publishers, 1967
• E. F. Johnson, Automatic Process Control, New York: McGraw Hill Company, 1967
• J. M. Douglas, Process Dynamics and Control, volumes 1 & 2, Englewood Cliffs, NJ:
Prentice-Hall Inc., 1972
• W. L. Luyben, Process Modeling Simulation and Control for Chemical Engineers, New
York: McGraw-Hill Company, 1973.
All these books are mainly concentrated around what is now called classical automatic
control theory, which is mainly characterized by the use of Laplace and frequency-domain
analysis of control systems. These particular methods of control systems analysis and design
were used extensively in electrical engineering and communication systems in the 1940s and
1950s and offered, even in the 1960s, a strong and comprehensive set of tools that could be
used efficiently to analyze and design process control systems. The time-domain approach,
which is based on using sets of first-order differential equations to describe system dynamics
(including feedback control systems), is an alternative method of formulating control sys-
tems theory, which has become known as the modern or state-space approach. Both classical
and modern approaches will be presented in this book in a parallel way and the reader will
soon realize that both approaches have their advantages and disadvantages. In any case, a
process control engineer must be able to understand and use system analysis and controller
synthesis tools from both approaches.
In the past 40 years, process control theory is considered a well-established field of
study and research in chemical engineering and has expanded its areas of application.
Automatic process control systems are in extensive use in the chemical process industries.
Several books were published in the 1980s and 1990s (a list of some of the most influen-
tial textbooks is presented at the end of the chapter) and the field can now be considered
as a fairly mature field. In addition, computer software tools that are now available make
application of the theory for controller synthesis, closed-loop system simulation and con-
troller prototyping an easy task. Technologies such as digitalization and the Internet of
Things (IoT) are expected to have a real impact in the years to come. Nonlinear con-
trol, model-predictive control and real-time optimization are now routinely installed in
advanced control systems. These developments have shifted the focus and the educational
needs in the field of process control. Future chemical engineers must have a solid under-
standing of the underlying theory but more importantly they must be able to combine this
Terms and concepts 15
knowledge with software tools in order effectively to design controllers for large-scale,
interacting and integrated plants.
LEARNING SUMMARY
Automatic control is an interdisciplinary field of study with many common elements between
different engineering disciplines such as electrical, mechanical and chemical engineering.
Chemical process control is a well-established field in chemical engineering and it involves
the application of automatic control in the chemical, petrochemical, food and related indus-
tries. Automatic operation is achieved mainly through feedback control systems, whose main
constituting elements are shown in Figure 1.3.
The controlled variable is measured continuously through a sensor and is fed back to the
controller, which can be an operator or a computer system. Controller actions are imple-
mented to the process through a final control element, which, in the case of chemical process
industries, is usually a control valve.
Chemical process control has a history of roughly one century but is advancing at a fast
pace by taking advantage of innovations in other fields, mainly related to computer soft-
ware and hardware. Advanced control algorithms are routinely implemented at several levels
of the control system. This book presents both approaches to process control theory: the
classical approach that is based on the frequency response and the Laplace domain, and
the modern approach that is based on the state-space approach in the time domain. It also
introduces basic software tools in commercial computer software necessary for the efficient
implementation of theory into practice.
TERMS AND CONCEPTS

