No abstract

No abstract

No abstract

This paper proposes a real-time supervised learning algorithm for building a neural-network-based fuzzy logic control and decision system (Fuzzy Neurul Network). The Fuzzy Neural Network (FNIV) is a feedforward multi-layered network which integrates the basic elements and functions of traditional fuzzy logic controller into a connectionist structure which has distributed learning abilities. Provided with training data, the proposed learning algorithm can learn the proper network structure and parameters simultaneously in real time. The structure learning decides the proper connection types and the number of hidden units which represent fuzzy logic rules and the number of fuzzy partitions. The parameter learning adjusts the node and link parameters which represent the membership functions. A novel measure, fuzzy similarity measure, which indicates the degree to which two fuzzy sets are equal has been developed. This measure is utilized in the structure learning. The backprojmgation scheme is incorporated into the proposed learning algorithm to perform the parameter learning. The proposed supervised learning algorithm provides an efficient way for constructing an FNN in real time and introduces a novel scheme to combine structure learning and parameter learning in neural networks. Simulation results are presented to illustrate the performance and applicability of the proposed learning algorithm.

No abstract