融合内外场退化数据的可靠性评估方法  

Reliability assessment method with integration of infield and outfield degradation data

作　　者：吴晓辉[1] 蔡忠义[2] 李全祥[1] 

机构地区：[1]中国飞行试验研究院,陕西西安710089 [2]空军工程大学装备管理与安全工程学院,陕西西安710051

出　　处：《合肥工业大学学报（自然科学版）》2017年第12期1589-1593,共5页Journal of Hefei University of Technology：Natural Science

基　　金：国家自然科学基金资助项目(71601183);陕西省自然科学基金资助项目(2014JQ2-7045)

摘　　要：针对性能退化过程服从Wiener过程的产品,文章运用贝叶斯统计推断法,提出了一种融合内场加速退化试验(accelerated degradation test,ADT)与外场退化信息的可靠性评估方法。考虑到内外场应力环境差异,提出了基于修正系数的Wiener过程双参数修正模型;建立了退化数据模型,得到各应力下分布参数估计值;将各加速应力下分布参数估计值折算到正常工作应力下,构成未知参数先验分布的数据样本;构建了外场退化数据下未知参数的后验分布函数,采用马尔科夫链蒙特卡洛(Markov chain Monte Carlo,MCMC)方法,得到参数的后验分布均值。实例分析结果验证了该文方法的正确性。Aiming at the degradation process of products which obeys Wiener prsessment method with accelerated degradation test（ADT） data and field degradation daward by using Bayesian statistical inference. Given the difference between field and laboratory stress environment, Wiener process double-parameters modified model with the correction fastructed. Degradation model is built to obtain estimated values of distribution parameters under eachstress. Estimated values under accelerated stress are converted into those undconstitute the data sample of prior distribution of unknown parameters. Posterior distribution function of unknown parameters under field degradation data is built and Markov chmethod is used to obtain mean values of posterior distribution. The accuracy and practical applicability of the presented method is verified by an example.

关 键 词：WIENER过程 退化数据融合 贝叶斯推断 应力环境差异 

分 类 号：TB114.3[理学—概率论与数理统计]