Research And Application Of Aerospace Product Reliability Growth Prediction Based On Variable Parental Environment Data

As during the process of selecting spaceflight product failure data,equipments under development phase all experienced reliability growth process that possessing varied population and varying environment.Therefore,conducting conversion to experimental data by introducing time environmental coefficient can obtain experimental failure data under standard environment,which also can conduct objective evaluating to products under specific condition,at the same time,it guarantees high accuracy of evaluating result.The article adopts AMSAA model that is often used in engineering field,utilizing failure data in experimental process collected by product and its similar type or similar products in development stage.Basing on comprehensive utilization of mean value and variance for test of goodness of fit statistics this basic thinking,it puts forward a kind of method to confirm time environmental referring factor by regarding minimum coefficient of dispersion of test of goodness of fit statistics as optimizing criterion,then conduct conversion to original failure data and solve all relative parameter estimation of model by utilizing time environmental referring factor.The article confirms a group of time environmental referring factor value that conforms to condition by improving method and solves data after conversion,then let MTBF single side lower confidence limit increase by 16% of product molding when confidence coefficient is 0.9,which is superior to literature result;And when confidence coefficients respectively are from 0.91 to 0.99,we find MTBF single side lower confidence limit of one spaceflight product molding is superior to literature result.Therefore,under different confidence levels,the method is reasonable and feasible,which conforms to engineering practice.