一种双时间尺度函数Wiener过程性能退化模型

为通过性能退化数据评估产品的可靠性，基于广义Wiener过程提出了一种双时间尺度函数模型以描述产品退化过程。采用极大似然估计(MLE)法求解模型参数，推导了其寿命分布(FTD)，进而进行可靠度分析和寿命评估。对模型进行了Monte Carlo模拟验证和实例应用，并将目前常用的混合效应Wiener过程模型作为参考方法进行了对比分析。结果表明:根据90%可靠度曲线的定义，该方法较参考方法的结果更为合理、准确。进一步的寿命评估结果表明，该方法与参考方法的平均寿命相差232.9h，90%可靠寿命结果相差109.4h，且参考方法计算结果偏于危险。 

An Wiener process perfermance degradation model with two transformed time scales function

A Wiener process degradation model with two transformed time scales is proposed for reliability assessment based on products deterioration information. The maximum likelihood estimation (MLE) for model parameters is constructed. Then the failure time distribution (FTD) is derived, the reliability and life can be inferred accordingly. A comprehensive Monte Carlo simulation study and a real application have been conducted, and the widely adopted Wiener process model with mixed effects is adopted as a reference for a better understanding. According to the definition of the 10th degradation percentile curve, comparative results illustrated that the proposed method can effectively improve the reliability assessment precision. Regarding further lifetime assessment, the median life derived from the proposed method was 232.9h smaller than that derived from the referenced method, and the 10th failure time distribution percentile was 109.4h. This showed that the results from the reference method were more dangerous in practical engineering.