Block diagram. A diagram that indicates the flow of information around the system where
each block denotes a component and the arrows the interactions among the different
components
Closed-loop system. A system with components connected in a circular (loop) structure. The
term closed-loop system is used in the context of feedback control, where a controller is
connected with the final control element, the process and the sensor in a circular pattern.
Controlled variable. The process variable that we want to maintain at a particular desired
value.
Controller. A device (or a human operator) that corrects any mismatch between the set point
and the controlled variable by adjusting the manipulated variable.
Disturbance variable. Any external variable that can affect the process but is not under our
control.
Error. The difference between the value of the set point and the value of the measured variable.
Manipulated variable. Process variable that is adjusted to bring the controlled variable back
or close to the set point.
Set point. The desired value of the controlled variable.
16 Introduction
FURTHER READING
The books that follow are the most frequently used textbooks in chemical process control
and presented according to the date of publication of their most recent edition.
Stephanopoulos, G., Chemical Process Control. Englewood Cliffs, NJ: Prentice Hall, 1984.
Luyben, W. L., Process Modeling Simulation and Control for Chemical Engineers, 2nd edn.
New York: McGraw-Hill, 1990.
Ogunnaike, B. A. and Ray, W. H., Process Dynamics, Modeling and Control. New York:
Oxford, 1994.
Shinskey, F. G., Process Control Systems: Application, Design, and Tuning. New York:
McGraw Hill, 1996.
Luyben, M. L. and Luyben, W. L., Essentials of Process Control. New York: McGraw-Hill, 1997.
Marlin, T. E., Process Control: Designing Processes and Control Systems for Dynamic
Performance, 2nd edn. New York: McGraw Hill, 2000.
Chau, P. C., Process Control, A First Course with MATLAB. Cambridge, UK: Cambridge
University Press, 2002.
Brosilow C. and Joseph, B., Techniques of Model-based Control. Engelwood Cliffs, NJ:
Prentice Hall, 2002.
Bequette, B. W., Process Control: Modeling, Design and Simulation. Engelwood Cliffs, NJ:
Prentice Hall, 2003.
Smith, C. A. and Corripio, A. B., Principles and Practice of Automatic Process Control, 3rd
edn. New York: Wiley, 2005.
Riggs, J. B. and Karim, M. N., Chemical and Bio-Process Control, 3rd edn. Boston, MA:
Pearson International Edition, 2006.
Coughanowr, D. R. and LeBlanc, S., Process Systems Analysis and Control, 3rd edn. New
York: McGraw-Hill Education, 2009.
Svrcek, W. Y., Mahoney, D. P. and Young, B. R., A Real-Time Approach to Process Control,
3rd edn. Chichester, UK: Wiley, 2013.
Seborg, D. E., Edgar, T. F., Mellichamp, D. A. and Doyle III, F. J., Process Dynamics and
Control, 4th edn. New York: Wiley, 2016.
Rohani, S. (ed.), Coulson and Richardson’s Chemical Engineering Volume 3B: Process Control,
4th edn. Cambridge, MA: Butterworth-Heinemann, 2017.
Information about standard symbols and convections used in P&IDs and sensor and actua-
tor technology can be found in the following books.
Meier, F. A. & Meier, C. A., Instrumentation and Control Systems Documentation. Research
Triangle Park, NC: International Society of Automation (ISA), 2011.
Hughes, T. A., Measurement and Control Basics, 5th edn. Research Triangle Park, NC:
International Society of Automation (ISA), 2015.
Dunn, W. C., Fundamentals of Industrial Instrumentation and Control. New York: Mc-Graw
Hill Education, 2018.
Toghraei, M., Piping and Instrumentation Diagram Development. New Jersey: John Wiley &
Sons, Inc., 2019.
Problems 17
PROBLEMS
1.1 Draw a block diagram for the control system generated when you drive an automobile.
What are the “sensors” that you use and what are the final control elements? What are
the disturbances?
1.2 Draw a block diagram for an automobile cruise control system. Select some informa-
tion from the internet about autonomous driving and list the sensors that you think
you need to achieve such a task.
1.3 Draw a block diagram for the control system that maintains the temperature in a home
refrigerator. Identify the potential disturbances.
1.4 Draw a block diagram for the control system for a home air-conditioning system. Iden-
tify the potential disturbances.
1.5 Collect information from the internet concerning the basic principles of temperature
sensors.
1.6 Collect information from the internet concerning the basic principles of pressure and
differential pressure sensors.
1.7 Collect information from the internet concerning the basic principles of flow metering
devices.
1.8 Suppose you were to develop the first sensor for measuring the liquid level in a tank
or the pressure in a closed vessel for gas storage. Think about the general principles
on which we can base the development of such a sensor. Search the internet to collect
information about the most used sensors for liquid-level or pressure measurement.
1.9 Suppose you have been asked to develop a system that follows the Sun and positions
a surface always to face the Sun (so that the Sun beam direction is normal to the cap-
turing surface). Think about the characteristics of the sensor that detects the position
of the Sun and the necessary actuators to move the surface. Search the internet, collect
information about the systems used in solar tracking control systems for photovoltaic
panels and prepare a short presentation.
1.10 Compare Figures 1.9 and 1.10 and discuss the common elements.
2 Dynamic Models for Chemical Process Systems
In this chapter we discuss the main principles of modeling the dynamics of chemical pro-
cesses. We start by stating the general inventory rate equation. The conservation of mass
and energy, which are special applications of the inventory rate equation, are then presented
and a number of representative examples are analyzed. The development of an appropriate
dynamic process model is the first step in understanding the underlying phenomena that
result in the transient behavior of a process system. Modeling will be essential in developing
appropriate control laws and in understanding the effect that a controller has on process
transients.
STUDY OBJECTIVES
After studying this chapter you should be able to:
• Possess basic understanding and skills in dynamic model development.
• Apply the appropriate inventory equations to model the dynamics of representative
chemical processes.
• Apply differential equation solvers of software packages like MATLAB or Maple, to
simulate the dynamic behavior of chemical process systems.
2.1 Introduction
A mathematical model is a representation of our knowledge about a physical system, which
is “translated” into a set of mathematical equations. The aim is to use the model to increase
our understanding of the real system’s behavior (model simulation can provide insights), and
also to design and optimize the operation of the process.
Dynamic models provide a quantitative description of the transient behavior of a process,
in addition to its steady-state characteristics. This is very useful in order to select proper
operating conditions for the process, so that undesirable transients are avoided. It is also
very important for the design of controllers, because, as we will see in subsequent chapters, a
controller can modify the dynamic behavior of a process, for better or for worse.
18
Another random document with
no related content on Scribd:
PIED DUCK.
Fuligula labradora, Gmel.

PLATE CCCXXXII. Male and Female.
Although no birds of this species occurred to me when I was in

