Fault isolation for multivariate nonlinear non-Gaussian systems using generalized entropy optimization principle

This paper is concerned with the fault isolation (FI) problem for multivariate nonlinear non-Gaussian systems by using a novel filtering method. The generalized entropy optimization principle (GEOP) is established for non-Gaussian systems with multiple faults and disturbances, where the statistic information including entropy and mean of the residual variable is maximized in the presence of the target fault as well as all the nuisance faults and disturbances, and is minimized in the absence of the target fault but in the presence of the nuisance faults and disturbances. Different from the existing results where the output is measurable for feedback, the fault isolation filter is designed and driven by the joint output stochastic distributions rather than its deterministic value. The error dynamics is represented by a multivariate nonlinear non-Gaussian system, for which new recursive relationships are proposed to formulate the joint probability density functions (JPDFs) of the residual variable in terms of the JPDFs of the noises and the faults. Finally, a simulation example is given to demonstrate the effectiveness of the proposed multivariate FI algorithms.

[1]  K. Loparo,et al.  Active probing for information in control systems with quantized state measurements: a minimum entropy approach , 1997, IEEE Trans. Autom. Control..

[2]  Hong Wang,et al.  Fault detection for non-linear non-Gaussian stochastic systems using entropy optimization principle , 2006 .

[3]  P. Frank,et al.  Survey of robust residual generation and evaluation methods in observer-based fault detection systems , 1997 .

[4]  Jie Chen,et al.  Robust Model-Based Fault Diagnosis for Dynamic Systems , 1998, The International Series on Asian Studies in Computer and Information Science.

[5]  Bruno O. Shubert,et al.  Random variables and stochastic processes , 1979 .

[6]  Michèle Basseville,et al.  Fault Detection and Isolation in Nonlinear Dynamic Systems: A Combined Input-Output and Local Approach , 1998, Autom..

[7]  Lei Guo,et al.  Minimum entropy filtering for multivariate stochastic systems with non-Gaussian noises , 2005, Proceedings of the 2005, American Control Conference, 2005..

[8]  Hong Wang,et al.  Minimum entropy control of closed-loop tracking errors for dynamic stochastic systems , 2003, IEEE Trans. Autom. Control..

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

[10]  Deniz Erdogmus,et al.  Generalized information potential criterion for adaptive system training , 2002, IEEE Trans. Neural Networks.

[11]  Tianyou Chai,et al.  Entropy Optimization Filtering for Fault Isolation of Nonlinear Non-Gaussian Stochastic Systems , 2009, IEEE Transactions on Automatic Control.

[12]  Fuwen Yang,et al.  Robust Kalman filtering for discrete time-varying uncertain systems with multiplicative noises , 2002, IEEE Trans. Autom. Control..

[13]  Athanasios Papoulis,et al.  Probability, Random Variables and Stochastic Processes , 1965 .

[14]  Jason L. Speyer,et al.  Optimal stochastic fault detection filter , 2003, Autom..