小子样条件下某型加速度计步降加速寿命试验优化设计  

Optimal design method for accelerometer step-down-stress accelerated life testing on condition of small sample

作　　者：罗赓 穆希辉 牛跃听 杜峰坡 陈建华[2] 王永南[2] 

机构地区：[1]中国华阴兵器试验中心,陕西华阴714200 [2]军械工程学院,石家庄050003 [3]军械技术研究所,石家庄050003

出　　处：《中国惯性技术学报》2015年第5期696-700,共5页Journal of Chinese Inertial Technology

基　　金：国家自然科学基金(61471385)

摘　　要：为了解决小样本条件下某型加速度计进行步降应力加速寿命试验的方案优劣难以评估的难题,提出一套适合于工程应用的优化设计方法。首先针对自然贮存试验的统计数据,采用基于极小卡方估计和拟合优度检验相结合的贮存可靠性评估方法以及最小二乘法建立寿命分布模型和加速方程。然后进行高温老化试验,依据失效模式分析机制和已建的失效树确定加速应力上限,完善加速寿命试验先验知识。最后对试验样本量进行可信度评定并根据先验知识采用基于思维进化算法优化BP神经网络的仿真方法进行优化设计,得到最优试验方案并通过试验进行验证。结果表明,该最优试验方案计算得到的先验参数值与原值之间的相对误差不超过1.7%,可以满足工程应用。An optimal design method is presented for accelerometer step-down-stress accelerated testing on condition of small sample. Firstly,the storage reliability evaluation method combining minimum chi-square estimation,goodness fit of test and the least square method are used to establish lifetime distribution function and acceleration equation for the statistical data of natural storage environmental test. Secondly,the accelerated stress upper bound is determined based on the failure mode analysis mechanism and the failure tree though high-temperature aging testing. Finally,the credibility evaluation is made for the test sample size,and an optimal experiment scheme is carried out based on MEA-BP neural network for Monte-Carlo simulation. The results show that the relative error between the prior parameter value calculated by the optimal scheme and the original values is less than 1.7%,meeting the requirements of the engineering applications.

关 键 词：加速寿命试验 优化设计 步降应力 小子样 加速度计 

分 类 号：TP114.3[自动化与计算机技术—控制理论与控制工程]