基于叠合度的维修性多源数据融合方法  

Research on Maintenance Multi-source Data Fusion Based on Coincidence Degree

作　　者：徐达 关矗 李闯 XU Da;GUAN Chu;LI Chuang(Department of Arms and Control,Army Academy of Armored Forces,Beijing 100072,China)

机构地区：[1]陆军装甲兵学院兵器与控制系,北京100072

出　　处：《火力与指挥控制》2020年第4期35-39,共5页Fire Control & Command Control

基　　金：军队“十三五”装备预研共用技术基金资助项目(41404010202)。

摘　　要：针对装备维修性试验数据样本量不足的问题,采用了数据融合方法进行扩充数据样本量,通过建立折合模型对多源先验数据进行转换,解决了与现场数据差异较大难以实现数据融合的问题,采用了改进的Bayes Bootstrap法对多源数据的分布参数进行拟合,建立基于叠合度的数据融合模型,对装备维修性多源先验数据进行融合,结合某型坦克试验数据进行算例分析,证明了该方法的可行性。Aiming at the problem of insufficient sample size of equipment maintainability test data,the data fusion method is used to fuse the multi-source maintainability prior data to expand the data sample size.The folding model is used to transform the multi-source prior data,which solves the problem that the data fusion is difficult to achieve with the difference of the field data.The improved Bayes Bootstrap method is proposed to fit the distribution parameters of the multi-source prior data.The data fusion model of the combined degree is used to fuse the multi-source maintainability prior data,and combined with a certain tank test data to analyze the example,which proves the feasibility of the method.

关 键 词：叠合度 维修性 BAYES BOOTSTRAP法 先验数据 多源数据融合 

分 类 号：TJ07[兵器科学与技术—兵器发射理论与技术]