Decentralized fault detection and diagnosis via sparse PCA based decomposition and Maximum Entropy decision fusion

Abstract This paper proposes an approach for decentralized fault detection and diagnosis in process monitoring sensor networks. The sensor network is decomposed into multiple, potentially overlapping, blocks using the Sparse Principal Component Analysis algorithm. Local predictions are generated at each block using Support Vector Machine classifiers. The local predictions are then fused via a Maximum Entropy algorithm. Empirical studies on the benchmark Tennessee Eastman Process data demonstrated that the proposed decentralized approach achieves accuracy comparable to that of the fully centralized approach, while offering benefits in terms of fault tolerance, reusability, and scalability.

[1]  Slobodan Vucetic,et al.  A boosting method for process fault detection with detection delay reduction and label denoising , 2011, KDD4Service '11.

[2]  Adam L. Berger,et al.  A Maximum Entropy Approach to Natural Language Processing , 1996, CL.

[3]  Christos Georgakis,et al.  Disturbance detection and isolation by dynamic principal component analysis , 1995 .

[4]  Ruxu Du,et al.  Fault diagnosis using support vector machine with an application in sheet metal stamping operations , 2004 .

[5]  Kevin M. Passino,et al.  Decentralized adaptive control of nonlinear systems using radial basis neural networks , 1999, IEEE Trans. Autom. Control..

[6]  Steven X. Ding,et al.  Decentralised fault detection of large-scale systems with limited network communications [Brief Paper] , 2010 .

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

[8]  Michael J. Piovoso,et al.  On unifying multiblock analysis with application to decentralized process monitoring , 2001 .

[9]  Heng Tao Shen,et al.  Principal Component Analysis , 2009, Encyclopedia of Biometrics.

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

[11]  Belur V. Dasarathy.  Decision fusion strategies in multisensor environments , 1991, IEEE Trans. Syst. Man Cybern..

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

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

[14]  B. M. Wise,et al.  UPSET AND SENSOR FAILURE DETECTION IN MULTIVARIATE PROCESSES , 1989 .

[15]  P.K. Varshney,et al.  Distributed fault detection via particle filtering and decision fusion , 2005, 2005 7th International Conference on Information Fusion.

[16]  A. D'Costa,et al.  Data versus decision fusion for distributed classification in sensor networks , 2003, IEEE Military Communications Conference, 2003. MILCOM 2003..

[17]  Rajagopalan Srinivasan,et al.  Multi-agent based collaborative fault detection and identification in chemical processes , 2010, Eng. Appl. Artif. Intell..

[18]  Theodora Kourti,et al.  Statistical Process Control of Multivariate Processes , 1994 .

[19]  Junghui Chen,et al.  Dynamic process fault monitoring based on neural network and PCA , 2002 .

[20]  Raghunathan Rengaswamy,et al.  A review of process fault detection and diagnosis: Part I: Quantitative model-based methods , 2003, Comput. Chem. Eng..

[21]  Ali Cinar,et al.  Multivariate statistical methods for monitoring continuous processes: assessment of discrimination power of disturbance models and diagnosis of multiple disturbances , 1995 .

[22]  Seungjin Choi,et al.  Independent Component Analysis , 2009, Handbook of Natural Computing.

[23]  Jaideep Srivastava,et al.  Unsupervised Learning Based Distributed Detection of Global Anomalies , 2010, Int. J. Inf. Technol. Decis. Mak..

[24]  Qi Cheng,et al.  Distributed Fault Detection with Correlated Decision Fusion , 2009, IEEE Transactions on Aerospace and Electronic Systems.

[25]  Hong Zhou,et al.  Decentralized Fault Diagnosis of Large-Scale Processes Using Multiblock Kernel Partial Least Squares , 2010, IEEE Transactions on Industrial Informatics.

[26]  R. Tibshirani,et al.  Sparse Principal Component Analysis , 2006 .

[27]  Age K. Smilde,et al.  Multiway multiblock component and covariates regression models , 2000 .

[28]  Raghunathan Rengaswamy,et al.  A review of process fault detection and diagnosis: Part III: Process history based methods , 2003, Comput. Chem. Eng..

[29]  Hosein Marzi,et al.  Real-time fault detection and isolation in industrial machines using learning vector quantization , 2004 .

[30]  Rameswar Debnath,et al.  A decision based one-against-one method for multi-class support vector machine , 2004, Pattern Analysis and Applications.

[31]  S. M. Magrabi,et al.  Decentralised fault detection and diagnosis in navigation systems for unmanned aerial vehicles , 2000, IEEE 2000. Position Location and Navigation Symposium (Cat. No.00CH37062).

[32]  Yi Liu,et al.  One-against-all multi-class SVM classification using reliability measures , 2005, Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005..

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

[34]  Marios M. Polycarpou,et al.  Decentralized fault detection in a class of large-scale nonlinear uncertain systems , 2009, Proceedings of the 48h IEEE Conference on Decision and Control (CDC) held jointly with 2009 28th Chinese Control Conference.

[35]  Tao Han,et al.  ART–KOHONEN neural network for fault diagnosis of rotating machinery , 2004 .

[36]  John Platt,et al.  Probabilistic Outputs for Support vector Machines and Comparisons to Regularized Likelihood Methods , 1999 .

[37]  Slobodan Vucetic,et al.  Decentralized Estimation Using Learning Vector Quantization , 2009, 2009 Data Compression Conference.

[38]  Kuang-Ching Wang,et al.  Value-Fusion versus Decision-Fusion for Fault-tolerance in Collaborative Target Detection in Sensor Networks , 2001 .

[39]  Michael I. Jordan,et al.  A Direct Formulation for Sparse Pca Using Semidefinite Programming , 2004, SIAM Rev..

[40]  Venugopal V. Veeravalli,et al.  Decentralized detection in sensor networks , 2003, IEEE Trans. Signal Process..

[41]  Richard O. Duda,et al.  Pattern classification and scene analysis , 1974, A Wiley-Interscience publication.

[42]  Chih-Jen Lin,et al.  LIBSVM: A library for support vector machines , 2011, TIST.

[43]  Zhi-Quan Luo,et al.  Universal decentralized detection in a bandwidth-constrained sensor network , 2004, IEEE Transactions on Signal Processing.

[44]  Marios M. Polycarpou,et al.  Distributed Fault Diagnosis With Overlapping Decompositions: An Adaptive Approximation Approach , 2009, IEEE Transactions on Automatic Control.

[45]  Adwait Ratnaparkhi,et al.  A Maximum Entropy Model for Part-Of-Speech Tagging , 1996, EMNLP.

[46]  S.J. Qin,et al.  Multiblock principal component analysis based on a combined index for semiconductor fault detection and diagnosis , 2006, IEEE Transactions on Semiconductor Manufacturing.

[47]  M. Alanyali,et al.  Distributed Detection in Sensor Networks With Packet Losses and Finite Capacity Links , 2006, IEEE Transactions on Signal Processing.

[48]  Zhi-Quan Luo,et al.  Universal decentralized detection in a bandwidth-constrained sensor network , 2005, IEEE Trans. Signal Process..

[49]  Yew Seng Ng,et al.  Evaluation of decision fusion strategies for effective collaboration among heterogeneous fault diagnostic methods , 2011, Comput. Chem. Eng..