指数型产品多阶段可靠性增长的Bayes模型

Bayes Model of Exponential Product Multi-stage Reliability Growth

基于D-S证据理论融合验前信息,建立了一种适用于小子样复杂系统多阶段可靠性增长分析的Bayes模型。选择Gamma分布作为失效率的先验分布,通过多源可靠性信息融合,将专家经验转换成概率分布,利用D-S证据理论融合多个专家信息,确定了先验分布参数,结合产品研制阶段试验数据,根据Bayes统计推断理论,给出了失效率、平均故障间隔时间(MTBF)和可靠度的Bayes点估计和置信下限。以兆瓦级直驱式风力发电机研制试验验证了该模型的有效性。

A multi-stage reliability growth Bayes model was built based on D-S evidence theory and prior information,that was adaptive to small-sample complicated systems.Gamma distribution was used as prior distribution of failure rate.Through integrated usage of multiple-source reliability infor-mations,the expertise was converted into probability distribution.Using D-S evidence theory and mul-tiple experts’informations,the prior distribution parameters were known.With test data at product development and Bayes statistical inference theory,the failure rate,MTBF,point estimation and lower confidence were obtained.The validity of this approach was illustrated by the development and test of M W direct-drive wind turbine .