摘要
针对传统的事件树分析方法的二态性和独立性假设的局限性,提出了基于贝叶斯网络的事故序列分析方法.该方法不仅可以对二态独立的事件树进行分析,还可以分析存在多态相依事件的事件树,而且大大缩减了图形的规模.另外利用贝叶斯网络还可以得到其他有用的信息,最后通过一个实例说明了该方法的有效性.
According to the limitations of hypothesis of binary-state and statistical independence in event tree analysis, a new method based on Bayesian networks is proposed. The new method can deal with not only binary-state and statistically independent events, but multi-state and statistically dependent events. At the same time, the size of the graph can be kept down and some additional information can be obtained. An example taken from literature verifies the validity of the proposed method.
出处
《哈尔滨工业大学学报》
EI
CAS
CSCD
北大核心
2009年第9期191-193,197,共4页
Journal of Harbin Institute of Technology
基金
国家杰出青年科学基金项目(70825006)
国家自然科学基金项目(70901024)
关键词
贝叶斯网络
事故序列
事件树
故障诊断
bayesian networks
accident sequence
event tree
failure diagnosis
