摘要
基于Weibull过程的可靠性增长试验Bayes分析,首先推导定时截尾情况下某些关键参数的Bayes验后分布形式。然后,在截尾时间内的期望故障数的验前分布服从Gamma分布情况下,导出简捷的验后分布形式。在利用该结果进行可靠性评定时,需采用矩相等的转换方法进行验前信息的转换和失效强度验后分布的估计。仿真表明,该方法精度较好,评估过程可靠。
In the Bayesian analysis of reliability growth test based on Weibull process, the Bayesian posterior distributions of some key parameters in the time truncated test is first developed. Then, by assuming the prior distribution of the expectation exception list in the truncated time to have the Gamma distribution form, thus a concise posterior distribution form is obtained, When using this result to carry out the reliability evaluation, the conversion of prior information and the estimation of failure intensity＇s posterior distribution are realized in the principle of moment equivalence. The simulation shows that this method can obtain believable evaluation results with reliable precision.
出处
《兵工自动化》
2007年第11期30-31,34,共3页
Ordnance Industry Automation
基金
国家部委级预研基金重点项目(xxxxxxxxxxx06KG0164)
教育部博士点基金(2005999807)
