This edition first published 2019 by John Wiley & Sons Singapore Pte. Ltd under exclusive licence granted by National Defence Industry Press for all media and languages (excluding simplified and traditional Chinese) throughout the world (excluding Mainland China), and with a non‐exclusive license for electronic versions in Mainland China.
© 2019 National Defence Industry Press
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by law. Advice on how to obtain permission to reuse material from this title is available at http://www.wiley.com/go/permissions.
The right of Yugeng Xi and Dewei Li to be identified as the authors of this work has been asserted in accordance with law.
Registered Offices
John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, USA
John Wiley & Sons Singapore Pte. Ltd, 1 Fusionopolis Walk, #07‐01 Solaris South Tower, Singapore 138628
Editorial Office
The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK
For details of our global editorial offices, customer services, and more information about Wiley products visit us at www.wiley.com.
Wiley also publishes its books in a variety of electronic formats and by print‐on‐demand. Some content that appears in standard print versions of this book may not be available in other formats.
Limit of Liability/Disclaimer of Warranty
MATLAB® is a trademark of The MathWorks, Inc. and is used with permission. The MathWorks does not warrant the accuracy of the text or exercises in this book. This work's use or discussion of MATLAB® software or related products does not constitute endorsement or sponsorship by The MathWorks of a particular pedagogical approach or particular use of the MATLAB® software.
While the publisher and authors have used their best efforts in preparing this work, they make no representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties, including without limitation any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives, written sales materials, or promotional statements for this work. The fact that an organization, website, or product is referred to in this work as a citation and/or potential source of further information does not mean that the publisher and authors endorse the information or services the organization, website, or product may provide or recommendations it may make. This work is sold with the understanding that the publisher is not engaged in rendering professional services. The advice and strategies contained herein may not be suitable for your situation. You should consult with a specialist where appropriate. Further, readers should be aware that websites listed in this work may have changed or disappeared between when this work was written and when it is read. Neither the publisher nor authors shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.
Library of Congress Cataloging-in-Publication Data
Names: Xi, Yugeng, 1946- author. | Li, Dewei, 1971- author.
Title: Predictive control : fundamentals and developments / Yugeng Xi, Shanghai Jiao Tong University, Shanghai, China, Dewei Li, Shanghai Jiao Tong University, Shanghai, China.
Description: First edition. | Hoboken, NJ : John Wiley & Sons, Inc., [2019] | Includes bibliographical references and index. |
Identifiers: LCCN 2019003713 (print) | LCCN 2019005389 (ebook) | ISBN 9781119119586 (Adobe PDF) | ISBN 9781119119579 (ePub) | ISBN 9781119119548 (hardback)
Subjects: LCSH: Predictive control.
Classification: LCC TJ217.6 (ebook) | LCC TJ217.6 .X53 2019 (print) | DDC 629.8–dc23
LC record available at https://lccn.loc.gov/2019003713
Cover Design: Wiley
Cover Images: © Jesica Leach/EyeEm/Getty Images, © science photo/Shutterstock
Predictive control was developed in the middle of the 1970s. It originally referred to a kind of advanced computer control algorithm that appeared in complex industrial processes. Because of the ability of real‐time solving the optimal control under constraints, it has received great attention from the industrial community and been successfully applied to chemical, oil refining, power, and other industries. Since the 1980s, the algorithms and applications of predictive control have been rapidly expanded. The commercial software of predictive control has undergone several generations of version updating and function expansion. Not only has it been applied in thousands of industrial processes around the world and achieved remarkable economic benefits but also the application fields have expanded rapidly from industrial processes to manufacturing, aerospace, transportation, environment, energy, and so on. Compared with applications, the development of predictive control theory lagged behind. However, since the 1990s, it has rapidly become a hotspot in the control field through adopting novel ideas and powerful tools from a new perspective. Especially the systematic progress in the synthesis theory of stable and robust predictive controllers has deepened the understanding of the essential mechanism of predictive control and constructed a rich theoretical system on qualitative synthesis of predictive control. Nowadays, predictive control is not only favored by industrial communities and regarded as the most representative advanced process control algorithm but it has also become a systematic theory of synthesizing stable and robust controllers characteristic of rolling optimization for uncertain systems.
After entering the twenty‐first century, with the scientific and technological development and the social progress, the requirement for control is becoming higher and higher. Rather than traditionally satisfying the stabilizing design, optimization is much more incorporated in control design in order to achieve a better control performance. In the meantime, optimization is restricted by more and more factors. In addition to traditional physical constraints such as actuator saturation, various constraints brought by technology, safety, economics, and sociality indices should be incorporated. The contradiction between the higher requirements and the more complicated constraints has become a new challenge in many application fields. Because of remarkable achievements in the field of industrial process control, predictive control naturally becomes the first choice to solve this kind of problem. In recent years, many application reports on adopting predictive control for solving various constrained optimization control problems appeared in the fields of advanced manufacturing, energy, environment, aerospace, and medicine, etc., which reflects the expectations of people for this advanced control technology and also motivates more people to become familiar with or master predictive control.
