Adaptive neural predictive control schemes for unknown nonlinear systems

Wei Wu, Jun Xian Chang, Chia Ju Wu

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Adaptive neural predictive control strategies for general nonlinear systems are proposed. The network weight update rule with discrete-time learning procedures which executes the minimal error between the feedforward neural network (INN) model output and plant output is obtained. The one-step-ahead neural predictive control combined with the 'dual' optimization algorithm serves as a rapid, reliable adaptation mechanism and guarantees the stable output regulation of a class of uncertain nonlinear systems. In principle, the off-line training algorithm on neural networks is reduced, and the state/parameter estimation design is obviated. Through closed-loop simulation demonstrations, the proposed control schemes have been successfully applied to two reactor system examples.

Original languageEnglish
Pages (from-to)107-117
Number of pages11
JournalJournal of the Chinese Institute of Chemical Engineers
Volume36
Issue number2
Publication statusPublished - 2005 Mar 1

Fingerprint

Nonlinear systems
Feedforward neural networks
Parameter estimation
Demonstrations
Neural networks

All Science Journal Classification (ASJC) codes

  • Chemistry(all)
  • Chemical Engineering(all)

Cite this

@article{4f6bea7a59424d19941c789f0e0cd68b,
title = "Adaptive neural predictive control schemes for unknown nonlinear systems",
abstract = "Adaptive neural predictive control strategies for general nonlinear systems are proposed. The network weight update rule with discrete-time learning procedures which executes the minimal error between the feedforward neural network (INN) model output and plant output is obtained. The one-step-ahead neural predictive control combined with the 'dual' optimization algorithm serves as a rapid, reliable adaptation mechanism and guarantees the stable output regulation of a class of uncertain nonlinear systems. In principle, the off-line training algorithm on neural networks is reduced, and the state/parameter estimation design is obviated. Through closed-loop simulation demonstrations, the proposed control schemes have been successfully applied to two reactor system examples.",
author = "Wei Wu and Chang, {Jun Xian} and Wu, {Chia Ju}",
year = "2005",
month = "3",
day = "1",
language = "English",
volume = "36",
pages = "107--117",
journal = "Journal of the Taiwan Institute of Chemical Engineers",
issn = "1876-1070",
publisher = "Taiwan Institute of Chemical Engineers",
number = "2",

}

Adaptive neural predictive control schemes for unknown nonlinear systems. / Wu, Wei; Chang, Jun Xian; Wu, Chia Ju.

In: Journal of the Chinese Institute of Chemical Engineers, Vol. 36, No. 2, 01.03.2005, p. 107-117.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Adaptive neural predictive control schemes for unknown nonlinear systems

AU - Wu, Wei

AU - Chang, Jun Xian

AU - Wu, Chia Ju

PY - 2005/3/1

Y1 - 2005/3/1

N2 - Adaptive neural predictive control strategies for general nonlinear systems are proposed. The network weight update rule with discrete-time learning procedures which executes the minimal error between the feedforward neural network (INN) model output and plant output is obtained. The one-step-ahead neural predictive control combined with the 'dual' optimization algorithm serves as a rapid, reliable adaptation mechanism and guarantees the stable output regulation of a class of uncertain nonlinear systems. In principle, the off-line training algorithm on neural networks is reduced, and the state/parameter estimation design is obviated. Through closed-loop simulation demonstrations, the proposed control schemes have been successfully applied to two reactor system examples.

AB - Adaptive neural predictive control strategies for general nonlinear systems are proposed. The network weight update rule with discrete-time learning procedures which executes the minimal error between the feedforward neural network (INN) model output and plant output is obtained. The one-step-ahead neural predictive control combined with the 'dual' optimization algorithm serves as a rapid, reliable adaptation mechanism and guarantees the stable output regulation of a class of uncertain nonlinear systems. In principle, the off-line training algorithm on neural networks is reduced, and the state/parameter estimation design is obviated. Through closed-loop simulation demonstrations, the proposed control schemes have been successfully applied to two reactor system examples.

UR - http://www.scopus.com/inward/record.url?scp=21044445155&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=21044445155&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:21044445155

VL - 36

SP - 107

EP - 117

JO - Journal of the Taiwan Institute of Chemical Engineers

JF - Journal of the Taiwan Institute of Chemical Engineers

SN - 1876-1070

IS - 2

ER -