This edition first published 2017
© 2017 John Wiley & Sons Ltd
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 Sandip Kumar Lahiri to be identified as the author 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 Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK
Editorial Office
111 River Street, Hoboken, NJ 07030, USA
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
In view of ongoing research, equipment modifications, changes in governmental regulations, and the constant flow of information relating to the use of experimental reagents, equipment, and devices, the reader is urged to review and evaluate the information provided in the package insert or instructions for each chemical, piece of equipment, reagent, or device for, among other things, any changes in the instructions or indication of usage and for added warnings and precautions. 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: Lahiri, Sandip Kumar, 1970– author.
Title: Multivariable predictive control : applications in industry / Sandip Kumar Lahiri, Supra International Private Ltd, Vadodara, India.
Description: First edition. | Hoboken, NJ, USA : Wiley, 2017. | Includes bibliographical references and index. | Description based on print version record and CIP data provided by publisher; resource not viewed.
Identifiers: LCCN 2017010553 (print) | LCCN 2017012540 (ebook) | ISBN 9781119243519 (pdf) | ISBN 9781119243595 (epub) | ISBN 9781119243601 (cloth)
Subjects: LCSH: Predictive control. | Multivariate analysis.
Classification: LCC TJ217.6 (ebook) | LCC TJ217.6 .L34 2017 (print) | DDC 629.8–dc23
LC record available at https://lccn.loc.gov/2017010553
Cover design by Wiley
Front Cover Image: FotoBug11/Shutterstock
Back Cover Image: Paulo Vilela/Shutterstock
To my parents, wife Jinia and two lovely children Suchetona and Srijon
Figure 1.1 | Flow scheme of a simple distillation column using multivariable model predictive controller |
Figure 1.2 | Hierarchy of plant‐wide control framework |
Figure 1.3 | Expected cost vs. benefits for different levels of controls |
Figure 1.4 | Typical benefit of MPC |
Figure 1.5 | MPC stabilization effect can increase plant capacity closer to its maximum limit |
Figure 1.6 | Reduced variability allows operation closer to constraints by shifting set point |
Figure 1.7 | Operating zone limited by multiple constraints |
Figure 1.8 | Opportunity loss due to operator action |
Figure 1.9 | Advance control implementations by one of the major MPC vendors |
Figure 1.10 | Spread of MPC application across the whole spectrum of chemical process industries |
Figure 2.1 | Optimum operating point vs. operator comfort zone |
Figure 2.2 | Different module of MPC |
Figure 2.3 | A general MPC calculation |
Figure 2.4 | Schematic of distillation column |
Figure 2.5 | Model of distillation column |
Figure 2.6 | CV prediction due to past MV change |
Figure 2.7 | Model reconciliation and bias update |
Figure 2.8 | Operating region of a distillation column with two manipulated variables and six controlled variables |
Figure 2.9 | Revised CV trajectory and steady state error |
Figure 2.10 | Develop a detail plan of MV movement to drive the steady state error to zero |
Figure 2.11 | Controlled variables predictions with and without control moves |
Figure 2.12 | Manipulated variables move plan for distillation column |
Figure 3.1 | Brief history of development of MPC technology |
Figure 4.1 | Different steps in MPC implementation project |
Figure 4.2 | Schematics of steps involved in MPC project with vendor |
Figure 5.1 | Benefit estimation procedure |
Figure 5.2 | Stabilizing effect of MPC and moving of set point closer to limit |
Figure 6.1 | Various probable reasons of failure of control loops |
Figure 6.2 | Valve sizing problem detection by process gain |
Figure 6.3 | Typical trends when valve stiction presents |
Figure 6.4 | Typical trends of the process having hysteresis and backlash |
Figure 7.1 | Different steps in functional design |
Figure 8.1 | Expectation matrix (√ definite response expected, X no response expected, ? response is doubtful) |
Figure 8.2 | Basic concept of step test |
Figure 9.1 | Advantages and disadvantages of various model structures |
Figure 9.2 | Flowchart of identification process |
Figure 9.3 | System identification structure |
Figure 10.1 | Types of soft sensors |
Figure 10.2 | Steps involved in developing reliable soft sensors |
Figure 10.3 | Artificial neural network architecture |
Figure 10.4 | Schematic of SVR using an e‐insensitive loss function |
Figure 11.1 | Different tuning parameters |
Figure 11.2 | Hard and soft limits |
Figure 12.1 | The schematic of interface of MPC controller and DCS |
Figure 12.2 | Schematic of online commissioning of the controller |
Figure 13.1 | Effect of move suppression (or MV weight) on CV and MV trajectory |
Figure 13.2 | Effect of CV give up on CV trajectory and CV error |
Figure 14.1 | Benefit loss over time |
Figure 14.2 | Contributing failure factors of postimplementation of MPC applications |
Figure 14.3 | Strategies for avoiding MPC failures |
Figure 16.1 | Major linear MPC companies and their products |
Figure 16.2 | Basic structure of MPC software |
Figure 16.3 | Comparison of different MPC identification technology |
Figure 16.4 | DMCplus product package |
Figure 16.5 | MPC project outline: Conventional vs. adaptive approach |
Figure 16.6 | Optimization in adaptive control mode |
Figure 16.7 | RMPCT product package |
Figure 16.8 | History of SMOC |
Figure 16.9 | SMOC product package |
Table 1.1 | Typical Payback Period of MPC |
Table 1.2 | Typical Benefits of MPC Implementation in CPI |
Table 1.3 | Typical Benefits of MPC implementation in Refinery |
Table 2.1 | Description of CV, MV, and DV in a Simple Distillation Column Shown in Figure 2.4 |
Table 5.1 | Typical Value of Factor β |
Table 5.2 | Average Value and Standard Deviation of Quality Parameters |
Table 6.1 | Typical Performance of Control Loops in Industry |
Table 6.2 | Ziegler‐Nichols Tuning Parameters |
Table 6.3 | Recommended PID Tuning Parameters |
Table 6.4 | IMC Tuning Parameters |
Table 8.1 | Difference between Normal Step Testing and PRBS Testing |
Table 11.1 | Simulation Initial Condition File for MVs |
Table 11.2 | Simulation Initial Condition File for CVs |
Table 11.3 | Controlled Variables with Their Limits for Simulation Studies |
Table 11.4 | Controlled Variables with Their Limits for Simulation Studies |
Table 14.1 | Retaining Initial MPC Benefits after 12 Months |
Table 14.2 | Commercial MPC Monitoring Tools |