Labrador, my son, John Woodhouse, and the young friends who
accompanied him on the 28th of July 1833 to Blanc Sablon, found,
placed on the top of the low tangled fir-bushes, several deserted
nests, which from the report of the English clerk of the fishing
establishment there, we learned to belong to the Pied Duck. They
had much the appearance of those of the Eider Duck, being very
large, formed externally of fir twigs, internally of dried grass, and
lined with down. It would thus seem that the Pied Duck breeds
earlier than most of its tribe. It is surprising that this species is not
mentioned by Dr Richardson in the Fauna Boreali-Americana, as it
is a very hardy bird, and is met with along the coasts of Nova Scotia,
Maine, and Massachusetts, during the most severe cold of winter.
My friend Professor MacCulloch of Pictou has procured several in
his immediate neighbourhood; and the Honourable Daniel Webster
of Boston sent me a fine pair killed by himself, on the Vineyard
Islands, on the coast of Massachusetts, from which I made the
drawing for the plate before you. The female has not, I believe, been
hitherto figured; yet the one represented was not an old bird.
The range of this species along our shores does not extend farther
southward than Chesapeake Bay, where I have seen some near the
influx of the St James River. I have also met with several in the
Baltimore market. Along the coast of New Jersey and Long Island it
occurs in greater or less number every year. It also at times enters
the Delaware River in Pennsylvania, and ascends that stream at
least as far as Philadelphia. A bird-stuffer whom I knew at Camden
had many fine specimens, all of which he had procured by baiting
fish-hooks with the common mussel, on a “trot-line” sunk a few feet
beneath the surface, but on which he never found one alive, on
account of the manner in which these Ducks dive and flounder when
securely hooked. All the specimens which I saw with this person,
male and female, were in perfect plumage; and I have not enjoyed
opportunities of seeing the changes which this species undergoes.
The Pied Duck seems to be a truly marine bird, seldom entering
rivers unless urged by stress of weather. It procures its food by
diving amidst the rolling surf over sand or mud bars; although at
times it comes along the shore, and searches in the manner of the
Spoonbill Duck. Its usual fare consists of small shell-fish, fry, and
various kinds of sea-weeds, along with which it swallows much sand
and gravel. Its flight is swift, and its wings emit a whistling sound. It is
usually seen in flocks of from seven to ten, probably the members of
one family.
Anas labradora, Lath. Ind. Ornith. vol. ii. p. 859.

Pied Duck, Anas labradora, Wils. Amer. Ornith. vol. viii. p. 91, pl. 69, fig. 6.
male.
Fuligula labradora, Ch. Bonaparte, Synopsis of Birds of United States, p.
391.
Pied Duck, Nuttall, Manual, vol. ii. p. 428.
Adult Male. Plate CCCXXXII. Fig. 1.

Bill nearly as long as the head, rather broader than high at the base,
the sides nearly parallel, but at the end enlarged by soft
membranous expansions to the upper mandible. The latter has the
dorsal outline at first straight and declinate, then direct and slightly
convex, at the extremity decurved; the ridge broad at the base,
convex toward the end; the sides sloping at the base, then convex,
the extremity broad and rounded, the unguis broadly obovate; the
margins soft, expanded toward the end, and with about 50 lamellæ,
of which the anterior are inconspicuous. Nasal groove oblong,
nostrils linear-oblong, sub-basal near the ridge. Lower mandible
flattened, curved upwards, with the angle very long and narrow, the
dorsal line very short, and nearly straight, the nearly erect edges with
about 30 large and prominent lamellæ; the unguis very broad.
Head of moderate size, oblong, compressed. Eyes small. Neck
rather short and thick. Body full, depressed. Feet very short, strong,
placed rather far behind; tarsus very short, compressed, with two
anterior series of rather small scutella, the sides and back part
reticulated with angular scales. Hind toe very small, with a free
membrane beneath; outer anterior toes double the length of the
tarsus, and nearly equal, the inner much shorter, and with a broad
marginal membrane. Claws small, slightly arched, compressed,
rather acute.
Plumage dense, soft, blended; feathers of the head and neck small,
oblong; those on the lower part of the cheeks very stiff, having the
terminal filaments more or less united into a horny plate. Wings
short, of moderate breadth, concave, acute; primary quills curved,
strong, tapering, the second very slightly longer than the first, the
rest rapidly graduated; secondary quills broad and rounded, the
inner elongated and tapering. Tail very short, much rounded, of
fourteen tapering feathers.
Bill with the basal space between the nostrils running into a rounded
point in the middle, pale greyish-blue; the sides of the base, and the
edges of both mandibles for two-thirds of their length, dull pale
orange; the rest of the bill black. Iris reddish-hazel. Feet light
greyish-blue, webs and claws dusky. Head and upper half of neck
white, excepting an elongated black patch on the top of the head and
nape. Below the middle of the neck is a black ring, from the hind part
of which proceeds a longitudinal band of the same colour, gradually
becoming wider on the back and rump; below the black ring
anteriorly is a broad band of white, passing backwards on each side
so as to include the scapulars. All the under parts black, excepting
the axillaries and lower wing-coverts. Upper wing-coverts and
secondary quills white, some of the inner quills with a narrow
external black margin; alula, primary coverts, and primary quills,
brownish-black. Tail brownish-black, tinged with grey, the shafts
black; upper tail-coverts dusky, minutely dotted with reddish-brown.
Length to end of tail 20 inches, to end of claws 22 1/2, to end of
wings 18 1/4; extent of wings 30; wing from flexure 9 1/4; tail 3 5/8; bill
along the ridge 1 3/4, along the edge of lower mandible 2 3/8; tarsus
1 1/2; middle toe 2 3/8, its claw 3/8, hind toe 4 1/2/8, its claw 1 1/2/8;
outer toe and claw slightly longer than middle: inner toe 1 7/8, its claw
2 1/2/ . Weight 1 lb. 14 1/2 oz.
8
Female. Plate CCCXXXII. Fig. 2.

The female is less than the male. The bill, iris, and feet are coloured
as in the male; sides of the forehead white (not in the figure, it having
been taken from a young bird). The general colour is brownish-grey,
darker on the head, cheeks, back, rump, and abdomen, of a lighter
tint approaching to ash-grey, on the throat, breast, wing-coverts, and
inner secondaries, which are margined externally with black; seven
or eight of the secondary quills white; the primaries and tail-feathers
as in the male.
Length to end of tail 18 1/4 inches, to end of claws 19 3/8, to end of
wings 17; extent of wings 29; wing from flexure 9; tail 3 1/2; bill along
the ridge 1 5/8, along the edge of lower mandible 2 1/8; tarsus 1 1/2;
hind toe and claw 3/4; middle toe and claw 2 1/2. Weight 1 lb. 1 oz.
GREEN HERON.
Ardea virescens, Linn.

