Adaptive control /
by Chang C. Hang, Tong H. Lee, Weng K. Ho.
Research Triangle Park, N.C. : Instrument Society of America, c1993.
x, 261 p. : ill. ; 26 cm.
More Details
added author
Research Triangle Park, N.C. : Instrument Society of America, c1993.
catalogue key
Includes bibliographical references and index.
A Look Inside
First Chapter

Chapter One

UNIT 1 Introduction and Overview

Welcome to this ISA's Independent Learning Module (ILM) Adaptive Control. The first unit of this self-study program provides the information needed to proceed through the course.

Learning Objectives--When you have completed this unit, you should:

A. Understand the general organization of the course.

B. Know the course objectives.

C. Know how to proceed through the course.

1-1. Course Coverage

This ILM on the fundamental techniques of adaptive control as applied to industrial process control problems covers the following:

A. A definition of adaptive control, its merits, and the limitations of its applications.

B. The common techniques for auto-tuning, self-tuning, and gain-scheduling of PID controllers.

C. The common techniques for adaptive advanced control, including the Smith predictor, feedforward, poleplacement, minimum-variance, generalized predictive control, and multivariable control.

The course will focus on the adaptive algorithms as well as the associated heuristics that ensure robustness and easy use by plant operators. Such heuristics actually take the software designer longer to incorporate because their computer codes are usually many times longer than the algorithms themselves.

No attempt is made in this ILM to be exhaustive in the presentation of -adaptive control techniques. In fact, certain techniques have been specifically omitted; an example is model reference adaptive control, which is more suitable for low order, well structured servomechanisms as used in robotic and aerospace systems. The unknown nonlinear nature, high order, and nonminimum phase dynamics, the presence of significant dead time, the possibly low signal-to-noise ratio, and the frequent occurrence of load disturbances are what characterize difficult industrial process control problems. We shall focus on practical adaptive control techniques for these problems.

1-2. Purpose

The purpose of this ILM is to present, in easily understood terms, the principles and practices of adaptive process control techniques. Although the course cannot substitute for actual field experience, it is designed to speed up the learning process during field training.

1-3. Audience and Prerequisites

The material in this ILM will be useful to control and instrument engineers who are concerned with the design, installation, and operation of modern process control systems. The course will also be useful to undergraduate and postgraduate students in universities who wish to gain some insight into the practical aspects of adaptive process control.

The prerequisite to this course is a basic knowledge of automatic control, which has been covered by an earlier ILM, Fundamentals of Process Control Theory. It would be useful to have some familiarity with the basic concepts of controller tuning and advanced control, especially when the student studies the material on adaptive advanced control. If necessary, the student can refer to another earlier ILM, Tuning of Industrial Control Systems. In terms of mathematical skills, it is not absolutely necessary for the student to be intimately familiar with the mathematics used in the presentation in order to understand the fundamentals of adaptive control. It is hoped that the barrier that extensive mathematics usually presents to the understanding of automatic control concepts will be kept to a minimum in this ILM.

1-4. Study Materials

This textbook is the only study material required in this course; it is one of ISA's ILM system. It is an independent, stand-alone textbook that is uniquely and specifically designed for self-study.

Contained in Appendix A is a list of suggested reading to provide additional reference and study materials for the student. The student will find it useful to study other ILMs available from ISA that present a broad range of specific applications of instrumentation and process control techniques.

1-5. Organization and Sequence

This ILM is organized into 10 separate units. The next unit presents the functions, classification, and appropriate applications of adaptive control. Unit 3 presents the general structure of a self-tuning controller, which gives the student a macro view of the major components of an adaptive controller and an introduction to the need for supervision for robust adaptive control. Units 4 and 5 are devoted to the principles, properties, and applications of auto-tuning and self-tuning techniques for PID controllers, namely, the relay feedback, pattern recognition, and correlation methods. Relevant controller tuning formulas and their improvement, process modeling from relay feedback, and a design to achieve specified gain margin and phase margin design are presented. Unit 6 is devoted entirely to the adaptive Smith predictor control of processes with long dead time. More details of online recursive parameter estimation methods are presented in Unit 7 to prepare the necessary background for understanding the adaptive techniques for model-based advanced control of single-loop, multi-loop, and multivariable systems in Units 8 and 9. The last unit presents recent developments in knowledge-based control as an extension of adaptive control, neural networks, and adaptive fuzzy control towards the realization of an intelligent control system.

As with all ILMs, the method of instruction is self-study. You select the pace at which you learn best. You may completely skip or browse through some units with which you feel you are intimately familiar and devote more time to other units that contain material new to you.

Each unit is designed in a consistent format with a set of specific learning objectives stated at the very beginning of the unit. Note these learning objectives carefully; the material in the unit will teach to these objectives. Each unit contains examples to illustrate specific concepts and exercises to test your understanding of these concepts. All of these exercises have solutions contained in Appendix B, against which you should check your solution.

1-6. Course Objectives

When you have completed this entire ILM, you should:

* Know when to apply adaptive control to compensate for significant process or disturbance parameter variations. * Be able to decide whether the simpler auto-tuning method or the (open-loop) gain-scheduling technique should be used instead of the more sophisticated, continuous (closed-loop) adaptive control technique. * Be able to select different adaptive control techniques for PID controllers . * Be more confident in applying the adaptive Smith predictor for tighter control of a process with long dead time. * Know when and how to apply adaptive techniques for other advanced control techniques, including cascade, feedforward, pole-placement, minimum variance, and generalized predictive control. * Appreciate the difficulty of and possible solutions for adaptive multivariable control. * Be familiar with new possibilities and developments in intelligent control.

1-7. Course Length

The basic premise of ISA's ILM system is that students learn best when they progress at their own pace. As a result, there will be a significant variation in the amount of time taken by individual students to complete this ILM. On the average, most students will complete this course in 50 to 60 hours.

Copyright © 1993 Instrument Society of America. All rights reserved.

This item was reviewed in:
SciTech Book News, January 1994
To find out how to look for other reviews, please see our guides to finding book reviews in the Sciences or Social Sciences and Humanities.

This information is provided by a service that aggregates data from review sources and other sources that are often consulted by libraries, and readers. The University does not edit this information and merely includes it as a convenience for users. It does not warrant that reviews are accurate. As with any review users should approach reviews critically and where deemed necessary should consult multiple review sources. Any concerns or questions about particular reviews should be directed to the reviewer and/or publisher.

  link to old catalogue

Report a problem