摘要
针对故障树分析方法在可靠性分析中的局限性,研究了贝叶斯网络在可靠性分析中的应用,给出了故障树向贝叶斯网络转化的方法,以及基于贝叶斯网络求解顶事件发生概率、最小路集、最小割集和底事件重要度的算法.最后对大亚湾核电站紧急停堆失效进行了分析,并与故障树分析的结果进行了比较,结果表明,基于贝叶斯网络的建模分析方法可以得到更丰富的信息.
According to the deficiency of fault tree analysis in reliability analysis, a new method based on Bayesian networks is proposed. This paper is aimed at converting fault trees to Bayesian networks, and solving the probability of top event, minima1 path sets, minima1 cut sets and importance based on the Bayesian networks. The comparison of the two methods is carded out by means of the Daya Bay nuclear power plant reactor protection problem in the end, and the results show that the proposed method could obtain more additional information.
出处
《系统工程理论与实践》
EI
CSCD
北大核心
2006年第6期95-100,共6页
Systems Engineering-Theory & Practice
基金
十五国防预研项目(41319020103)
关键词
贝叶斯网络
故障树分析
最小路集
最小割集
故障诊断
Bayesian networks
fault tree analysis
minimal path sets
minimal cut sets
fault diagnosis
