首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于模糊神经网络的动态非线性系统辨识研究
引用本文:胡玉玲,曹建国.基于模糊神经网络的动态非线性系统辨识研究[J].系统仿真学报,2007,19(3):560-562.
作者姓名:胡玉玲  曹建国
作者单位:1. 北京建筑工程学院自动化系,北京,100044
2. 中国科学院声学研究所,北京,100080
基金项目:北京建筑工程学院校科研和教改项目
摘    要:针对静态模糊神经网络对动态系统辨识精度低的特点,在5层静态模糊神经网络基础上进行了优化和改进,形成了可将暂态信息记忆于网络的动态回归层的动态模糊神经网络,来提高对动态系统的辨识能力。同时给出了参数的动态自适应学习算法。通过仿真实验,证明提出的动态模糊神经网络对动态非线性系统的辨识,可以取得较好的辨识精度,较快的网络收敛速度,为动态非线性系统的辨识提供新的思路。

关 键 词:动态模糊神经网络  动态系统  动态自适应学习算法  辨识
文章编号:1004-731X(2007)03-0560-03
收稿时间:2005-11-28
修稿时间:2006-09-25

Research on Identification of Dynamic Nonlinear System Based on Fuzzy Neural Network
HU Yu-ling,CAO Jian-guo.Research on Identification of Dynamic Nonlinear System Based on Fuzzy Neural Network[J].Journal of System Simulation,2007,19(3):560-562.
Authors:HU Yu-ling  CAO Jian-guo
Institution:1.Department of Electrical Engineering and Automation, Beijing Institute of Civil Engineering and Architecture, Beijing 100044, China; 2.Institute of Acoustics Chinese Academy of Sciences, Beijing 100080, China
Abstract:Aiming at improving the identification ability of dynamic system, a new dynamic fuzzy-neural network was proposed based on the traditional static fuzzy neural networks (TFNN). The new fuzzy-neural network adopted a recurrent layer to memory the temporal information. The inference algorithm and dynamic adaptive learning algorithm were deduced to demonstration the system. It is demonstrated that proposed DFNN is better than TFNN used in dynamic nonlinear system identification by simulation.
Keywords:dynamic fuzzy-neural networks  dynamic system  dynamic adaptive learning algorithm  identification
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号