摘要
针对一类含有不确定性的非线性系统,对于执行器故障情形,提出了一种新的鲁棒故障诊断方法.首先设计了自适应观测器结构,并利用最小二乘法给出了故障估计递推算法.对系统中的不确定性,算法中采用域值处理技术以实现鲁棒故障估计.在此基础上,分析了该方法的鲁棒性,可检测性和稳定性.最后,给出了仿真实例,结果证明了该方法的有效性.
A new robust actuator fault diagnosis method for nonlinear systems with modeling uncertainties is proposed. First, an adaptive observer structure is designed. And an iterative algorithm of the fault estimation is given according to least minimum square theory. For modeling uncertainties of the system, dead region value treatment technique is adopted to realize robust fault evaluation in the algorithm. Robustness, sensitivity and stability of the method are analyzed. Finally, a simulation example is used to illustrate the effectiveness of the method.
出处
《自动化学报》
EI
CSCD
北大核心
2004年第4期601-607,共7页
Acta Automatica Sinica
基金
国家教委博士点基金(20011045008)资助~~
关键词
故障诊断
非线性
自适应观测器
执行器
鲁棒性
Actuators
Algorithms
Computer simulation
Iterative methods
Mathematical models
Observability
Robustness (control systems)
Sensitivity analysis
System stability
Theorem proving
