Study And Application Of Fuzzy-Neural Network Control

Fuzzy Neural Network(FNN) is an active branch in the intelligent control. It iscomposed of the neural network and fuzzy logic system organically. And it is good atself-learning and self-tuning. But there are still some problems in it. In this dissertation,the following theories and application of FNN are discussed.1. The large account of compute work of updating weights and long training timeusually discourage the FNN' s on-line application in industry. In this paper, oneoptimization methods are proposed: Only these linking weights corresponding to thecontrol rules that affect the control performance significantly are updated in order toreduce the compute works and speed up the training progress.2.Resource-allocating network is a famous dynamical learning method . It can bejudged whether new rules should be added and what is the number of the added rules.But the rules which were added can't be removed. This paper presents a method. So notonly the parameters of the FNN can be optimized, but also the structure of the neuralnetwork can be adjusted adaptively.