文献类型：期刊文献

中文题名：基于观测器的故障诊断技术的鲁棒性

英文题名：Robustness of Observer-based Fault Diagnosis Technology

作者：赵瑾[1];顾幸生[1];申忠宇[2]

机构：[1]华东理工大学信息科学与工程学院,上海200237;[2]南京师范大学电气与自动化工程学院,南京210042

年份：2005

卷号：20

期号：8

起止页码：939

中文期刊名：控制与决策

外文期刊名：Control and Decision

收录：CSTPCD;;EI(收录号：2005389374935);Scopus;北大核心:【北大核心2004】；CSCD:【CSCD2011_2012】；

基金：上海市科技局重大科技攻关项目(04dz11008);江苏省教育厅自然科学研究项目(04KJD510111)

语种：中文

中文关键词：故障诊断技术;未知输入观测器;复合故障估计器;鲁棒性;不确定性

外文关键词：Fault diagnosis technology; Unknown input observers; Compound fault estimator; Robustness;Uncertainness

摘要：基于观测器的动态系统故障诊断技术的可行性和应用性,关键是对系统不确定性因素具有良好的鲁棒性.根据基于观测器的故障诊断系统鲁棒性的特点,强调通过增加观测器自身的鲁棒性,提高线性不确定系统的故障诊断鲁棒性,并对利用未知输入观测器方法增强鲁棒性进行讨论.提出估计复合故障的一种新的复合故障估计器的设计方法,给出其存在条件以及理论证明,并提出相应的算法.
A key for the feasibility and applicability of the techniques of observer-based fault diagnosis for dynamic systems is a satisfactory robustness with respect to system uncertainties. According to the features of studying robustness of observer-based fault diagnosis systems, emphasis is put on robustness improvement of fault diagnosis in linear uncertain systems by increasing robustness of observers, and enhancement of robustness by using unknown input observers. A new design method of compound fault estimator is developed to estimate compound faults. The existence condition of a estimator is given and a related computing method is proposed.