基于SFT理论的系统可靠性评估方法改造研究  

Reliability Assessment Method Based on Space Fault Tree

作　　者：崔铁军[1,2,3] 马云东[3] 

机构地区：[1]辽宁工程技术大学安全科学与工程学院,辽宁阜新123000 [2]矿山热动力灾害与防治教育部重点实验室,辽宁阜新123000 [3]大连交通大学辽宁省隧道与地下结构工程技术研究中心,辽宁大连116028

出　　处：《模糊系统与数学》2015年第5期173-182,共10页Fuzzy Systems and Mathematics

基　　金：国家自然科学基金资助项目(61350003);教育部高校博士学科点专项科研基金资助项目(20102121110002)

摘　　要：为了确定基本事件发生概率随工作条件变化而变化的一类系统可靠性,利用空间故障树修改T-S模糊故障树和贝叶斯网络系统可靠性评估方法,解决了原有方法的系统状态计算结果的存在性问题;系统状态的模糊化问题;基本事件发生概率确定问题。并列举了这类系统的一个实例,即电气系统T,其中的元件故障发生概率与其使用时间t和工作温度c有关。提出了解决上述问题的修改后方法,并计算系统T的故障发生概率分布。就本例研究表明:系统三种状态(0,0.5,1)的概率最值位置接近;P(T=1)的有效范围>P(T=0.5)的有效范围>P(T=0)的有效范围;P(T=0)的最大值<P(T=0.5)的最大值,与实际的系统可靠性特点相同。另一方面,T=0过渡到T=0.5时系统对使用时间t敏感;T=0.5过渡到T=1对使用温度c敏感。In order to determine a class of system reliability with the basic events, the probability of them changes with working conditions, reliability assessment method based On T-S Fuzzy Fault Tree and Bayesian Network is modified by space fault tree. with the method, the problems in the original method are solved such as the existence of the system state calculation results; The fuzzy of the system state; Basic event probability determination. Citing an example of this kind of system, the electrical system T, the component failure probability is related to its use time t and working temperature c. the modified method is put forward to solve these problems, and the failure probability distribution of system T is calculated. For the results of this case study showed that the most value location of the probability of the system three states （0,0.5,1） are close to each other; the effective range of P（T=I） ＆ the effective range of P（T= 0. 5） 〉 the effective range of P（T=0） ; A maximum of P（T=0） 〈 the maximum value of P（T=0.5）, the same as the actual situation of the system reliability. On the other hand, the system is sensitive to use time t with the transition from T= 0 to T= 0. 5 and is sensitive to temperature c with the transition from T=0.5 to T=1.

关 键 词：系统可靠性 T-S模糊故障树 贝叶斯网络 空间故障树 

分 类 号：X913[环境科学与工程—安全科学]