基于遗传算法的模糊神经网络控制器设计及其稳定性分析

Design of a Kind of Fuzzy Neural Network Controllers Based on Genetic Algorithm and Analysis of Its Stability

首先根据联结主义思想模糊控制器设计问题转化为对模糊神经网络参数的设计和优化，然后通过遗传算法对模糊神经网络的参数进行集中优化，得到了被控对象的一个最优或次优的控制器———模糊神经控制器，稳定性分析为此设计提供了理论依据．仿真实例证实了其性能较常规模糊控制器优越．

Based on the thought of connection,firstly the problem of design of fuzzy controller is changed into that of design and optimization of fuzzy neural networks in this paper.Secondly,an optimal or suboptimal fuzzy controller of the plant is obtained by comprehensivelly optimizing the parameters of fuzzy neural networks using genetic algorithms.Analysis of stability provides theoretical support of the design.Finally,a simulation result verifies that it has a better performance than normal fuzzy controller.