小子样复杂产品可靠性试验评定专家信息融合方法  

Fusion of Expert Information in Reliability Test and Evaluation of Small Sample Complex Products

作　　者：吴琳丽[1] 潘光[1] 

机构地区：[1]西北工业大学航海学院,陕西西安710072

出　　处：《机械与电子》2014年第11期3-6,共4页Machinery & Electronics

基　　金：国家自然科学基金资助项目(51279165);高等学校博士学科点专项科研基金资助项目(20126102110009)

摘　　要：运用Bayes方法进行小子样复杂产品可靠性试验评定时,先验分布的获取与表示是一个关键问题。研究了利用专家信息进行先验分布融合的方法,对基于AHP方法的先验分布权重因子的确定及评价判断矩阵的一致性检验等问题进行了分析。以某水中兵器导航系统为例,运用专家信息,在获得评价判断矩阵的基础上,对多源先验信息的融合进行了仿真验证。The acquisition and representation of the prior distribution is a key problem in reliability test and evaluation of small sample complex prod-ucts by using Bayes method.In this paper,the ap-plication of expert information in prior distribution fusion is studied.The method of the weight factor’ s calculation and the judgment matrix’s consisten-cy checking based on AHP is presented.In allu-sion to the underwater weapon’s guided system, with the judgment matrix based expert informa-tion,the fusion of multiple sources prior informa-tion is validated by simulation.

关 键 词：小子样复杂产品 可靠性试验评定 层次分析法 专家信息 

分 类 号：TB114.3[理学—概率论与数理统计] TJ630.1[理学—数学]