极小子样加速退化试验数据可靠性评估方法

Reliability Estimation Method for Extremely Small Sample Accelerated Degradation Test Data

针对极小子样的退化数据,对原始退化试验数据序列进行分段,依次从每段数据中随机抽取,组成新序列的方法扩充为大样本。考虑固定影响因素和随机影响因素,采用混合参数模型对扩充的样本进行建模,并采用two-stage方法估计模型参数。由最小二乘法估计每一样本的退化轨迹模型参数,计算混合参数模型的参数,由混合参数模型推导出产品失效时间累积概率分布函数。通过与自然贮存的大样本数据的比较表明,运用分段随机抽样方法对加速试验后的数据序列进行再抽样,可有效扩充试验样本量,无需虚拟扩充数据,解决了加速退化试验评估失效时间分布时样本量少的问题。

To assess the accelerated degradation failure time distribution of product,a method to expand extremely small sample for large sample is proposed,which divides the original degradation test data into sections and randomly selects partial data from each section.Then,the sampled data forms a new test sample.Through this method,the sample size could be expanded.Considering the fixed factors and random factors,the mixed parameter model is adopted for the modeling of the degradation paths of the expanded samples.The model parameters are estimated by the two-stage approach.First,each degradation path model parameter is evaluated with least square method.Then,the parameters of the mixed model are calculated based on the first stage’s estimation.The cumulative probability distribution function of the product’s degradation failure time could be derived through the degradation model.As compared with the large sample data of natural storage products,using the subsection random sampling method to resample the accelerated test data can effectively expand the test sample size without generating virtual data and solve the problem of insufficient samples when evaluating the failure time distribution in accelerated degradation test.