Research on Failure Analysis Model of Complex Mechanical System
-
摘要: 将动态故障概率与面向对象贝叶斯网络(OOBN)相结合,在系统结构功能关系和故障模式影响分析(FMEA)的基础上,建立一种适合于复杂机械系统的动态面向对象贝叶斯网络(DOOBN)故障分析模型,该方法考虑了组件故障概率随时间的变化及组件可靠性退化对系统功能状态的影响,诊断模型具有层次性、动态性,利用该模型可以方便的对复杂系统进行故障传播的分析和关联故障的定位。在ISG-发动机上的应用表明该方法既可以简化系统的故障分析模型,又使模型描述更符合系统的实时状态。Abstract: Combining dynamic fault probability with object oriented Bayesian network (OOBN), we propose a fault diagnosis model based on dynamic object oriented Bayesian network(DOOBN) with both the functional analysis and the failure mode effect analysis(FMEA). This methodology can adapt to construct fault diagnosis model of complex system. Considering the fault probability of component increases with the working time, the system function state is influenced by degradation of component reliability. This model is hierarchical and dynamic, it can be used to analyze failure propagation of, and it also facilitates the isolation of dependent failure. The proposed method was applied to ISG-engineand results suggest that this method can not only simplify the diagnosis model but also describe the actual state of the system.
-
Key words:
- failure analysis /
- failure isolation /
- failure mode /
- object oriented Bayesian network
-
[1] 王远航,邓超,吴军,等.基于混合型专家系统的重型机床故障诊断[J].计算机集成制造系统,2010,16(10):2139-2147 Wang Y H, Deng C, Wu J, et al. Mixed-expert-system-based fault diagnosis for heavy machine[J]. Computer Integrated Manufacturing Systems, 2010,16(10):2139-2147(in Chinese) [2] Leung D, Romagnoli J. Dynamic probabilistic model-based expert system for fault diagnosis[J]. Computers & Chemical Engineering, 2000,24(11):2473-2492 [3] Liu X F, Ma L, Mathew J. Machinery fault diagnosis based on fuzzy measure and fuzzy integral data fusion techniques[J]. Mechanical Systems and Signal Processing, 2009,23(3):690-700 [4] Li X G. Application of Bayesian networking in fault diagnosis of car[J]. Highways & Automotive Applications, 2006,(2):21-24 [5] Bobbioa A, Portinalea L, Minichinob M, et al. Improving the analysis of dependable systems by mapping fault trees into Bayesian networks[J]. Reliability Engineering & System Safety, 2001,71(3):249-260 [6] Lerner U, Parr R, Koller D, et al. Bayesian fault detection and diagnosis in dynamic systems[C]//Proceedings of the 17th National Conference on Artificial Intelligence (AAAI-00). Austin, Texas:AAAI Press, 2000:531-537 [7] Duan R X, Tu J L, Dong D C. A novel hybrid approach of fault tree and bayesian networks for fault diagnosis[J]. Journal of Computational Information Systems, 2010,6(11):3605-3612 [8] Kawahara Y, Fujimaki R, Yairi T, et al. Diagnosis method for spacecraft using dynamic bayesian networks[C]//Proceedings of the 8th International Symposium on Artificial Intelligence, Robotics and Automation in Space-iSAIRAS'. Munich, Germany:European Space Agency, 2005:5-8 [9] Nunnari G, Cannavò F, Vrânceanu R. Bayesian networks approach for a fault detection and isolation:a case of study[M]//Abraham A, de Baets B, Köppen M, et al. Applied Soft Computing Technologies:The Challenge of Complexity. Berlin Heidelberg:Springer, 2005:173-183 [10] Yu D C, Nguyen T C, Haddawy P. Bayesian network model for reliability assessment of power systems[J]. IEEE Transactions on Power Systems, 1999,14(2):426-432 [11] 宋保维.系统可靠性设计与分析[M].西安:西北工业大学出版社,2008 Song B W. System reliability design and analysis[M]. Xi'an:Northwestern Polytechnical University Press, 2008(in Chinese) [12] Bangso O, Wuillemin P H. Top-down construction and repetitive structures representation in Bayesian Networks[C]//Proceedings of the 30th International Florida Artificial Intelligence Research Symposium Conference. Florida, USA:AAAI Press, 2000:282-286 [13] Koller D, Pfeffer A. Object-oriented bayesian networks[C]//Proceedings of the 30th Annual Conference on Uncertainty in Artificial Intelligence (UAI-97). Providence, Rhode Island:AUAI, 1997:302-313 [14] Weber P, Suhner M C. An application of Bayesian networks to the performance analysis of a process[C]//Proceedings of Lm 13, ESREL 2002 European Conference. Lyon, France:ESREL,;002 [15] Yang S H, Zhou Y C, Baras J. Compositional analysis of dynamic bayesian networks and applications to complex dynamic system decomposition[J]. Procedia Computer Science, 2013,16:167-176
点击查看大图
计量
- 文章访问数: 214
- HTML全文浏览量: 34
- PDF下载量: 8
- 被引次数: 0