PLATE CCCXXXIII. Male and Young.
This species is more generally known than any of our Herons, it

being very extensively dispersed in spring, summer, and early
autumn. It ranges along our many rivers to great distances from the
sea, being common on the Missouri and its branches, from which it
spreads to all such localities as are favourable to its habits. To the
north of the United States, however, it is very seldom seen, it being
of rare occurrence even in Nova Scotia. At the approach of winter it
retires to the Florida and Lower Louisiana, where individuals,
however, reside all the year, and many remove southward beyond
the limits of our country. I have observed their return in early spring,
when arriving in flocks of from twenty to fifty individuals. They would
plunge downwards from their elevated line of march, cutting various
zigzags, until they would all simultaneously alight on the tops of the
trees or bushes of some swampy place, or on the borders of miry
ponds. These halts took place pretty regularly about an hour after
sun-rise. The day was occupied by them, as well as by some other
species, especially the Blue, the Yellow-crowned, and Night Herons,
all of which at this period travelled eastward, in resting, cleansing
their bodies, and searching for food. When the sun approached the
western horizon, they would at once ascend in the air, arrange their
lines, and commence their flight, which, I have no doubt, continued
all night. You may therefore, good Reader, conclude that Herons are
not only diurnal habits when feeding, but also able to travel at night
when the powerful impulse of migration urges them from one portion
of the country to another. But although on their northward journey,
the Green Herons travel in flocks, it is a curious fact, that, unlike our
smaller Waders, Ducks, Geese, and Cranes, they usually return
southward at the approach of winter, singly or in very small flocks.
Stagnant pools or bayous, and the margins of the most limpid
streams, are alike resorted to by this species for the purpose of
procuring food. It is little alarmed by the presence of man, and you
may often see it close to houses, on the mill-dams, or even raising
its brood on the trees of gardens. This is often the case in the
suburbs of Charleston in South Carolina, where I have seen several
nests on the same live oak in the grounds of the Honourable Joel
Poinset, as well as in those of other cities of the Southern States.
The gentleness, or as many would say, the stupidity of this bird is
truly remarkable, for it will at times allow you to approach within a
few paces, looking as unconcernedly upon you as the House
Sparrow is wont to do in the streets of London.
Although they not unfrequently breed in single pairs; they also
associate, not only forming communities of their own kind, but
mingling with the larger species of their tribe, and with the Boat-tailed
Grakles, and other birds. On the 23d May 1831, I found two nests of
the Green Heron on one of the Florida Keys, close to some of Ardea
rufescens and A. cœrulea. Now and then a dozen or more of their
nests are found on a bunch of vines in the middle of a pond, and
placed within two or three feet of the water; while in other cases,
they place their tenements on the highest branches of tall cypresses.
In our Middle Districts, however, and especially at some distance
from the sea, it is very seldom that more than a single nest is seen in
one locality.
The nest of the Green Heron, like that of almost every other species
of the tribe, is flat and composed of sticks, loosely arranged, among
which are sometimes green twigs with their leaves still attached. The
eggs are three or four, seldom more, an inch and three-eighths in
length, an inch and one-eighth in breadth, nearly equally rounded at
both ends, and of a delicate sea-green colour. According to the
locality, they are deposited from the middle of March to the beginning
of June. In the Southern States, two broods are frequently reared,
but in the Middle and Northern Districts, seldom more than one.
The young, which are at first of a deep livid colour, sparingly covered
here and there, and more especially about the head, with longish
tufts of soft hair-like down, of a brownish colour, remain in the nest
until nearly able to fly; but if disturbed, at once leave their couch, and
scramble along the branches, clinging to them with their feet, so as
not to be easily drawn off.
After the spring migration is over, the flight of this species is rather
feeble, and when they are passing from one spot to another, they
frequently use a stronger flap of their wings at intervals. On such
occasions, they scarcely contract their neck; but when travelling to a
considerable distance, they draw it in like all other species of the
tribe, and advance with regular and firm movements of their wings.
When alighting to rest, they come down with such force, that their
passage causes a rustling sound like that produced by birds of prey
when pouncing on their quarry, and on perching they stretch up their
neck and jerk their tail repeatedly for some time, as they are also
wont to do on any other occasion when alarmed.
The Green Herons feed all day long, but, as I think, rarely at night.
Their food consists of frogs, fishes, snails, tadpoles, water lizards,
crabs, and small quadrupeds, all of which they procure without much
exertion, they being abundant in the places to which they usually
resort. Their gait is light but firm. During the love-season they exhibit
many curious gestures, erecting all the feathers of their neck,
swelling their throat, and uttering a rough guttural note like qua, qua,
several times repeated by the male as he struts before the female.
This note is also usually emitted when they are started, but when
fairly on wing they proceed in silence. The flesh of this species
affords tolerable eating, and Green Herons are not unfrequently
seen in the markets of our Southern cities, especially of New
Orleans.
The young attain their full beauty in the second spring, but continue
to grow for at least another year. The changes which they exhibit,
although by no means so remarkable as those of Ardea rufescens
and A. cœrulea, have proved sufficient to cause mistakes among
authors who had nothing but skins on which to found their decisions.
I have given figures of an adult in full plumage, and of an immature
bird, to enable you to judge how carefully Nature ought to be studied
to enable you to keep free of mistakes.
Ardea virescens, Linn. Syst Nat. vol i. p. 238.—Lath. Ind. Ornith. vol. ii. p.
684.—Ch. Bonaparte, Synopsis of Birds of United States, p. 307.
Green Heron, Ardea virescens, Wils. Amer. Ornith. vol. viii. p. 97, pl. 61,
fig. 1.—Nuttall, Manual, vol. ii. p. 63.
Adult Male. Plate CCCXXXIII. Fig. 1.

