摘要
提出了一种基于神经网络的自组织模糊系统，它能够根据输入输出数据灵活地划分模糊集合．由于采用模糊聚类方法和梯度下降法分两步对该系统进行训练，其收敛速度要比传统的ＢＰ算法快得多．仿真结果表明该系统结构简单，学习速度快，规则数少。
A selforganizing neuralnetworkbased fuzzy system is proposed in this paper.It can partition the input spaces in a flexible way based on the distribution of the training data.By combining both the nearest neighborhood clustering scheme and the gradient descent method,the learning speed converges much faster than the original backpropagation algorithm.Simulation results suggest that the SONNFS has merits of simple structure,fast learning speed,fewer fuzzy logic rules and relatively high modeling accuracy.
出处
《控制理论与应用》
EI
CAS
CSCD
北大核心
1999年第3期455-457,共3页
Control Theory & Applications
关键词
梯度下降法
神经网络
自组织模糊系统
fuzzy logic
neural network
nearest neighborhood clustering scheme
gradient descent method
backpropagation learning scheme
