Factoring Dynamic Bayesian Networks using Possible Conflicts

##plugins.themes.bootstrap3.article.main##

##plugins.themes.bootstrap3.article.sidebar##

Carlos J. Alonso-Gonzalez Noemi Moya Gautam Biswas

Abstract

Dynamic Bayesian Networks (DBNs) are temporal probabilistic graphical models that represent in a very compact way dynamic systems. They have been used for model based diagnosis of complex systems because they naturally cope with uncertainties in the diagnosis process, particularly sensor uncertainty in noisy environments. A caveat of DBN is the complexity of the inference procedure which is usually performed with Particle Filtering algorithms. Recently, factoring has been proposed to decompose a DBN into subsystems, distributing the diagnosis process and reducing the computational burden. This paper proposes decomposing a system with Possible Conflicts (PCs) and, afterwards, building a DBN factor from each resultant PC. The method can be systematically applied to a state space representation of a dynamic system to obtain minimal observable subsystems with analytical redundancy. Assuming single fault hypothesis and known fault modes, the method allows performing consistency based fault detection, isolation and identification with the unifying formalism of DBN. The three tank system benchmark has been used to illustrate the approach. Two fault scenarios are discussed and a comparison of the behaviors of a DBN of the complete system with the DBN factors is also included.

How to Cite

J. Alonso-Gonzalez, C., Moya, N., & Biswas, G. (2010). Factoring Dynamic Bayesian Networks using Possible Conflicts. Annual Conference of the PHM Society, 2(2). https://doi.org/10.36001/phmconf.2010.v2i1.1941
Abstract 57 | PDF Downloads 12

##plugins.themes.bootstrap3.article.details##

Keywords

PHM

References
( Alonso-Gonzlez et al., 2008 ) C. Alonso-Gonzlez, J. J. Rodrguez, O. Prieto, and B. Pulido. Machine learning and model based diagnosis using possible conflicts and system decomposition. In Proceeding of the 19th International Workshop on Principles of Diagnosis, DX08, pages 215–222, Blue Mountains, Australia, September 2008.

(Arulampalam et al., 2002 ) M. S. Arulampalam, S. Maskell, N. Gordon, and T. Clapp. A tutorial on particle filters for online nonlinear/non-gaussian bayesian tracking. IEEE Transactions on Signal Processing, 50(2):174188, 2002.

(Biswas et al., 2003 ) Gautam Biswas, Gyula Simon, Nagabhushan Mahadevan, Sriram Narasimhan, John Ramirez L, and Gabor Karsai I. A robust method for hybrid diagnosis of complex systems. In 5th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes (SAFEPROCESS, pages 1125–1131, 2003.

( Bregon et al., 2009 ) A. Bregon, B. Pulido, G. Biswas, and X. Koutsoukos. Generating possible conflicts from bond graphs using temporal causal graphs. In Proceeding of the 23rd European Conference on Modelling and Simulation, ECMS09, Madrid, Spain, 2009.

(Dearden and Clancy, 2001 ) R. Dearden and D. Clancy. Particle filters for real-time fault detection in planetary rovers. In the 12th International Workshop on Principles of Diagnosis, pages 1–6, 2001.

(Gelso et al., 2008 ) E. R. Gelso, G. Biswas, S. M. Castillo, and J. Armengol. A comparison of two methods for fault detection: a statistical decision, and an interval-based approach. In Proceeding of the 19th International Workshop on Principles of Diagnosis, DX08, Blue Mountains, Australia, 2008.

(Koller and Lerner, 2001 ) D. Koller and U. Lerner. Sequential Monte Carlos Methods in Practice, chapter Sampling in factored dynamic systems. Springer, 2001.

(Lerner et al., 2000 ) Uri Lerner, Ronald Parr, Daphne Koller, and Gautam Biswas. Bayesian fault detection and diagnosis in dynamic systems. In Prooccedings of the AAAI/IAAI, page 531537, 2000.

( Mosterman and Biswas, 1999 ) P. Mosterman and G. Biswas. Diagnosis of continuous valued systems in transient operating regions. IEEE Transactions on Systems, Man, and Cybernetics, 29(6):554–565, 1999.

(Moya et al., 2010 ) N. Moya, G. Biswas, C.J. AlonsoGonzalez, and X. Koutsoukos. Structural observability. application to decompose a system with possible conflicts. In Submitted to the 21th International Workshop on Principles of Diagnosis, DX10, June 2010.

(Murphy, 2002 ) Kevin Patrick Murphy. Dynamic Bayesian Networks: Representation, Inference and Learning. PhD thesis, University of California, Berkeley, 2002.

(Narasimhan, 2007 ) S. Narasimhan. Automated diagnosis of physical systems. In Proceedings of ICALEPCS07, pages 701–705, 2007.

(Pulido and Alonso-Gonzalez, 2004 ) B. Pulido and C. Alonso-Gonzalez. Possible conflicts: a compilation technique for consistency-based diagnosis. Part B: Cybernetics, IEEE Transactions on Systems, Man, and Cybernetics, 34(5):2192–2206, Oct. 2004.

(Roychoudhury et al., 2008 ) I. Roychoudhury, G. Biswas, and X. Koutsoukos. Comprehensive diagnosis of continuous systems using dynamic bayes nets. In Proceeding of the 19th International Workshop on Principles of Diagnosis, DX08, Blue Mountains, Australia, September 2008.

( Roychoudhury et al., 2009 ) I. Roychoudhury, G. Biswas, and X. Koutsoukos. Factoring dynamic bayesian networks based on structural observability. In In 48th IEEE Conference on Decision and Control (CDC 2009), 2009.

(Roychoudhury, 2009 ) I. Roychoudhury. Distributed Diagnosis of Continuous Systems: Global diagnosis through local analysis. PhD thesis, Graduate School of the Vanderbilt University, August 2009.
Section
Technical Papers

Most read articles by the same author(s)

1 2 > >>