基于BQEA的多因素影响下系统故障概率变化范围研究

Study on the change range of systemfault probability under the influence of multiple factors based on BQEA

为了研究多因素影响下系统在不同因素变化过程中的故障概率变化范围,提出了一种基于BQEA的分析方法。利用BQEA具有3条基因链且能容纳多个量子的特点,将因素与量子对应,因素变化与量子状态变化对应,在空间故障树框架内确定系统故障概率的变化范围。空间故障树理论提供了因素与元件故障关系的特征函数,从而得到各元件故障概率分布,通过元件组成系统的结构得到系统故障概率分布表达式。该表达式作为BQEA的优化对象从而获得极小值和极大值,即系统故障概率变化范围。结果表明,使用BQEA确定该变化范围是可行性的。所得变化范围与解析法得到的故障概率分布近似,更适合表征多因素条件下的故障概率变化范围,且能降低计算复杂度并提高效率。

To study the change range of system fault probability under the influence of multiple factors,an analysis method based on BQEA is proposed.Since BQEA has three gene chains and can hold many quanta,the change range of system fault probability is determined in the frame of space fault tree by corresponding factors with quanta and corresponding changes of factors with quantum states.The space fault tree theory provides the characteristic function of the relationship between factors and component faults to obtain the probability distribution of each component fault.The expression of the system fault probability distribution is obtained by the structure of the component-system.Take the expression as the optimization object of BQEA,we can get the minimum and maximum value of the expression,that is,the change range of system fault probability.Besides,this study illustrates the methods of chromosome coding,solution space transformation,chromosome update and mutation.The system is composed of five components,and the influence factors are use temperature and use time.According to the existing research,the system structure is obtained,and then the characteristic function of these five elements is used for the analysis of the system fault change range.This method sacrifices the local accuracy of the fault probability distribution but characterizes the change range of fault probability in different ranges,which is almost the same as the known distribution.The results show that it is feasible to use BQEA for determining the change range.The obtained change range is similar to the fault probability distribution obtained by the analytical method.BQEA is more suitable for characterizing the change range of fault probability under multi-factor conditions.While ensuring that it is similar to the analytical results,it can reduce the computational complexity and improve efficiency.