航空学报 > 2001, Vol. 22 Issue (6): 556-558

自组织模糊CMAC神经网络及其非线性系统辨识

王源, 胡寿松, 齐俊伟   

  1. 南京航空航天大学自动化学院 江苏南京 210016
  • 收稿日期:2000-10-08 修回日期:2001-02-25 出版日期:2001-12-25 发布日期:2001-12-25

SELF-ORGANIZING FUZZY CMAC NEURAL NETWORK AND ITS NONLINEAR SYSTEM IDENTIFICATION

WANG Yuan, HU Shou-song, QI Jun-wei   

  1. College of Automation, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China
  • Received:2000-10-08 Revised:2001-02-25 Online:2001-12-25 Published:2001-12-25

摘要:

针对CMAC的特点,提出了联想度的概念,并由此设计了一种自组织模糊 CMAC神经网络( SOFC-MAC)及其学习算法,证明了SOFCMAC能以任意精度对非线性特性一致逼近。该网络具有学习速度快,逼近精度高及局部泛化能力等特点。歼击机系统特征模型辨识仿真验证表明了该方法的有效性。

关键词: CMAC, 模糊神经网络, 系统辨识

Abstract:

A concept of association degree is proposed and further a self-organizing fuzzy CMAC neural network and its learning algorithm are presented based on CMAC. And it is proved that the approximations provided by the SOFCMAC can be made arbitrarily accurate. The proposed network capable of local generalization is characterized by fast learning, accurate approximation, \%etc\%. In this paper, the network is used in fighter identification and satisfactory result is obtained.

Key words: CMAC, fuzzy neur al networ ks, system identification