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

一类非线性系统的多故障检测方法研究
引用本文:纪志成,苏晓丹,沈艳霞.一类非线性系统的多故障检测方法研究[J].系统仿真学报,2007,19(9):1979-1982.
作者姓名:纪志成  苏晓丹  沈艳霞
作者单位:江南大学电气自动化研究所,江苏,无锡,214122
基金项目:江苏省高技术研究发展计划项目
摘    要:针对一类非线性系统,提出了一种基于最优滤波的多故障检测观测器设计方法。引入神经网络逼近系统非线性部分,由Lyapunov稳定性理论设计神经网络权值调整规则.采用对等空间方法设计最优滤波矩阵滤除干扰故障噪声,从而对目标故障进行检测和跟踪。仿真结果验证了该方法的正确性和可行性。

关 键 词:多故障检测  神经网络  观测器  稳定性  对等空间法
文章编号:1004-731X(2007)09-1979-04
收稿时间:2006-03-24
修稿时间:2006-11-02

Study on Multiple-fault Detecting Method for Non-linear System
JI Zhi-cheng,SU Xiao-dan,SHEN Yan-xia.Study on Multiple-fault Detecting Method for Non-linear System[J].Journal of System Simulation,2007,19(9):1979-1982.
Authors:JI Zhi-cheng  SU Xiao-dan  SHEN Yan-xia
Institution:Institute of Electrical Automation, Southern Yangtze University, Wuxi 214122, China
Abstract:A novel multiple-faults diagnosis method for non-linear system was proposed. The nonlinear part was approximated by neural network, and the on-line tuning rule of neural network was designed by Lyapunov stability theory. Parity space approach was introduced to design the optimal filter matrix, which used to filter out the nuisance faults, so the target fault could be detected and tracked. Finally, the simulation example shows the validity and the applicability of the proposed method.
Keywords:multiple faults detection  neural network  observer  stability  parity space approach
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