考虑性能分布信息的多态系统可靠性评估模型

Reliability Assessment Model of Multi-state Systems Considering Random Distribution Information of Performance

针对通用生成函数只能用于求解多态部件离散性能状态的情形,在分析其用于评估多态系统可靠性实质的基础上,定义了多状态系统可用度、输出性能期望和性能失效值等可靠性评价指标参数。考虑部件及系统性能随机变量的分布信息,分别对离散型、连续型以及既有离散又有连续的多性能状态系统的可靠性进行分析建模。从两部件的一般串并联系统推广至多个部件的多态系统上去,扩展了多状态系统可靠性分析的理论及应用范畴。最后通过对文献中案例的对比计算验证了所提方法的科学性与合理性,证明了该方法具有更强的工程实践价值且能处理多种不确定性的信息融合问题。

Universal generating function technology can only be used to solve discrete performance state of multi-state components.After analyzing the essence of universal generating function,reliability parameters including availability,mean output performance and performance deficiency were defined.Considering the random distribution of output performance in multi-state systems,reliability of multi-performance state systems containing discrete random variables,continuous random variables,both discrete and continuous random variables were respectively analyzed and modeled.From the general serial-parallel system containing only two components to multiple parts,both theory and application of multi-state system reliability were expanded.Finally,comparison and calculation of cases in other literatures verify the effectiveness and rationality of the possibility method proposed above,which also proves the method has a stronger engineering practice value and can solve information fusion of multiple uncertainties.