基于Bayes network含失效节点的网络可靠性评估  

Reliability evaluating of networks with imperfect nodes based on Bayes network

作　　者：李振[1] 孙新利[1] 姬国勋[1] 刘志勇[2] 

机构地区：[1]第二炮兵工程学院103教研室,西安710025 [2]第二炮兵装备研究院二所,北京100085

出　　处：《系统工程理论与实践》2011年第10期1974-1984,共11页Systems Engineering-Theory & Practice

基　　金：第二炮兵工程学院创新性探索研究项目(XY2010JJB23)

摘　　要：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.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 associated computational complexities.The provided example shows correctness and applicability of the algorithm.The method can be applied to reliability evaluating of networks with imperfect nodes and common cause failure.Moreover,based on BN inference,as network failure probability and importance degrees of each component,posterior probability of arbitrary component failure and common cause failure occurrence after network failure that are useful for fault diagnosis can also be obtained.

关 键 词：网络可靠性 失效节点 共因失效 BAYES NETWORK 

分 类 号：TP202.1[自动化与计算机技术—检测技术与自动化装置]