基于均匀设计理论的非矩形试验区域恒加寿命试验优化设计方法
|
摘要
针对基于Escobar和Meeker设计思想的非矩形试验区域最优试验方案设计,当边界较为复杂时失效概率最大点难以求得的问题,以正常应力下寿命估计值的渐近方差最小为目标,在非矩形区域内选择两个应力水平点分别作为最高和最低应力水平点,以最高、最低应力水平以及各试验样本分配比例为设计变量,提出了一种基于均匀设计理论的非矩形试验区域恒加寿命试验方案优化设计方法。模拟计算结果表明,与基于Escobar和Meeker设计思想的最优试验方案相比较,所提出的设计方法具有相同的估计精度,为非矩形试验区域加速寿命试验方案设计提供了一种简便的方法。
Aiming at the problems of hardly finding out the points of the maximum probability of failure on the complicated boundary curves when designed the optimal multiple stress accelerated life test plan on the non-rectangle test region based on Escobar-Meeker method,minimum variance of product life estimated value at normal working stresses was taken as design object for constant stress accelerated life test method.After selecting two stress level points as the highest level point and the lowest level point on the non-rectangle test region,the highest stress level,the lowest stress level of each accelerated factor and allocation proportion of test samples were adopted as design variables;the optimum design method of multiple stress accelerated life test on the non-rectangle test region was built based on uniform design theory.The computer simulation results show that the optimized test plan has the same estimation precision of test data compared to the optimized test plan based on the Escobar-Meeker method.The research fruits provide a simple method for the test plan design of multiple stress accelerated life test on the non-rectangle test region.
Aiming at the problems of hardly finding out the points of the maximum probability of failure on the complicated boundary curves when designed the optimal multiple stress accelerated life test plan on the non-rectangle test region based on Escobar-Meeker method,minimum variance of product life estimated value at normal working stresses was taken as design object for constant stress accelerated life test method.After selecting two stress level points as the highest level point and the lowest level point on the non-rectangle test region,the highest stress level,the lowest stress level of each accelerated factor and allocation proportion of test samples were adopted as design variables;the optimum design method of multiple stress accelerated life test on the non-rectangle test region was built based on uniform design theory.The computer simulation results show that the optimized test plan has the same estimation precision of test data compared to the optimized test plan based on the Escobar-Meeker method.The research fruits provide a simple method for the test plan design of multiple stress accelerated life test on the non-rectangle test region.
引文
[1]Nelson W.Accelerated Testing Statistical Models,Test Plans and Data Analysis[M].New York:A Wileyinte-science Publication,JohnWiley&Sons,1990.
[2]陈文华,冯红艺,钱萍,等.综合应力加速寿命试验方案优化设计理论与方法[J].机械工程学报,2006,42(12):101-105.Chen Wenhua,Feng hongyi,Qian Ping,et al.Optimal Design of Multiple Stresses Accelerated Life Test Plan[J].Chinese Journal of Mechanical Engineering,2006,42(12):101-105.
[3]Gao Liang,Chen Wenhua,Liu Juan,et al.Design Criteria for Planning Multiple Stress Accelerated Life Test[C]//The Proceedings on 2011 9th International Conference on Reliability Maintainability&Safety.Guiyang,2011:1141-1146.
[4]陈文华,钱萍,马子魁,等.基于定时测试的综合应力加速寿命试验方案优化设计[J].仪器仪表学报,2009,30(12):2345-2550.Chen Wenhua,Qian Ping,Ma Zikui,et al.Optimal Design of Multiple Stresses Accelerated Life Test Plan under type I Censoring[J].Chinese Journal of Scientific Instrument,2009,30(12):2345-2550.
[5]Nelson W B.Accelerated Testing:Statistical Models,Test Plans,Data Analyses[M].New York:Wiley,1990.
[6]Park J W,Yum B J.Optimal Design of Accelerated Life Testing with Two Stress[J].Naval Research Logistics,1996,43(6):863-884.
[7]Guo Huairui,Pan Rong.D-optimal Reliability Test Design for Two-stress Accelerated Life Tests[C]//Proceedings of the 2007IEEE IEEM.Singapore,2007:1236-1240.
[8]Chen Wenhua,Gao Liang,et al.Optimal Design of Multiple Stress Accelerated Life Test Plan on the Non-rectangle Test Region[J].Chinese Journal of Mechanical Engineering,2012,25(6):1231-1237.
[9]Escobar L A,Meeker W Q.Planning Accelerated Life Tests with Two or More Experimental Factors[J].Technometrics,1995,37(4):411-427.
[10]方开泰,马长兴.正交与均匀试验设计[M].北京:科学出版社,2001.
[11]Gao Liang,Chen Wenhua,Qian Ping,et al.Optimal Design of Multiple Stresses Accelerated Life Test Plan Based on Transforming the Multiple Stresses to Single Stress[J].Chinese Journal of Mechanical Engineering,2014,28(6):1125-1132.
[2]陈文华,冯红艺,钱萍,等.综合应力加速寿命试验方案优化设计理论与方法[J].机械工程学报,2006,42(12):101-105.Chen Wenhua,Feng hongyi,Qian Ping,et al.Optimal Design of Multiple Stresses Accelerated Life Test Plan[J].Chinese Journal of Mechanical Engineering,2006,42(12):101-105.
[3]Gao Liang,Chen Wenhua,Liu Juan,et al.Design Criteria for Planning Multiple Stress Accelerated Life Test[C]//The Proceedings on 2011 9th International Conference on Reliability Maintainability&Safety.Guiyang,2011:1141-1146.
[4]陈文华,钱萍,马子魁,等.基于定时测试的综合应力加速寿命试验方案优化设计[J].仪器仪表学报,2009,30(12):2345-2550.Chen Wenhua,Qian Ping,Ma Zikui,et al.Optimal Design of Multiple Stresses Accelerated Life Test Plan under type I Censoring[J].Chinese Journal of Scientific Instrument,2009,30(12):2345-2550.
[5]Nelson W B.Accelerated Testing:Statistical Models,Test Plans,Data Analyses[M].New York:Wiley,1990.
[6]Park J W,Yum B J.Optimal Design of Accelerated Life Testing with Two Stress[J].Naval Research Logistics,1996,43(6):863-884.
[7]Guo Huairui,Pan Rong.D-optimal Reliability Test Design for Two-stress Accelerated Life Tests[C]//Proceedings of the 2007IEEE IEEM.Singapore,2007:1236-1240.
[8]Chen Wenhua,Gao Liang,et al.Optimal Design of Multiple Stress Accelerated Life Test Plan on the Non-rectangle Test Region[J].Chinese Journal of Mechanical Engineering,2012,25(6):1231-1237.
[9]Escobar L A,Meeker W Q.Planning Accelerated Life Tests with Two or More Experimental Factors[J].Technometrics,1995,37(4):411-427.
[10]方开泰,马长兴.正交与均匀试验设计[M].北京:科学出版社,2001.
[11]Gao Liang,Chen Wenhua,Qian Ping,et al.Optimal Design of Multiple Stresses Accelerated Life Test Plan Based on Transforming the Multiple Stresses to Single Stress[J].Chinese Journal of Mechanical Engineering,2014,28(6):1125-1132.