Bill longer than the head, straight, rather slender, tapering to a very
acute point, higher than broad at the base, compressed towards the
end. Upper mandible with its dorsal line very slightly arched, the
ridge broad and rather flattened at the base, narrowed towards the
end, the sides sloping, erect towards the edges, which are sharp and
direct, the tip acute. Nasal depression long, with a groove extending
to near the point; nostrils basal, linear, longitudinal. Lower mandible
with the angle very long and narrow, the dorsal line sloping upwards,
the sides sloping outwards and nearly flat, the edges sharp, the tip
acuminate.
Head oblong, much compressed. Neck long. Body very slender,
much compressed. Feet rather long, moderately stout; tibia bare for
about an inch; tarsi of moderate length, covered with hexagonal
scales of which some of the anterior are much larger and
scutelliform. Toes rather long and slender, with numerous scutella
above; hind toe stout, second and fourth nearly equal, third much
longer; claws rather long, slender, arched, compressed, acute, that
of middle toe expanded and serrated on the inner edge.
A large space extending from the bill to behind the eye bare.
Plumage very soft, loose, and blended; feathers of the hind head
elongated and erectile, as are those of the neck generally, but
especially of its hind and lower anterior parts; of the fore part of the
back much elongated and acuminate, scapulars very large. Wings
short, very broad, rounded; second and third quills equal and
longest, first and fourth equal and but slightly shorter, the rest slowly
graduated; secondaries broad and rounded. Tail very short, even, of
twelve, broad, soft feathers.
Bill greenish-black above, bright yellow beneath. Iris and bare part
about the eye also bright yellow. Feet, greenish-yellow, claws dusky.
Upper part of the head and nape glossy deep green. Neck purplish-
red, tinged with lilac behind, and having anteriorly a longitudinal
band of white, spotted with dusky-brown; a similar white band along
the base of lower mandible to beyond the eye. Elongated feathers of
the back greyish-green, in some lights bluish-grey, with the shafts
bluish-white; the rest of the back similar; the upper tail-coverts and
tail bluish-green; the lateral feathers slightly margined with white.
Scapulars, wing-coverts, and inner secondaries, deep glossy green,
bordered with yellowish-white; primary quills and outer secondaries
greyish-blue tinged with green. Lower parts pale purplish-brown
tinged with grey; axillary feathers purplish-grey, as are some of the
lower wing-coverts; lower tail-coverts greyish-white.
Length to end of tail 17 3/4 inches, to end of wings 17 1/2, to end of
claws 24, to carpal joint 11 1/4; extent of wings 27; wing from flexure
7 5/8; tail 3 3/4; bill along the ridge 2 1/4, along the edge of lower
mandible 3 1/4; bare part of tibia 10/12; tarsus 2; hind toe 7/8, its claw
1/2; middle toe 1 1/8, its claw 2 1/2/8; inner toe 1 5/8, its claw 1/4; outer
toe 1 3/8, its claw 1/4. Weight 7 1/2 oz.
The Female is considerably smaller, but otherwise similar.
Length to end of tail 17 inches, to end of wings 17, to end of claws
21 3/4; extent of wings 25. Weight 6 1/4 oz.
Young fully fledged. Plate CCCXXXIII.

The bill dull greyish-green, the lower mandible lighter: bare space
around the eye greenish-blue, with the exception of a streak of
yellow at the upper part. Iris yellow. Feet greenish-yellow, duller than
in the adult. The hind neck light brownish-red, the fore part of the
neck and all the under parts white, longitudinally streaked with
brownish-red, some of the long feathers on the sides of the neck
also white. At this age there are no elongated feathers on the back,
which is greenish-blue, as well as the scapulars, and tail-feathers.
Wing as in the adult, but the smaller feathers on its anterior part
more red, the coverts with a small triangular tip of white, and the
quills narrowly tipped and margined with the same.
Length to end of tail 17 1/2 inches, to end of wings 17, to end of
claws 23; extent of wings 25. Weight 6 1/2 oz.
The roof of the mouth is anteriorly a little concave, with a median
prominent line; the palate convex; the lower jaw with a kind of joint
about an inch from the base, its intercrural membrane or skin very
extensile. The tongue is 1 7/12 inch long, very slender, trigonal,
emarginate at the base, with a groove along the middle, and pointed.
Posterior apertures of nares linear, 1/2 inch long. Œsophagus, a, b c,
10 inches long, its walls delicate, its diameter at the upper part 1 1/4
inch, gradually contracting to 1/2 inch at its entrance into the thorax.
The lobes of the liver unequal, the right 1 inch 5 twelfths long, the left
11 twelfths; the gall-bladder large, 7 twelfths long. The stomach, c, d,
is membranous, of an oblong form, 9 twelfths long, 10 twelfths in
breadth; its tendons elliptical, twelfths by 3 twelfths. The
proventriculus, c c, 9 twelfths long, with a complete belt of oblong
glandules. There is a small roundish pyloric lobe e. Intestine, f, g, 2
feet 11 inches long, its diameter uniform, 1 twelfth, or about the
thickness of a crow’s quill. Rectum enlarged to 3 twelfths, and 3 1/4
inches long, its cœcal extremity rounded, and only one-twelfth long.
The trachea is 7 1/4 inches long, of nearly uniform diameter,
averaging 2 twelfths; the rings 160, nearly circular and ossified. The
bronchial half-rings about 18. The lateral muscles are very
inconspicuous; sterno-tracheals; and a pair of inferior laryngeal,
going to the first bronchial rings.
The Herons generally differ from the other Grallæ in having the
œsophagus much wider and similar to that of the fish-eating
palmipedes; the stomach in a manner membranous, like that of the
rapacious land-birds, without lateral muscles or strung epithelium;
the intestine extremely slender, and the anterior extremity of the
large intestine or rectum furnished with a single cœcum, in place of
two, as in almost all other birds.
BLACK-BELLIED PLOVER.
Charadrius Helveticus, Wilson.

