Subspace-based fault detection robust to changes in the noise covariances

作者: Michael Döhler , Laurent Mevel

DOI: 10.1016/J.AUTOMATICA.2013.06.019

关键词: MathematicsFault detection and isolationLinear systemAlgorithmSubspace topologyStatistical hypothesis testingResidualRobustness (computer science)Ambient noise levelSpeech recognitionLTI system theoryControl and Systems EngineeringElectrical and Electronic Engineering

摘要: The detection of changes in the eigenstructure a linear time invariant system by means subspace-based residual function has been proposed previously. While enjoying some success its applicability particular context vibration monitoring, robustness this framework against noise properties not properly addressed yet. In paper, new robust is and statistics covariances shown. complete theory for hypothesis testing fault derived numerical illustration provided.

参考文章(53)
Lucien M Le Cam, Asymptotic methods in statistical theory Springer-Verlag New York, Inc.. ,(1986)
Alireza Esna Ashari, Laurent Mevel, Input-Output Subspace-Based Fault Detection* IFAC Proceedings Volumes. ,vol. 45, pp. 204- 209 ,(2012) , 10.3182/20120829-3-MX-2028.00113
Tony Gustafsson, A 4SID-Based Change-Detection Algorithm IFAC Proceedings Volumes. ,vol. 29, pp. 4604- 4609 ,(1996) , 10.1016/S1474-6670(17)58408-6
Bob Schalkwijk, J. M. Cohen, Mexico City : México Spring Books. ,(1965)
Ron J. Patton, Paul M. Frank, Robert N. Clarke, Fault diagnosis in dynamic systems: theory and application Prentice-Hall, Inc.. ,(1989)
A. Benveniste, M. Basseville, L. Mevel, Convergence rates for eigenstructure identification using subspace methods conference on decision and control. ,vol. 2, pp. 1550- 1554 ,(2000) , 10.1109/CDC.2000.912080
Rolf Isermann, Fault-Diagnosis Systems ,(2006)