论文部分内容阅读
随着系统结构与功能复杂性的不断增加,系统的失效过程呈现复杂的动态特性,如顺序相关性、功能相关性以及冗余备份。基于马尔科夫过程的动态故障树分析方法能够较好地解决复杂系统的建模问题,然而在求解时该方法却是在全局状态空间中考虑部件不同状态之间的转移,随着底事件以及逻辑门数量的增加,马尔可夫模型的计算量将呈指数增长。为了解决该问题,笔者采用贝叶斯网络结合动态故障树的方法进行分析,利用贝叶斯网络的双向推理能力进行可靠性评估。实例分析表明,该方法能够较好地解决具有动态特性的系统可靠性分析问题,而且具有较高的求解精度。
With the increasing complexity of the system structure and function, the failure process of the system presents complex dynamic characteristics such as order dependency, function correlation and redundant backup. The method of dynamic fault tree analysis based on Markov process can solve the modeling problem of complex systems well. However, in the solution, this method considers the transition between different states of components in the global state space. With the bottom event and As the number of logic gates increases, the computation of Markov models will increase exponentially. In order to solve this problem, the author uses the Bayesian network combined with the dynamic fault tree approach to analyze the reliability of the Bayesian network using two-way reasoning ability. The case study shows that this method can solve the problem of system reliability analysis with dynamic characteristics better and has higher solution accuracy.