动态故障诊断中的立体因果建模与不确定性推理方法

被引：6
作者：
董春玲 [1 ]
赵越 [2 ]
张勤 [1 ,2 ]
机构：
[1] 清华大学计算机科学与技术系
[2] 清华大学核能与新能源技术研究院
来源：
清华大学学报(自然科学版) | 2018年 / 58卷 / 07期
基金：
中国博士后科学基金;
关键词：
故障诊断; 时序因果建模; 概率推理; 动态不确定性; 动态负反馈;
D O I：
10.16511/j.cnki.qhdxxb.2018.26.029
中图分类号：
TM623 [核电厂（核电站）];
学科分类号：
摘要：
为满足复杂系统的动态、实时和高可靠性的故障诊断需求,克服动态不确定因果图(dynamic uncertain causality graph,DUCG)及其他概率图模型的局限,该文在DUCG理论的基础上扩展其时序因果表达与推理方法,建立了立体DUCG(Cubic DUCG)理论模型。采用动态的手段处理动态问题,以"逐步生长"的立体因果建模取消了时序模型中常见的Markov假设限制,以穿越式因果连接准确地表达动态系统下故障的产生、演变和发展;直观地刻画和处理动态负反馈等复杂故障逻辑因果关系;给出了严谨、高效的动态推理算法。宁德核电站1号机组CPR1000模拟机二回路系统上的故障实验结果表明:Cubic DUCG诊断推理准确、高效,能有效处理负反馈等复杂动态情形。
引用
收藏
页码：614 / 622
页数：9
相关论文
共 24 条
  • [1] COMPARISON OF TWO TYPES OF EVENT BAYESIAN NETWORKS: A CASE STUDY[J] . S. F. Galan,G. Arroyo-Figueroa,F. J. Diez,L. E. Sucar. &nbspApplied Artificial Intelligence . 2007 (3)
  • [2] Temporal bayesian network of events for diagnosis and prediction in dynamic domains
    Arroyo-Figueroa, G
    Sucar, LE
    [J]. APPLIED INTELLIGENCE, 2005, 23 (02) : 77 - 86
  • [3] Time-sliced temporal evidential networks:The case of evidential HMM with application to dynamical system analysis. SERIR L,RAMASSO E,ZERHOUNI N. Proceedings of 2011IEEE Conference on Prognostics and Health Management (PHM) . 2011
  • [4] Continuous time Bayesian networks. NODELMAN U. . 2007
  • [5] The infinite latent events model. WINGATE D,GOODMAN N D,ROY D M.et al. Proceedings of the 25th Conference on Uncertainty in Artificial Intelligence . 2009
  • [6] Dynamic uncertain causality graph for knowledge representation and reasoning:Utilization of statistical data and domain knowledge in complex cases. ZHANG Q,YAO Q Y. IEEE Transactions on Neural Networks and Learning Systems . 2017
  • [7] A dynamic-Bayesian-network-based fault diagnosis methodology considering transient and intermittent faults. CAI B P,LIU Y,XIE M. IEEE Transactions on Automation Science and Engineering . 2017
  • [8] Model event/fault trees with dynamic uncertain causality graph for better probabilistic safety assessment. ZHOU Z X,ZHANG Q. IEEE Transactions on Reliability . 2017
  • [9] Dynamic Bayesian networks:Representation,inference and learning. MURPHY K P. . 2002
  • [10] Dynamic uncertain causality graph applied to dynamic fault diagnoses of large and complex systems. ZHANG Qin,GENG Shichao. IEEE Transactions on Reliability . 2015
123