II型截尾步进应力加速寿命试验的优化设计  

OPTIMAL DESIGN FOR STEP-STRESS ACCELERATED LIFE TEST WITH CENSORING II

作　　者：王煜[1] 

机构地区：[1]青海师范大学数学系,青海西宁810008

出　　处：《西南石油大学学报（自然科学版）》2007年第3期122-125,共4页Journal of Southwest Petroleum University(Science & Technology Edition)

基　　金：国家社科基金项目(2006CTJ004)。

摘　　要：为在短时间内评价长寿命产品的可靠性指标,一般采用恒定应力、序进应力和步进应力等加速寿命试验法,以加速产品失效,估计产品指标。针对步进应力加速寿命试验,在多参数加速寿命方程下,以各失效机理对数均值渐进方差之和最小为目标,提出了II型截尾步进应力加速寿命试验的优化设计模型,并通过理论推导,得到了相应的优化设计方案。获得了定数截尾步加试验在满足截尾数据单调的约束条件下,使对数均值渐近方差最小的优化设计方案为各应力下截尾数据相等的结论。该成果是对简单步进应力加速寿命试验的拓展和延伸,据此可指导和组织试验,提高评价结果的精度,在石油工程等领域中具有理论和实用价值。In order to evaluate the reliability index of the long-life products in a short period of time, constant-stress, progressive stress and step-stress accelerated life tests are usually used to speed up failure products and estimate their indicators. In view of step-stress accelerated life test, in multi-parameter accelerated life equation, aiming at the minimun sumation of every failure mechanism and the variance of the MLE, optimal design model of Type II step-stress accelerated life tests, through the theoretical derivations, the corresponding optimal design plan is obtained. The main conclusion is that for a step-stress test which satisfies the condition of intercept data is monotonic.The optimal plan making minimun of the variance of the MLE is: The intercept data is equal to every stress.The results are the expansion and extension of simple step-stress accelerated life tests, which would guide and organize experiments, improve the accuracy of evaluation findings. It is of theoretical and practical value.

关 键 词：可靠性分析 应力 优化设计 Ⅱ型截尾 失效机理 

分 类 号：O213.2[理学—概率论与数理统计]