A new class of fuzzy neural networks with application in system modeling

Abstract

This paper presents a new class of fuzzy neural networks and its application to system modeling. The optimal set of fuzzy associative memory rules and weights can be identified for a given set of sample data. Test results of modeling a nonlinear system show that the proposed fuzzy neural network can accurately model a complicated nonlinear system and the model is quite robust in the sense that its modeling accuracy is much insensitive to sample data.<>