Adaptive RBF model for model-based control

Author

Yu, D.L. ; Yu, D.W. ; Gomm, J.B. ; Page, G.F.

Author_Institution

Sch. of Eng., Liverpool John Moores Univ., UK

Volume

1

fYear

2004

fDate

15-19 June 2004

Firstpage

78

Abstract

An adaptive radial basis function (RBF) neural network model is developed for nonlinear systems dynamics using the recursive orthogonal least squares (ROLS) algorithm. The model is oriented to on-line control. A center bank is formed and its associated R matrix is updated on-line. A pruning method is used to select significant centers that used for prediction. The developed adaptive model is evaluated in real data modeling of a multivariable reactor rig and compared with a non-adaptive RBF model.

Keywords

adaptive control; least squares approximations; matrix algebra; multivariable control systems; neurocontrollers; nonlinear control systems; nonlinear dynamical systems; radial basis function networks; recursive estimation; R matrix; RBF neural network model; adaptive radial basis function model; model based control; multivariable reactor rig; nonlinear dynamical systems; online control; pruning method; recursive orthogonal least square algorithm; Adaptive control; Adaptive systems; Control systems; Inductors; Least squares methods; Neural networks; Nonlinear systems; Predictive models; Programmable control; Radial basis function networks;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Control and Automation, 2004. WCICA 2004. Fifth World Congress on

Print_ISBN

0-7803-8273-0

Type

conf

DOI

10.1109/WCICA.2004.1340489

Filename

1340489