Bayesian networks and mutual information for fault diagnosis of industrial systems

The purpose of this article is to present and evaluate the performance of a new procedure for industrial process diagnosis. This method is based on the use of a bayesian network as a classifier. But, as the classification performances are not very efficient in the space described by all variables of the process, an identification of important variables is made. This feature selection is made by computing the mutual information between each process variable and the class variable. The performances of this method are evaluated on the data of a benchmark problem: the Tennessee Eastman Process. Three kinds of faults are taken into account on this complex process. The objective is to obtain the minimal recognition error rate for these 3 faults. Results are given and compared with results of other authors on the same data.

[1]  John C. Young,et al.  A Practical Approach for Interpreting Multivariate T2 Control Chart Signals , 1997 .

[2]  Arthur K. Kordon,et al.  Fault diagnosis based on Fisher discriminant analysis and support vector machines , 2004, Comput. Chem. Eng..

[3]  Douglas C. Montgomery,et al.  Introduction to Statistical Quality Control , 1986 .

[4]  S. W. Roberts.  Control chart tests based on geometric moving averages , 2000 .

[5]  Nir Friedman,et al.  Bayesian Network Classifiers , 1997, Machine Learning.

[6]  Stephen P. Gurden,et al.  Standardized Q‐statistic for improved sensitivity in the monitoring of residuals in MSPC , 2000 .

[7]  Pedro Larrañaga,et al.  Supervised classification with conditional Gaussian networks: Increasing the structure complexity from naive Bayes , 2006, Int. J. Approx. Reason..

[8]  N. Lawrence Ricker,et al.  Decentralized control of the Tennessee Eastman Challenge Process , 1996 .

[9]  Manabu Kano,et al.  Comparison of multivariate statistical process monitoring methods with applications to the Eastman challenge problem , 2002 .

[10]  B. Bakshi.  Multiscale PCA with application to multivariate statistical process monitoring , 1998 .

[11]  Richard D. Braatz,et al.  Fault Detection and Diagnosis in Industrial Systems , 2001 .

[12]  Sang Joon Kim,et al.  A Mathematical Theory of Communication , 2006 .

[13]  Nola D. Tracy,et al.  Decomposition of T2 for Multivariate Control Chart Interpretation , 1995 .

[14]  Christos Georgakis,et al.  Plant-wide control of the Tennessee Eastman problem , 1995 .

[15]  A. R. Crathorne,et al.  Economic Control of Quality of Manufactured Product. , 1933 .

[16]  George W. Irwin,et al.  Improved principal component monitoring of large-scale processes , 2004 .

[17]  Lakhmi C. Jain,et al.  Introduction to Bayesian Networks , 2008 .

[18]  J. E. Jackson.  Multivariate quality control , 1985 .

[19]  ChangKyoo Yoo,et al.  Statistical monitoring of dynamic processes based on dynamic independent component analysis , 2004 .

[20]  Igor Kononenko,et al.  Semi-Naive Bayesian Classifier , 1991, EWSL.

[21]  E. F. Vogel,et al.  A plant-wide industrial process control problem , 1993 .

[22]  Bhaskar D. Kulkarni,et al.  Knowledge incorporated support vector machines to detect faults in Tennessee Eastman Process , 2005, Comput. Chem. Eng..

[23]  David G. Stork,et al.  Pattern classification, 2nd Edition , 2000 .