A new artificial neural network based fuzzy inference system with moving consequents in if-then rules and selected applications

In this paper a new artificial neural network based fuzzy inference system (ANNBFIS) has been described. The novelty of the system consists in the mov ing fuzzy consequent in if-then rules. The location of this fuzzy set is de termined by a linear combination of system inputs. This system also automat ically generates rules from numerical data. The proposed system operates wi th Gaussian membership functions in premise part. Parameter estimation has been made by connection of both gradient and least-squares methods. For ini tialization of unknown parameter values of premises, a preliminary fuzzy c- means clustering method has been employed. For evaluation of the number of if-then rules, the indexes of Xie-Beni and Fukujama-Sugeno have been applie d. The applications to prediction of chaotic time series, pattern recogniti on and system identification are considered in this paper as well. (C) 1999 Published by Elsevier Science B.V. All rights reserved.