基于贝叶斯网络的卫星太阳翼驱动机构系统可靠性建模与评估

摘要

Along with the increase of complication in engineering systems, there exist many complex dynamic characteristics within the system failure process, such as sequence dependency, functional dependency and the spares. A Markov-based dynamic fault tree can figure out the modeling of a system with these characteristics;however, it faces with the issue of state space explosion as the number of components in the system increases. This paper combines a Bayesian network with the dynamic fault tree for system reliability analysis and utilizes the inference capability of the Bayesian network for the reliability assessment and inference of fault probability. The analysis ofpracticalexamplesprovestheeffectivenessofthismethodinengineeringsystems.%　　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。