对数正态分布场合下产品加固性能的Bayes评估方法
Bayesian Assessment for Product's Radiation Hardening\\Performance under Lognormal Distribution
-
摘要: 总结了评估产品抗辐射能力实验中常常遇到的三类数据. 针对产品抗辐射能力服从对数正态分布、实验样本数据为成败型实验数据的情形, 运用Bayes方法给出了在小样本情况下, 产品平均抗辐射能力大于给定指标要求的后验概率的计算方法以及在给定置信度下产品平均抗辐射能力置信下限的计算方法. 讨论了未知参数先验分布的确定方法, 并给出了评估方法的具体例子.Abstract: We summarize three kinds of data often occurred in experiments which are done to estimate the product's radiation hardening (RH) performance. A method based on Bayesian theory is proposed in this paper. This method is applied to evaluate the product's RH performance if the product's RH performance abides by the lognormal distribution and the samples are Zero-One data. The prior distributions of unknown parameters are also discussed, and two examples are given to illustrate the proposed method.