基于离散时间贝叶斯网络的动态故障树分析方法

提出了一种基于离散时间贝叶斯网络的动态故障树分析方法．首先给出优先与门、顺序相关门、备件门、功能相关门等动态逻辑门向离散时间贝叶斯网络的转化方法，在得到动态故障树对应的离散时间贝叶斯网络之后，再利用贝叶斯网络推理算法计算、诊断和预计顶事件概率、重要度等常规分析结果．对数字飞控计算机系统进行的分析表明，该方法能够保证较高的求解精度，其相对误差均保持在0．4%以内，而且易于扩展到多态和非确定性逻辑关系的情形．

Dynamic Fault Tree Analysis Method Based on Discrete-Time Bayesian Networks

A new dynamic fault tree analysis method based on discrete-time Bayesian networks is proposed. The conversion of the dynamic logic gates, such as priority and gate, sequence enforcing gate, spare gate and functional dependency gate to discrete-time Bayesian networks is studied. The top-event probability, importance measures and other diagnosis and prediction results can be obtained subsequently using the classical algorithms based on Bayesian networks. The comparison with Markov chain method is carried out by means of the digital aviation control computer system, and the results show that the proposed method can guarantee high accuracy, and the relative errors are no more than 0.4% . In addition, the proposed method can be easily extended to multi-state systems with non-determinate logic relations.