基于RBF神经网络的传感器非线性故障鲁棒诊断

被引:5
作者
贾明兴
王福利
何大阔
机构
[1] 东北大学信息科学与工程学院
[2] 东北大学信息科学与工程学院 辽宁沈阳 
[3] 辽宁沈阳 
关键词
故障诊断; 传感器; 非线性; 神经网络; 鲁棒性;
D O I
暂无
中图分类号
TP277 [监视、报警、故障诊断系统];
学科分类号
0804 ; 080401 ; 080402 ;
摘要
针对一类非线性系统,传感器非线性故障情形,提出了新的故障诊断方法·该方法采用状态变量扩展技术将传感器故障转化为系统故障进行诊断,RBF神经网络对传感器故障的导函数进行估计,网络权值在线调整,进而实现故障的实时估计·对于系统中存在的不确定性,故障诊断方法应用阈值处理技术,使算法具有一定鲁棒性·对于给出的算法,证明了Lyapunov稳定性·最后,给出了仿真实例,结果验证了该方法的正确性·
引用
收藏
页码:719 / 722
页数:4
相关论文
共 8 条
[1]  
Process fault diagnosis with parameter estimation observers. Isermann R. Proceedings IFAC Digital Computer Application to Process Control[C] . 1984
[2]  
On-line approximation methods for robust fault detection. Vemuri A T,Polycarpou M M. Proceedings 13th World Congress IFAC[C] . 1996
[3]  
Universal approximation using radial basis function networks. Park J,Sandberg T W. Neural Computing and Applications . 1991
[4]  
Fault detection and isolation for a class of nonlinear systems using an adaptive observer. Yang H,Saif M. Proceedings of American Control Conference[C] . 1997
[5]  
On-line actuator fault diagnosis for a class of nonlinear system. Jia M X,Wang F L. Control and Decision . 2004
[6]  
Incipient fault diagnosis of dynamical systems using online approximators. Demetriou M A,Polycarpou M M. IEEE Transactions on Automatic Control . 1998
[7]  
On-line fault detection and isolation of nonlinear system. Chan C W,Cheung K C,Wang Y,et al. Proceedings of the American Control Conference[C] . 1999
[8]  
Sliding mode observers for fault detection and isolation. Edwards C,Spurgeon S K,Patton R J. Automatica . 2000