Fault Detection and Isolation Based on the Constrained GLR Test

Abstract The problem of fault diagnosis (detection and isolation) in safety-critical manufacturing systems is addressed in the paper. Such an application necessitates to carefully define the notion of “failure” for a safety-critical manufacturing system and to design a fault detection and isolation algorithm by explicitly taking into account this definition. To integrate the information on the system failures in decision-making algorithm, it is proposed to use the constrained generalized likelihood ratio test. It is shown that due to a more accurate definition of the failure the statistical properties of such a diagnosis algorithm can be improved.

[1]  Louis L. Scharf,et al.  Matched subspace detectors , 1994, IEEE Trans. Signal Process..

[2]  I. Nikiforov,et al.  Advanced RAIM algorithms for safe navigation based on the constrained GLR test , 2007, 2007 American Control Conference.

[3]  Igor Nikiforov,et al.  Advanced RAIM Algorithms: First Results , 2005 .

[4]  G. Lorden.  PROCEDURES FOR REACTING TO A CHANGE IN DISTRIBUTION , 1971 .

[5]  I. V. Nikiforov,et al.  A lower bound for the detection/isolation delay in a class of sequential tests , 2003, IEEE Trans. Inf. Theory.

[6]  Michèle Basseville,et al.  FAULT ISOLATION FOR DIAGNOSIS: NUISANCE REJECTION AND MULTIPLE HYPOTHESES TESTING , 2002 .

[7]  Mitra Fouladirad,et al.  Optimal fault detection with nuisance parameters and a general covariance matrix , 2008 .

[8]  I. Nikiforov,et al.  Handling nuisance parameters in systems monitoring , 2005, Proceedings of the 44th IEEE Conference on Decision and Control.

[9]  Igor V. Nikiforov.  A simple recursive algorithm for diagnosis of abrupt changes in random signals , 2000, IEEE Trans. Inf. Theory.

[10]  Florent Retraint,et al.  $\varepsilon$ -Optimal Non-Bayesian Anomaly Detection for Parametric Tomography , 2008, IEEE Transactions on Image Processing.

[11]  Jason L. Speyer,et al.  A generalized Shiryayev sequential probability ratio test for change detection and isolation , 1999, IEEE Trans. Autom. Control..

[12]  I. Nikiforov.  On two new criteria of optimality for the problem of sequential change diagnosis , 1995, Proceedings of 1995 American Control Conference - ACC'95.

[13]  A. Wald.  Tests of statistical hypotheses concerning several parameters when the number of observations is large , 1943 .