摘要
基于前向模糊神经网络 ANFIS提出了一种新型的动态模糊神经网络 ( DFNN) ,将模糊逻辑、神经网络和 PID控制器三者的优点有机地融合在一起 .通过在 ANFIS的归一化层和输出层之间加入递归层 ,构成了动态模糊神经网络 ( DFNN) ,并推导了基于 BP的反传学习算法 .与 ANFIS和 PID控制器相比 ,DFNN具有更好的控制效果 .DFNN的参数具有明确的物理意义 ,可根据专家的经验选择初值 ,加快了网络的收敛速度 ;由于 DFNN为动态神经网络 。
A novel dynamical neuro fuzzy network (DFNN) is provided based on the forward neuro fuzzy network ANFIS, which combines the advantages of fuzzy system, neural network and PID algorithm. DFNN is constructed by adding a recurrent layer between the normalized layer and output layer of ANFIS, and backpropogation learning algorithm (BP) is given to DFNN. Compared with ANFIS and PID controller, DFNN has better control results. The parameters of DFNN have the implicite meaning and their initial values can be chosen by the experience of expert which increase the converging speed of network. Because of the dynamical ability of DFNN, it has the stronger ability of handling the dynamical system.
出处
《北京理工大学学报》
EI
CAS
CSCD
2000年第5期589-592,共4页
Transactions of Beijing Institute of Technology
基金
部级预研项目
