无失效数据下弹上机电产品加速贮存寿命评估方法  

Accelerated storage life assessment method under zero-failure data for electromechanical products of missile

作　　者：张雷雷 刘曦 刘旭琳[1] 杨鸿均 张峰[2] ZHANG Leilei;LIU Xi;LIU Xulin;YANG Hongjun;ZHANG Feng(Xi’an Institute of Modern Control Technology,Xi’an 710065,China;School of Methanics,Civil Engineering and Architecture,Northwestern Polytechnical University,Xi’an 710129,China)

机构地区：[1]西安现代控制技术研究所,陕西西安710065 [2]西北工业大学力学与土木建筑学院,陕西西安710129

出　　处：《系统工程与电子技术》2023年第7期2287-2294,共8页Systems Engineering and Electronics

基　　金：武器装备预先研究项目(JWCX1901)资助课题。

摘　　要：针对现有加速贮存寿命评估方法难以有效应对无失效数据、两参数Weibull模型拟合高可靠、长寿命机电产品寿命分布精度不高的问题,提出了基于E-Bayes理论和三参数Weibull寿命分布模型的弹上机电产品加速贮存寿命评估方法。对两参数Weibull在拟合高可靠、长寿命机电产品寿命分布时存在的的精度不足问题进行了论述,给出了三参数Weibull下的弹上机电产品寿命分布模型。在阿伦尼乌斯模型、Nelson累积失效假定等基础上,建立了单次步进应力加速贮存寿命试验数据向多次常温应力定时截尾试验数据的等效折算方法,并引入E-Bayes理论实现了对定时截尾试验无失效数据下失效概率的有效求解。针对复杂等效折算公式中的Weibull参数估计问题,设计了粒子群优化算法实现对参数的高精度求解,并给出了基于蒙特卡罗法的贮存寿命置信下限评估方法,从而实现对弹上机电产品的高效贮存寿命评估。以某型弹上惯性测量单元为应用对象,验证了所提方法的有效性和合理性。To solve the problem of existing accelerated storage life assessment method not applicable to zero-failure case,and low precision of two-parameter Weibull distribution model in fitting high-reliability and long-life electromechanical products,an accelerated storage life assessment method for mechanical and electrical products of missile is proposed based on E-Bayes theory and three-parameter Weibull distribution model.Firstly,the insufficient precision problem of two-parameter Weibull distribution in fitting high-reliability and long-life electromechanical products is discussed,and the life distribution model of missile electromechanical products is established based on three-parameter Weibull distribution model.Single stepping accelerated storage life test results are equivalent to several time truncated test results based on Arrhenius model and Nelson failure assumption,and the E-Bayes definition and theorem is introduced to solve the problem of failure probability estimation under zero-failure case.For the Weibull parameters estimation problem in complex equivalent conversion formula,the particle swarm optimization(PSO)algorithm is given to achieve the high precision solution for parameters,and the Monte Carlo method is designed to solve the lower confidence limit of reliable storage life,finally achieving the purpose of efficient storage life assessment for electromechanical products of missile.The validity and rationality of the proposed method is tested and verified through taking certain missile inertial measurement unit(IMU)as application object.

关 键 词：弹上机电产品 加速贮存寿命 E-Bayes理论 三参数WEIBULL分布 粒子群优化算法 

分 类 号：TB114.3[理学—概率论与数理统计]