PLATE CCCXXXIV. Male in Summer, Young, and Adult in
Winter.
This beautiful bird makes its appearance on our southern coasts in

the beginning of April, as I had many opportunities of observing in
the course of my journey along the shores of the Gulf of Mexico, in
the spring of 1837. Instead of being congregated in large flocks, as is
the case during their southward migration in autumn, they are seen
coming in small numbers, but at short intervals, so as almost to form
a continuous line. They travel chiefly by night, and rest for a great
part of the day along the margins of the sea, either reposing on the
sands in the sunshine, or searching the beaches for food. After dusk
their well-known cries give note of their passage, but by day they
remain silent, even when forced to betake themselves to flight. On
such occasions they generally wheel over the waters, and not
unfrequently return to the spot which they had at first selected. I
have traced this species along the whole of our eastern coast, and
beyond it to the rugged shores of Labrador, where my party procured
a few, on the moss-covered rocks, although we did not then find any
nests, and where some young birds were obtained in the beginning
of August.
Individuals of this species spend the summer months in the
mountainous parts of Maryland, Pennsylvania, and Connecticut,
where they breed. I found their nests near the waters of the
Delaware and the Perkioming Creek, when I resided in the first of
these States, and in the same localities as those of Totanus
Bartramius, as well as in ploughed fields. The nest is merely a slight
hollow with a few blades of grass. The eggs are four, an inch and
seven and a-half eighths in length, an inch and three-eighths in their
greatest breadth; their ground-colour yellowish-white, tinged with
olivaceous, and pretty generally covered with blotches and dots of
light brown, and pale purple, the markings being more abundant
toward the small end. Their form is similar to that of the egg of the
Guillemot, that is, broadly rounded at the large end, then tapering,
with the sides nearly straight, and the narrow end rounded. When
sitting, these birds will remain until they are almost trodden upon. On
being started, they fly off a few yards, alight running, and use all the
artifices employed on such occasions to induce the intruder to set
out in pursuit. The young leave the nest almost immediately after
they are hatched, and should one approach them the parents
become very clamorous, and fly around until they are assured of the
safety of their brood, when they take a long flight, and disappear for
a time. Unless during the breeding season, they are exceedingly
shy; but their anxiety for their young renders them forgetful of the
danger which they incur in approaching man. The young, when two
or three weeks old, run with great celerity, and squat in perfect
silence when apprehensive of danger. When they are able to fly,
several families unite, and betake themselves to the sea-shore,
where other flocks gradually arrive, until at length, on the approach
of cold weather, almost all of them begin to move southward.
Although the great body of these Plovers pass beyond the limits of
the United States, some remain on the shores of the Floridas during
winter. In their habits they are more maritime than the Golden
Plovers, which, when migrating, generally advance over the land.
The flight of this bird is swift, strong, and well sustained. When
roaming over large sand-bars, they move in compact bodies, whirling
round, and suddenly veering, so as alternately to exhibit their upper
and lower parts. At this time old and young are intermixed, and many
of the former have lost the black so conspicuous on the neck and
breast in summer. During winter, or as long as they frequent the sea-
shore, they feed on marine insects, worms, and small shell-fish; and
when they are in the interior, on grasshoppers and other insects, as
well as berries of various kinds, on which they fatten so as to
become tolerably good eating.
This species is known in Pennsylvania by the name of Whistling
Field Plover, suggested by the loud and modulated cries which it
emits during the love-season. In the Eastern States, as well as in
Kentucky, it is called the Bull-head; but in the South its most
common appellation is Black-bellied Plover. I have seen it, though
sparingly, along the shores of the Ohio, probably during its passage
from the north.
As its habits agree with those of the Plovers generally, and its form is
similar to that of the Golden Plover and other species, the only
difference being the presence of a rudimentary hind toe, it was
scarcely necessary to distinguish it generically from Charadrius, as
many recent authors have done.
Tringa helvetica, Linn. Syst. Nat. vol. i. p. 250.—Lath. Ind. Ornith. vol. ii. p.
728.
Tringa squatarola, Linn. Syst. Nat. vol. i. p. 252.—Lath. Ind. Ornith. vol. ii.
p. 729.
Black-bellied Plover, Charadrius helveticus, Wils. Amer. Ornith. vol. vii.
p. 41, pl. 59, fig. 4. Summer.
Charadrius helveticus, Ch. Bonaparte, Synopsis of Birds of United States,
p. 298.
Grey Lapwing, Vanellus melanogaster, Richards. and Swains. Fauna
Bor.-Amer. vol. ii. p. 370.
Black-bellied or Swiss Plover, Nuttall, Manual, vol. ii. p. 26.
Adult Male in summer. Plate CCCXXXIV. Fig. 1.