In order to meet the different needs for research and application of predictive control in a wide range of fields, this book tries to give a brief overview on research and application in the field of predictive control through introducing the basic problems and solutions of predictive control, and unscrambling its representative research progress, so as to provide readers with different needs with basic knowledge and background. Reviewing the development of predictive control during the past 40 years, it can be concluded that predictive control has become a diversified disciplinary branch, including various development tracks from theory and methods to applications, with different purposes and characteristics. Each contains rich contents with its own concerns and the fundamental knowledge that needs to become familiar. For example, the research on predictive control applications focuses on how to apply the predictive control principles and algorithms to implementing optimization‐based control for a specific plant. There is a need to become acquainted with predictive control algorithm and its implementation technology, particularly on how to select or formulate a predictive control algorithm for a specific problem, how to solve the optimization problem, and how to tune the parameters, etc. The research on predictive control methods focuses on how to develop effective implementation modes in terms of specific system properties, structural characteristics, or implementation environments, so that predictive control can be practically applicable or applied with a better performance and higher efficiency. It is necessary to understand the existing methods and strategies commonly used in predictive control, as well as the specified structures and research branches deriving from them. It is important that the ideas behind these methods need to be caught from the perspective of information and control, and taken as a reference for extension. The research of predictive control theory focuses on solving difficult or new problems brought on by deep research and background expansion. As a prerequisite, it is necessary to get acquainted with the basic ideas, solving methods, and required tools of mature theory on stable and robust predictive control synthesis. The implied novel ideas for successfully solving various difficulties in the existing literature need to be carefully explored and taken for reference, and the existing difficulties and problems need to be clarified.
In view of the above requirements and characteristics, this book attempts to make a comprehensive introduction to predictive control on the aspects of basic principles and algorithms, system analysis and design, algorithm development and applications according to its historical development process. The purpose is to help readers understand the basic principles and algorithms of predictive control, as well as to get acquainted with the most fundamental contents of predictive control theories, methods, and application techniques, in order to provide the basis and reference for researchers and engineers in the field of predictive control to go deep into theoretical research, to develop high‐level industrial applications, and to extend predictive control to more application fields.
The book consists of five parts. The first part (Chapters 1 and 2 and Sections 5.1 and 5.2) gives a brief overview of the development trajectory of predictive control and introduces its methodological principles and basic algorithms. The second part (Chapters 3 and 4) introduces system analysis and design of classical predictive control algorithms. On the one hand, for the dynamic matrix control (DMC) algorithm, which is based on a nonparametric model and is commonly used in industry, its control mechanism and system performance are analyzed and its parameter tuning is discussed. On the other hand, the relationship between DMC and the generalized predictive control (GPC) algorithm is clarified, and then the quantitative relationships between design parameters and the closed‐loop system performance are uniformly derived. The third part (Chapters 6 and 7) introduces the qualitative synthesis theory of predictive control, including stable predictive controller synthesis and robust predictive controller synthesis. Emphasis is put on unscrambling basic problems as well as solutions, and introducing some representative works. The fourth part (Chapters 8, 9, and Section 5.3) introduces the development of methods and strategies oriented to the characteristics and requirements of predictive control applications. Control structures, optimization concepts and strategies useful in industrial applications, as well as various decomposition algorithms commonly used in networked large‐scale systems, are presented. According to the characteristics of nonlinear systems, some practical and effective algorithms and strategies are also introduced. The fifth part (Chapters 10 and 11) concerns implementation technology and application examples of predictive control, with a detailed introduction of industrial predictive control technology and an overview on predictive control applications in other fields. The universality of predictive control principles is illustrated and potentially extended to general control problems in a dynamic uncertain environment. All of the above parts make a full view of predictive control from general concepts and basic algorithms, quantitative analysis, qualitative synthesis, method and strategy developments to application techniques. They are interdependent and combined organically. The book not only introduces the relevant knowledge on theory and applications but also runs through the methodological principles of predictive control, which will help readers to get rid of the limitations of specific algorithms and problems, deepen their understanding of the essential characteristics of predictive control, and broaden their thinking in predictive control research and application from a higher stand.
This book is based on the book Prediction Control (second edition) (Chinese version), published by the National Defense Industry Press in 2013. The major contents come from the research results of our group in the direction of predictive control during the past 30 years. In order to comprehensively reflect the field of predictive control, some pioneering and representative research work in this field has also been included and unscrambled. At the time of publication of this book, the authors would like to thank the National Natural Science Foundation of China for its long‐standing support for our predictive control research. We would also like to thank our colleagues in academia and industry. It is the helpful discussions and cooperation with them that have enabled us to deepen our understanding and receive new inspiration. Over the past 30 years, colleagues and students in our group have worked together with us in the field of predictive control, and we do appreciate their contributions. Special thanks is given to those whose work directly contributed to this book, including Professor Xiaoming Xu, Professor Shaoyuan Li, doctoral students Hao Sun, Jian Yang, Jian Fang, Jun Zhang, Xiaoning Du, Ning Li, Chungang Zhang, Baocang Ding, Bing Wang, Shu Lin, and Zhao Zhou, and master students Junyi Li, Hanyu Gu, Hui Qin, Shu Jiang, Nan Yang, Yang Lu, Yunwen Xu, and Yan Luo. It is their efforts and contributions that enrich the content of this book. The Chinese Defense Science and Technology Publishing Fund has provided financial support for the first and second editions of the Chinese version, and the authors would also like to express their deep gratitude.