[科普中国]-定量过程控制

内容简介

Quantitative Process Control Theory一书介绍针对化工被控对象，如何使用新颖的控制系统设计理论进行分析与设计。书中提出的易于使用的理论不需要设计者选取权函数，但是可以根据定量的工程化的性能指标（如超调等）进行控制器设计与整定。2

书中每一章，均有总结性的内容提要介绍介绍该章主要研究内容，也提供了足够的习题用于提高和测试对于内容的掌握程度。对于绝大多数的理论结果都给出了数学证明，同时书中的例子取材于真实的化工过程场景，包括造纸过程、热带磨机、磁悬浮系统、核反应堆、蒸馏塔/重油分馏器、套管冷却反应堆、导弹、直升机/飞机和麻醉等。

书中的理论成果来自于作者近二十年的理论研究，对于解决单输入单输出（SISO）和多输入多输出（MIMO）控制系统的定量性能指标问题，提供了一种独特且切实可行的设计方案。书中很多材料已经在应用于课堂教学，适合于工程学科高年级本科生、研究生、鲁棒控制入门阶段的研究者，以及想掌握新型控制设计方法的工程师。

作者简介张卫东，男，1967年生。上海交通大学自动化系教授，博士生导师，国家杰出青年科学基金获得者, 德国洪堡学者，上海市优秀学科带头人。现任上海市优化与控制软件工程研究中心主任、上海高校船舶自动化工程研究中心主任和上海交通大学自动化系副主任。长期从事控制理论及应用研究。在学术上提出了定量过程控制的学术思想, 相关成果已总结成英文专著在CRC Press出版; 发表SCI论文90篇, 论文被SCI他引459次，获得国家发明专利授权26项；理论成果在造纸、发电、精密仪器和海洋工程等领域得到成功应用。荣获全国高校霍英东优秀青年教师奖和2项省部级自然科学二等奖等12项奖励, 培养的学生3次荣获上海市优秀博士生学位论文。曾担任以色列国家基金海外评委、国家创新人才计划评审专家和国家科技奖励评审专家。3

本书特色对于诸如超调等工程实际中的性能指标提供了定量解决方案；

对于SISO/MIMO时滞系统提供了解析设计方法；

提出一种无权函数的最优控制设计方法，不需要使用复杂的数学工具；

提供了一种高效快速地设计控制的方案；

对与理论与应用均有所侧重；

为经典的控制理论与后发展起的鲁棒控制理论建立起桥梁；

书中涵盖大量的有效的物理系统案例，提供具有差异化难度的习题。

该书曾荣获2015年上海普通高校优秀教材奖、上海交通大学第十五届优秀教材 。4

前言Since the Industrial Revolution, control systems have played important roles in improving product quality, saving energy, reducing emissions, and relieving the drudgery of routine repetitive manual operations. In the past hundred years, many theories have been proposed for control system design. However, there are three main problems when some of these advanced control theories are applied to industrial systems:

1. These theories depend on empirical methods or trial-and-error methods in choosing weighting functions.

2. Both the design procedures and results are complicated for understanding and using.

3. The controllers cannot be designed or tuned for quantitative engineering performance indices (such as overshoot or stability margin).

In this book, an improved theory called the Quantitative Process Control Theory is introduced to solve these problems. This new theory has three features:

1. When using the theory, the designer is not required to choose a weighting function.

2. The design is suboptimal and analytical. It is easy to understand and use.

3. The controller can be designed or tuned for quantitative engineering performance indices.

These features enable the controller to be designed efficiently and quickly.

Mathematical proofs are provided in this book for almost all results, especially when they contribute to the understanding of the subjects presented. This will, I believe, enhance the educational value of this book. As few concepts as possible are introduced and as few mathematical tools as possible are employed, so as to make the book accessible. Examples are presented at strategic points to help readers understand the subjects discussed. Chapter summaries are included to highlight the main problems and results. At the end of each chapter, exercises are provided to test the reader’s ability to apply the theory he/she has studied. They are an integral part of the book. There is no doubt that a serious attempt to solve these exercises will greatly improve one’s understanding.

The methods developed here are not confined to process control. They areequally applicable to aeronautical, mechanical, and electrical engineering. To stress this point, examples with different backgrounds are adopted. With a few exceptions, these examples are based on real plants, including

· Paper-making machine · Heat exchanger · Hot strip mill · Maglev · Nuclear reactor · Distillation column/Heavy oil fractionator · Jacket-cooled reactor · Missile · Helicopter/Plane · Anesthesia

The book is divided into 14 chapters. Important topics that are coveredinclude

Introduction and review of classical analysis methods (Chapter 2)

Essentials of the robust control theory (Chapter 3)

H∞ and H2 proportional-integral-derivative controllers for stable plants with time delay (Chapters 4 and 5)

Quasi-H∞ and H2 controllers for stable plants with time delay (Chapter 6)

Quasi-H∞ and H2 controllers for integrating plants with time delay (Chapter 7)

Quasi-H∞ and H2 controllers for unstable plants with time delay (Chapter 8)

Complex control strategies, including two degrees-of-freedom control, cascade control, anti-windup control, and feedforward control (Chapter 9)

Analysis of multi-input/multi-output control systems (Chapter 10)

Classical multi-input/multi-output system design, including decentralized control and decoupling control (Chapter 11)

Quasi-H∞ decoupling control for plants with time delay (Chapter 12)

H2 optimal decoupling control for plants with time delay (Chapter 13)

Multivariable H2 optimal control (Chapter 14)

This book is intended for a wide variety of readers. It is appropriate forhigher level undergraduates and graduates in engineering, beginners in the research area of robust control, and engineers who want to learn new design techniques. It is assumed that readers have had an undergraduate course in classical control theory. A prior course on optimal control or process control would be helpful but is not a requirement.

This book has grown out of 15 years of research. The procedure is alwaysmuch harder than anyone anticipates. I received financial support from the National Science Foundation of China, the Alexander von Humboldt Foundation, Germany, and the National Science Fund for Distinguished Young Scholars, China, which enabled me to pursue the research. I am vastly indebted to many people who have helped and inspired me to start, continue, and complete this book.

My first thanks goes to Professor Shengxun Zhang and Professor YouxianSun, Zhejiang University. They brought me into the area of process control. I am grateful for the continuing help and support from Professor Xiaoming Xu, Professor Yugeng Xi, Professor Songjiao Shi, Professor Zuohua Tian, and Professor Xinping Guan at Shanghai Jiaotong University. I am also greatly indebted to Professor F. Allg¨ower and Professor C.A. Floudas, who hosted me at the University of Stuttgart and Princeton University, respectively, as a visiting professor during the writing of this book.

The first six chapters of this book have been classroom tested for severalyears at Shanghai Jiaotong University. Many students have contributed their time to the book. I would like to thank my PhD students F. S. Alc´antara Cano, Danying Gu, Daxiao Wang, and Mingming Ji for particularly helpful suggestions.

The book makes limited use of the material from several books. In particular, I want to express my sincere appreciation to Morari and Zafiriou (1989), Doyle et al. (1992), and Dorf and Bishop (2001).

Family members are a source of special encouragement in a job of thismagnitude, and I send love and thanks to my parents and my son in this regard.

Lastly, I thank my wife, Chen Lin. She read the manuscripts of differentversions and made corrections in her spare time. She gave hundreds of suggestions on editing, grammar, and technical problems. This book would not be the same without her enormous care and patience.