Bill as long as the head, straight, somewhat compressed, stout.
Upper mandible with the dorsal line straight and slightly sloping for
more than half its length, then bulging a little and arched to the tip,
which is rather acute, the sides flat and sloping at the base, convex
towards the end, where the edges are sharp and inclinate. Nasal
groove extended to a little more than half the length of the mandible;
nostrils sub-basal, linear, open and pervious. Lower mandible with
the angle rather long and narrow, the sides at the base erect and
nearly flat, the dorsal line ascending and slightly convex, the edges
sharp and involute towards the narrow tip.
Head of moderate size, roundish, the forehead much rounded. Eyes
large. Neck rather short. Body ovate, rather full. Feet rather long,
slender; tibia bare for a considerable space; tarsus rather
compressed, covered all round with reticulated hexagonal scales;
toes of moderate length, slender; the first extremely diminutive, with
an equally minute claw; the second shorter than the fourth, which is
much exceeded by the third; the anterior toes are rather broadly
marginate, the web between the third and fourth extending to the
second joint of the latter, that between it and the second smaller.
Claws small, compressed, slightly arched, acute.
Plumage soft, blended, the feathers broad and rounded. Wings long
and pointed, primary quills tapering, the first longest, the second a
quarter of an inch shorter, the rest rapidly graduated; secondaries,
short, broad, obliquely rounded; the inner tapering and elongated.
Tail rather short, slightly rounded, of twelve rounded feathers.
Bill and claws black. Iris and feet greyish-black. Forehead yellowish-
white, the rest of the head and the hind neck greyish-white, spotted
with dusky. The upper parts are variegated with black, yellowish-
brown, and white, the feathers being tipped with the latter. The hind
part of the rump, the upper tail-coverts, and the tail-feathers, white,
transversely barred with brownish-black, the tail tipped with white,
and having four dark bars on the middle feathers, and seven or eight
on the outer webs of the rest. Alula, primary coverts, and primary
quills brownish-black, the coverts terminally margined with white;
shafts of the primaries about the middle, and part of the inner web
towards the base, white; the inner six with a white patch on the outer
web towards the base, and margined with white externally; the outer
secondary feathers white at the base and margined with the same;
the inner dusky, with marginal white triangular spots. A narrow ring
round the eye, and a broad longitudinal band on the side of the neck
white; loral space, cheeks, fore part of neck, breast, and axillar
feathers, black; the rest of the lower parts white, the lower primary
coverts grey towards the end.
Length to end of tail 11 3/4 inches, to end of wings 12 1/2, to end of
claws 14; extent of wings 25; wing from flexure 8 1/4; tail 3 1/4; bill
along the ridge 1 1/4, along the edge of lower mandible 1 1/4; bare
part of tibia 3/4; tarsus 2; hind toe 1 1/2/8; middle toe 1 1/8, its claw 1/4;
outer toe and claw 1 1/8; inner toe 1. Weight 6 1/2 oz.
The Female resembles the male, but has the black of the lower parts
of a less deep tint.
Young about a week old. Plate CCCXXXIV. Fig. 2.
Bill and feet dull greenish-brown. Iris brown. The general colour is
pale brownish-yellow, mottled with dusky; a whitish ring round the
eye; tail with a black band, rump whitish, primary quills dusky, the
outer edges of the secondaries whitish.
The young in autumn.
Bill greyish-black; feet bluish-grey. The upper parts brownish-black,
spotted with white, some of the spots yellow; the wings and tail as in
the adult, but the latter tinged with grey, and having eight dark bars
on all the feathers. The fore part and sides of the neck, and the sides
of the body, greyish-white, mottled with brownish-grey; axillary
feathers brownish-black; the rest of the lower parts white.
Adult in winter. Plate CCCXXXIV. Fig. 3.
The adult in winter has the upper parts light greyish-brown, the
margins of the feathers much lighter; the sides and fore parts of the
neck pale grey, with dark grey streaks and spots; lower parts white.
In other respects the colours are as in summer.
In an adult male of this species, there is a double row of papillæ on

the roof of the mouth. The tongue is 1 inch long, slender, tapering,
emarginate and papillate at the base, grooved above, horny on the
back. The œsophagus a, is 5 inches long, at its upper part 4 twelfths
in diameter, enlarged to 1/2 inch on the lower part of the neck. The
proventriculus, b, oblong, its greatest diameter 8 twelfths, its
glandules oblong and about a twelfth in length. The stomach, c, d, e,
f, is a very powerful gizzard of an irregular roundish form, 1 inch 5
twelfths long, 1 inch 3 1/2 twelfths in breadth; its lateral muscles very
large and distinct, the right d, 4 twelfths thick, the left, e, 3 twelfths,
the tendons large; the epithelium thick, longitudinally rugous, and of
a reddish colour. The intestine g, h, i, is 2 feet 2 inches long, its
diameter about 2 twelfths; the cœca 2 inches 2 twelfths long, their
diameter at the base half a twelfth, toward the end 2 twelfths; the
rectum 3 twelfths in diameter, and 2 1/2 inches long.
In the stomach were several shrimps. The lobes of the liver very
unequal, the right being 2 1/4 inches in length, the other 1 4/12. No
gall-bladder.
The trachea is wide, flattened, membranous, 4 twelfths broad at the
upper part, gradually diminishing to 2 twelfths, its rings which are
very slender, about 100. The lateral muscles exceedingly thin, but
becoming more distinct towards the lower part; the sterno-tracheal
slender. Bronchi of moderate length, of about 20 half-rings.

