基于多目标函数优化的恒应力加速试验设计  

Design of Constant Stress Accelerated Test Based on Multi-objective Function Optimization

作　　者：张芳[1] 朱艳辉 苏林 吕萌 ZHANG Fang;ZHU Yanhui;SU Lin;LV Meng(China Meteorological Administration Training Centre,Beijing 100081,China;Troops 323812 of the PLA,Beijing 100071,China;Intellectual Property Office,Beijing 100088,China;Shijiazhuang Campus of Army Engineering University,Shijiazhuang 050005,China)

机构地区：[1]中国气象局气象干部培训学院,北京100081 [2]中国人民解放军323812部队,北京100071 [3]国防知识产权局,北京100088 [4]陆军工程大学石家庄校区,河北石家庄050003

出　　处：《电子产品可靠性与环境试验》2024年第5期45-49,共5页Electronic Product Reliability and Environmental Testing

摘　　要：对恒应力加速寿命试验进行了优化设计。首先,选取正常应力下可靠性指标估计的准确程度作为评价加速试验方案优劣的准则,推导出正常应力下对数寿命的极大似然估计(MLE)渐近方差的估计值作为单一优化目标函数;然后,在单一优化目标函数的基础上,考虑到试验代价的因素,推导出恒应力加速寿命试验优化设计的多目标函数;最后,将智能优化算法应用到单一目标优化函数和多目标优化函数的求解,通过实例验证解决了加速试验多目标函数优化的求解问题。Constant stress acceleration life test is designed optimally.Firstly,the accuracy of the reliability index estimation under normal stress is selected as the criterion to evaluate the quality of accelerated test scheme,and the asymptotic variance of maximum likelihood estimation of logarit hmic life under normal stress is derived as a single optimization objective function.Secondly,on the basis of a single optimization objective function,the multiple objective function of constant stress accelerated life test optimization design test is derived by considering the test cost.Finally,the intelligent optimization algorithm is applied to the solution of the single objective optimization function and the multiple objective optimization function.The problem of the multiple objective function optimization of accelerated test is solved through example verification.

关 键 词：加速寿命试验 优化设计 多目标函数优化 智能算法 

分 类 号：TB112[理学—数学] TB114.37[理学—应用数学]