摘要
提出变结构模糊神经网络控制及其学习算法，并对变结构模糊神经网络学习规律进行研究，变结构模糊神经网络中的模糊化神经网络（ＦＦＮＮ）、模糊推理神经网络（ＥＦＮＮ）和模糊决策神经网络（ＤＦＮＮ）都是结构可变的，可分开进行模糊隶属函数及模糊推理的学习，其学习过程符合人脑由粗到精的认识规律，学习收敛速度比一般模糊神经网络快，具有很好的适应性。
A model of variable structure fuzzy neural network and its variable structure learning algorithm are proposed in this paper. The structure of neural network fuzzifier, fuzzy inference engine and defuzzifier are variable. The fuzzy memberships and neural network fuzzy inference engine may be trained simultaneously. Its learning regulation accords with man's learning methods. Its speed of convergence is faster than that of ordinary fuzzy neural network and it is of wide adaptability.
出处
《华南理工大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
1999年第3期88-93,共6页
Journal of South China University of Technology(Natural Science Edition)
基金
广东省自然科学基金
关键词
神经网络
学习规律
变结构控制
模糊控制
variable structure fuzzy neural network
learning regulation
learning algorithm
