摘要
本文提出一类基于神经网络的模糊控制.首先,它利用具有动态结构的BP网络进行模糊推理,实现模糊控制的最优推理过程.其次,它采用另一具有动态结构的BP网络校正现有的控制规则,实现规则自组织,在控制过程中不断优化控制性能,从而使控制的效果更加理想.
A kind of fuzzy control based on neural network is proposed in this paper. Firstly,back propagation neural network which has dynamic structure is used to do fuzzy inference in order to implement the optimal inference process of fuzzy control. Secondly,it employes another dynamic structured BP neural network to adjust existing control rules. Therefore the self--organization of rules is implemented.The controlperformance is constantly optimized during Control process.So the control result will be better.
出处
《控制理论与应用》
EI
CAS
CSCD
北大核心
1996年第6期738-744,共7页
Control Theory & Applications
基金
国家攀登计划认识科学(神经网络)重大关键项目
广东省自然科学基金
关键词
模糊控制
自组织
神经网络
自动控制
fuzzy control
dynamic structured BP neural network
self-organize