维修性试验验前信息融合方法

Research on Fusion Method of Prior Information in Maintainability Test

针对小子样维修性试验因数据异总体而导致验前信息融合精度低的问题,提出了基于信息可用度的验前信息折合模型,解决了因数据异总体无法融合的问题;利用随机加权法对验前分布参数进行了拟合,准确地表述了验前分布;提出了基于支持度的融合权值确定方法,提高了各验前分布融合权值分配的合理性;最后,通过某装备维修性试验验前信息融合实例,表明了该方法具有一定的工程应用价值。

In order to solve the problem of low precision of prior information fusion caused bydifferent data population in maintainability test,the conversion model of prior information is proposedbased on the availability of information,the problem of unable to fuse the information because ofdifferent data population is solved.The parameters of prior distribution are fitted by random weightingmethod,prior distribution is described exactly.Weight determination method based on the degree ofsupport is proposed,and the reasonability of weighting fusion is improved.Finally,the engineeringapplication value of the fusion method proposed by this thesis is proved by the information fusioninstance of a certain equipment in maintainability test.