一种新型模糊神经网络函数逼近器  

A New Type Fuzzy Neural Network Function Approximator

作　　者：马恒[1] 贾传荧[1] 

机构地区：[1]大连海事大学航海技术研究所

出　　处：《计算机工程与应用》2004年第33期92-94,共3页Computer Engineering and Applications

摘　　要：论文提出了一种新型模糊小脑模型神经网络(NFCMAC),它采用模糊隶属度函数作为接收域函数,能够获得较常规CMAC连续性强且有解析微分的复杂函数近似,具有计算量少,学习效率高等优点。同时研究了NFCMAC接受域函数的映射方法、隶属度函数及其参数的选取规律和学习算法。仿真结果表明NFCMAC具有良好的泛化能力和逼近精度,具有较高的收敛速度。A new type fuzzified Cerebellar Model Articulation Controller CMAC(NFCMAC)is proposed,in which the fuzzy membership functions are utilized as the receptive field functions.By using NFCMAC,the approximation of complex functions can be obtained which is more continuous than using conventional CMAC.The mapping of receptive field functions,the selection law of membership with its parameters and the learning algorithm are presented in the paper.The simulation results show that NFCMAC has good generalization,proper approximate accuracy and high convergence speed.

关 键 词：函数逼近器 新型模糊小脑模型神经网络 接受域函数 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]