Identification and predictive control of a multistage evaporator

被引：22

作者：

Atuonwu, J. C. ^{[1
]}

Cao, Y. ^{[1
]}

Rangaiah, G. P. ^{[2
]}

Tade, M. O. ^{[3
]}

机构：

[1] Cranfield Univ, Sch Engn, Bedford, England

[2] Natl Univ Singapore, Dept Chem & Biomol Engn, Singapore 117548, Singapore

[3] Curtin Univ Technol, Dept Chem Engn, Perth, WA, Australia

来源：

CONTROL ENGINEERING PRACTICE | 2010年 / 18卷 / 12期

关键词：

Multiple-effect evaporators; Nonlinear model predictive control; Nonlinear system identification; Recurrent neural networks; Automatic differentiation; RECURRENT NEURAL-NETWORKS; SYSTEM-IDENTIFICATION; BACKPROPAGATION; REACTORS; INVERSE;

D O I：

10.1016/j.conengprac.2010.08.002

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

A recurrent neural network-based nonlinear model predictive control (NMPC) scheme in parallel with PI control loops is developed for a simulation model of an industrial-scale five-stage evaporator. Input-output data from system identification experiments are used in training the network using the Levenberg-Marquardt algorithm with automatic differentiation. The same optimization algorithm is used in predictive control of the plant. The scheme is tested with set-point tracking and disturbance rejection problems on the plant while control performance is compared with that of PI controllers, a simplified mechanistic model-based NMPC developed in previous work and a linear model predictive controller (LMPC). Results show significant improvements in control performance by the new parallel NMPC-PI control scheme. (C) 2010 Elsevier Ltd. All rights reserved.

引用

页码：1418 / 1428

页数：11

共 47 条

[1] Nonlinear system identification for predictive control using continuous time recurrent neural networks and automatic differentiation [J].