Efficient Neural and Fuzzy Models for the Identification and Control of Nonlinear Systems


AOUICHE Abdelaziz
Department of Electrical Engineering, University Laarbi Tebessi of Tebessa, Algeria LABGET Laboratory, Faculty of Sciences and Technology
SOUDANI Mouhamed Salah
Mining engineering institute, University Laarbi Tebessi of Tebessa
AOUICHE El Moundher
School of Electrical and Electronic Engineering, Hebei University of Technology


The purpose of this study is to identify and control nonlinear dynamical systems under some ambiguity by fuzzy inference systems (FISs) and artificial neural networks (ANNs). Due to the basic ability of FISs and ANNs to approximate unknown functions and to update different inputs and parameters, they are able to control systems which are complicated for linear controllers. The results indicate the FISs and ANNs (Back Propagation Algorithm) used were very efficient with better performance and good durability in modeling and control of nonlinear systems.

February 5, 2024