Cubic causality modeling and uncertain inference method for dynamic fault diagnosis
DONG Chunling1, ZHAO Yue2, ZHANG Qin1,2
1. Department of Computer Science and Technology, Tsinghua University, Beijing 100084, China; 2. Institute of Nuclear and New Energy Technology, Tsinghua University, Beijing 100084, China
Abstract：Complex systems need dynamic, real-time, reliable fault diagnostics but current methods have some shortcomings. This paper expands the dynamic uncertain causality graph method (DUCG) for temporal causality modeling and reasoning theory to correct the limits of the DUCG method and other probabilistic graphical models. A Cubic DUCG is developed that is characterized by a true dynamic model of dynamic problems. The cubic causality graph abandons the restriction of the Markov assumption usually used in temporal models with the fault formation, evolution, and development in dynamic systems represented by allowing causal connections to penetrate among any number of time-slices. The negative feedback dynamics is modelled intuitively combined with a reliable dynamic inference algorithm. Fault tests on the secondary loop of Ningde Nuclear Power Plant Unit 1 (CPR1000) simulator show that Cubic DUCG is accurate, efficient, and capable of dealing with the complex dynamics including negative feedback.
[1] HELLER K A, TEH Y W, G R R D, UNIT G. Infinite hierarchical hidden Markov models[J]. Journal of Machine Learning Research, 2009, 5:224-231. [2] PETROPOULOS A, CHATZIS S P, XANTHOPOULOS S. A hidden Markov model with dependence jumps for predictive modeling of multidimensional time-series[J]. Information Sciences, 2017, 412-413:50-66. [3] SERIR L, RAMASSO E, ZERHOUNI N. Time-sliced temporal evidential networks:The case of evidential HMM with application to dynamical system analysis[C]//Proceedings of 2011 IEEE Conference on Prognostics and Health Management (PHM). Montreal, Canada:IEEE, 2011:1-10. [4] SCHOCKAERT S, DE COCK M. Temporal reasoning about fuzzy intervals[J]. Artificial Intelligence, 2008, 172(8-9):1158-1193. [5] SEZER S, ATALAY A E. Dynamic modeling and fuzzy logic control of vibrations of a railway vehicle for different track irregularities[J]. Simulation Modelling Practice and Theory, 2011, 19(9):1873-1894. [6] RUAN S, ZHOU Y K, YU F L, et al. Dynamic multiple-fault diagnosis with imperfect tests[J]. IEEE Transactions on Systems, Man, and Cybernetics, Part A:Systems and Humans, 2009, 39(6):1224-1236. [7] SINGH S, KODALI A, CHOI K, et al. Dynamic multiple fault diagnosis:Mathematical formulations and solution techniques[J]. IEEE Transactions on Systems, Man, and Cybernetics, Part A:Systems and Humans, 2009, 39(1):160-176. [8] ARROYO-FIGUEROA G, SUCAR L E. Temporal Bayesian network of events for diagnosis and prediction in dynamic domains[J]. Applied Intelligence, 2005, 23(2):77-86. [9] GALAN S F, ARROYO-FIGUEROA G, DIEZ F J, SUCAR L E. Comparison of two types of event Bayesian networks:A case study[J]. Applied Artificial Intelligence, 2007, 21(3):185-209. [10] NODELMAN U. Continuous time Bayesian networks[D]. Palo Alto:Stanford University, 2007. [11] WINGATE D, GOODMAN N D, ROY D M, et al. The infinite latent events model[C]//Proceedings of the 25th Conference on Uncertainty in Artificial Intelligence. Montreal, Canada:AUAI Press, 2009:607-614. [12] MURPHY K P. Dynamic Bayesian networks:Representation, inference and learning[D]. Berkeley:University of California, 2002. [13] CAI B P, LIU Y, XIE M. A dynamic-Bayesian-network-based fault diagnosis methodology considering transient and intermittent faults[J]. IEEE Transactions on Automation Science and Engineering, 2017, 14(1):276-285. [14] CODETTA-RAITERI D, PORTINALE L. Dynamic Bayesian networks for fault detection, identification, and recovery in autonomous spacecraft[J]. IEEE Transactions on Systems, Man, and Cybernetics:Systems, 2015, 45(1):13-24. [15] ZHANG Q. Dynamic uncertain causal graph for knowledge representation and reasoning:Discrete DAG cases[J]. Journal of Computer Science and Technology, 2012, 27(1):1-23. [16] ZHANG Q, DONG C L, CUI Y, et al. Dynamic uncertain causality graph for knowledge representation and probabilistic reasoning:Statistics base, matrix, and application[J]. IEEE Transactions on Neural Networks and Learning Systems, 2014, 25(4):645-663. [17] ZHANG Q, GENG S C. Dynamic uncertain causality graph applied to dynamic fault diagnoses of large and complex systems[J]. IEEE Transactions on Reliability, 2015, 64(3):910-927. [18] ZHANG Q, ZHANG Z. Dynamic uncertain causality graph applied to dynamic fault diagnoses and predictions with negative feedbacks[J]. IEEE Transactions on Reliability, 2016, 65(2):1030-1044. [19] DONG C L, ZHAO Y, ZHANG Q. Assessing the influence of an individual event in complex fault spreading network based on dynamic uncertain causality graph[J]. IEEE Transactions on Neural Networks and learning systems, 2016, 27(8):1615-1630. [20] ZHANG Q. Dynamic uncertain causality graph for knowledge representation and reasoning:continuous variable, uncertain evidence, and failure forecast[J]. IEEE Transactions on Systems, Man, and Cybernetics:Systems, 2015, 45(7):990-1003. [21] ZHOU Z X, ZHANG Q. Model event/fault trees with dynamic uncertain causality graph for better probabilistic safety assessment[J]. IEEE Transactions on Reliability, 2017, 66(1):178-188. [22] DONG C L, WANG Y J, ZHANG Q, et al. The methodology of dynamic uncertain causality graph for intelligent diagnosis of vertigo[J]. Computer Methods and Programs in Biomedicine, 2014, 113(1):162-174. [23] 董春玲, 张勤. 用于不确定性故障诊断的权重逻辑推理算法研究[J]. 自动化学报, 2014, 40(12):2766-2781.DONG C L, ZHANG Q. Research on weighted logical inference for uncertain fault diagnosis[J]. Acta Automatica Sinica, 2014, 40(12):2766-2781.(in Chinese) [24] ZHANG Q, YAO Q Y. Dynamic uncertain causality graph for knowledge representation and reasoning:Utilization of statistical data and domain knowledge in complex cases[J]. IEEE Transactions on Neural Networks and Learning Systems, 2017, 99:1-15.
2018, Vol. 58 
