引用本文:胡寿松，周 川，胡维礼，陈庆伟，苏红云.基于RBF神经网络观测器的非线性系统鲁棒故障检测方法[J].控制理论与应用,1999,16(6):853~857.[点击复制]
Hu Shousong，Zhou Chuan，Hu Weili ， Chen Qingwei，Su Hongyun.An Approach to Robust Fault Detection for NonlinearSystem Based on RBF Neural Network Observer*[J].Control Theory and Technology,1999,16(6):853~857.[点击复制]
基于RBF神经网络观测器的非线性系统鲁棒故障检测方法
An Approach to Robust Fault Detection for NonlinearSystem Based on RBF Neural Network Observer*
摘要点击 954  全文点击 385  投稿时间：1998-08-25  修订日期：1999-07-25
查看全文  查看/发表评论  下载PDF阅读器
DOI编号  
  1999,16(6):853-857
中文关键词  故障检测  神经网络  观测器  鲁棒性
英文关键词  fault detection  neural network  observer  robustness
基金项目  
作者单位
胡寿松，周 川，胡维礼，陈庆伟，苏红云  
中文摘要
      针对一类仿射非线性动态系统，提出一种基于神经网络非线性观测器的鲁棒故障检测与隔离的新方法．采用RBF神经网络逼近观测器系统中的非线性项，提高了状态估计的精度，证明了状态估计误差稳定且渐近收敛到零；同时提出了一种新的网络权值调整指标方法，提高了神经网络故障分类器的泛化能力，从而保证该方法对被监测系统的建模误差和外部扰动具有良好的鲁棒性．
英文摘要
      A new robust fault detection and isolation (FDI) method based on neural network observer is presented for a class of affine nonlinear dynamic system. A radial basis function neural network is used to approximate the nonlinear item of the monitored system to improve the accuracy of state estimation, and the state estimation error is proved to be zero asymptotically. On the other hand, a new index of weight tuning is adopted to improve the robustness of neural network fault classifier for the modelling error and di.sturbance.