Understanding Process Dynamics and Control 1st Edition Costas Kravaris Download PDF

Uploaded by

Copyright:

Available Formats

Understanding Process Dynamics and Control 1st Edition Costas Kravaris Download PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Understanding Process Dynamics and Control 1st Edition Costas Kravaris Download PDF

Uploaded by

Copyright:

Available Formats

Full download test bank at ebookmeta.

Understanding Process Dynamics and Control 1st

For dowload this book click LINK or Button below

Download More ebooks from https://ebookmeta.com

Advanced Chemical Process Control 1st Edition Morten

Basic Process Engineering Control Paul Serban Agachi

Process Operational Safety and Cybersecurity A Feedback

Theory of Applied Robotics Kinematics Dynamics and

Process Control: Modeling, Design, and Simulation 2nd

Advanced Dynamics Modeling, Duality and Control of

Flight Dynamics, Simulation, and Control: For Rigid and

Theory of Applied Robotics Kinematics Dynamics and

Costas Kravaris is Professor of Chemical Engineering at Texas A&M University,

Ioannis K. Kookos is Professor of Process Systems Engineering, in the Department

Books in the Series

Cambridge University Press is part of the University of Cambridge.

Irene, Michael, Evangeline and Cosmas

Natasa, Kostas and Georgia

Preface page xvii

2 DYNAMIC MODELS FOR CHEMICAL PROCESS SYSTEMS 18

3.4 The Choice of Reference Steady State Affects the Mathematical

4 CONNECTIONS OF FIRST-ORDER SYSTEMS 115

5 SECOND-ORDER SYSTEMS 144

5.5 Frequency Response and Bode Diagrams of Second-Order Systems

6 LINEAR HIGHER-ORDER SYSTEMS 171

7 EIGENVALUE ANALYSIS – ASYMPTOTIC STABILITY 215

Learning Summary 244

8 TRANSFER-FUNCTION ANALYSIS OF THE INPUT–OUTPUT BEHAVIOR 251

9 FREQUENCY RESPONSE 297

10 THE FEEDBACK CONTROL SYSTEM 327

10.4 Further Examples of Process Control Loops 335

11 BLOCK-DIAGRAM REDUCTION AND TRANSIENT-RESPONSE CALCULATION

12 STEADY-STATE AND STABILITY ANALYSIS OF THE CLOSED-LOOP SYSTEM 377

13 STATE-SPACE DESCRIPTION AND ANALYSIS OF THE CLOSED-LOOP SYSTEM 409

13.3 Time Discretization of the Closed-Loop System 422

14 SYSTEMS WITH DEAD TIME 437

15 PARAMETRIC ANALYSIS OF CLOSED-LOOP DYNAMICS – ROOT-LOCUS DIAGRAMS 484

16 OPTIMAL SELECTION OF CONTROLLER PARAMETERS 541

16.3 Calculation of Optimal Controller Parameters for Quadratic Criteria 557

18 MULTI-INPUT–MULTI-OUTPUT SYSTEMS 613

19 SYNTHESIS OF MODEL-BASED FEEDBACK CONTROLLERS 641

19.7 Model-Based Controller Synthesis for MIMO Systems 674

20 CASCADE, RATIO AND FEEDFORWARD CONTROL 683

APPENDIX A LAPLACE TRANSFORM 719

APPENDIX B BASIC MATRIX THEORY 743

Scope of the Book

Organization of the Book

Topic From the book Hours

Topic From the book Hours

Costas Kravaris and Ioannis K. Kookos, October 2020

• Identify the underlying reasons for the need of control systems.

1.1 What is Process Control?

1.2 Feedback Control System: Key Ideas, Concepts and Terminology

• Why does the liquid level vary during everyday operation?

Figure 1.1 A “primitive” level control system.

Figure 1.2 A “primitive” temperature-control system.

Controller Valve Process Sensor

Figure 1.3 Elements of a “primitive” temperature-control system.

Figure 1.4 A computer-based temperature-control system.

4 CONNECTIONS OF FIRST-ORDER SYSTEMS 115

5 SECOND-ORDER SYSTEMS 144

6 LINEAR HIGHER-ORDER SYSTEMS 171

7 EIGENVALUE ANALYSIS – ASYMPTOTIC STABILITY 215

8 TRANSFER-FUNCTION ANALYSIS OF THE INPUT–OUTPUT BEHAVIOR 251

9 FREQUENCY RESPONSE 297

10 THE FEEDBACK CONTROL SYSTEM 327

12 STEADY-STATE AND STABILITY ANALYSIS OF THE CLOSED-LOOP SYSTEM 377

13 STATE-SPACE DESCRIPTION AND ANALYSIS OF THE CLOSED-LOOP SYSTEM 409

14 SYSTEMS WITH DEAD TIME 437

15 PARAMETRIC ANALYSIS OF CLOSED-LOOP DYNAMICS – ROOT-LOCUS DIAGRAMS 484

16 OPTIMAL SELECTION OF CONTROLLER PARAMETERS 541

18 MULTI-INPUT–MULTI-OUTPUT SYSTEMS 613

19 SYNTHESIS OF MODEL-BASED FEEDBACK CONTROLLERS 641

20 CASCADE, RATIO AND FEEDFORWARD CONTROL 683

APPENDIX A LAPLACE TRANSFORM 719

APPENDIX B BASIC MATRIX THEORY 743