融合专家经验和寿命试验数据的产品可靠性评估  

Reliability evaluation for products by fusing expert knowledge and lifetime data

作　　者：贾祥 郭波[1] JIA Xiang;GUO Bo(College of Systems Engineering,National University of Defense Technology,Changsha 410073,China)

机构地区：[1]国防科技大学系统工程学院,长沙410073

出　　处：《控制与决策》2022年第10期2600-2608,共9页Control and Decision

基　　金：国家自然科学基金项目(71801219);湖南省自然科学基金项目(2019JJ50730)。

摘　　要：专家经验是可靠性工程中常见的一类可靠性数据,通过将其与产品的寿命试验数据融合,可以扩充可靠性信息,为产品可靠性的评估提供新的思路.对此,利用Bayes理论,考虑不同类型和不同形式的专家经验,通过验前矩拟合的方法将其转化为产品寿命分布参数的验前分布.进一步,根据寿命试验数据确定似然函数,推断分布参数的验后分布,可求得数据融合后产品的可靠度和剩余寿命等可靠性评估结果.通过蓄电池算例分析,表明所提出方法的应用及其有效性.The expert knowledge is common in reliability engineering. It would provide a new insight on the estimation of reliability by incorporating the expert knowledge and lifetime data. This problem is studied in this paper through Bayesian theory considering different categories and forms of expert knowledge. First, the prior distribution of distribution parameters are elicited from expert knowledge by fitting prior moments. Further, the posterior distribution of distribution parameters are obtained by integrating the likelihood function from lifetime data and prior distribution. Finally, the estimations of reliability and remaining useful lifetime are computed from posterior distribution. Through an example for battery, the proposed method is applied and proved to be effective.

关 键 词：可靠性评估 专家经验 信息融合 矩拟合 Bayes理论 

分 类 号：TB114.3[理学—概率论与数理统计]