基于Bayes network含失效节点的网络可靠性评估

针对Bayes network (BN)能很好地表示变量的不确定性和相关性, 并能直接进行不确定性推理的优势, 提出了基于BN含失效节点的网络可靠性评估方法. 建模方面, 给出了通过节点分割去环构建有向无环图及条件概率表的具体方法; 推理方面, 通过节点消隐, 减小BN规模, 有效降低推理复杂度. 分析了算法复杂度, 并通过算例证明了算法的有效性和适用性. 由于BN建模及推理的灵活性, 该方法可研究同时含节点失效、共因失效的网络可靠性, 况且通过BN推理除得到网络失效概率和部件重要度外, 还可得到网络失效条件下任一部件失效或者共因失效发生的概率, 为故障诊断和维护提供指导.

Reliability evaluating of networks with imperfect nodes based on Bayes network

Since Bayes network(BN) can faultlessly represent randomness and dependence of variables, and can make probabilistic inference directly,a method for reliability evaluating of networks with imperfect nodes based on BN was presented in this paper.Procedure for building directed acyclic graph by node partition and conditional probability table were proposed.Furthermore,the method reduced BN inference complexity through node blanking.The proposed method is efficient and effective,followed by the analysis of ass...