基于性能退化数据的可靠性评估

传统的可靠性评估方法一般基于失效寿命数据，对于高可靠长寿命产品，很难通过寿命试验和加速寿命试验获得失效寿命时间，这种情况下产品的性能退化数据可提供可靠性评估的重要信息。在基于失效数据可靠性评估及性能可靠性评估方法的基础上，提出了基于退化轨迹与基于性能退化量分布两种可靠性评估方法，以对具有退化失效机理的高可靠长寿命产品进行可靠性评估。基于退化轨迹的可靠性评估方法首先选取合适的退化轨迹模型，利用退化数据对退化轨迹进行模型拟合得到模型参数，然后根据退化轨迹外推得到伪失效寿命，最后根据完全寿命试验数据的统计分析方法进行统计分析得到产品可靠性信息。基于性能退化量分布的可靠性评估方法首先计算不同测量时刻产品性能退化量的统计特征，对其进行曲线拟合得到统计特征随时间变化的函数，即性能退化量分布参数，最后利用可靠性与性能退化量分布的关系评估产品的可靠性。本文最后给出了两种可靠性评估方法的应用案例，验证了方法的正确性与有效性。

Reliability Assessment Based on Performance Degradation Data

The traditional reliability assessment methods are based on failure lifetime data. The failure lifetime data of high- reliability long-lifetime products are obtained difficultly by life test and accelerated life test. Degradation data can provide with useful information about the reliability assessment for these products. In this paper, the method based on performance degradation path and the ones based on distribution of performance degradation measures are proposed to assessing reliability of high-reliable long-lifetime products with degradation failure mechanism. In the first method, we choose an appropriate model of degradation path, and then fit degradation data to the degradation path with which pseudo failure life data can be obtained by extrapolating the degradation path to a certain critical value, so reliability information can be obtained by the analysis of pseudo failure life data. In the second method, we first compute the statistical characteristics of the amount of degradation at different measurements, and then fit the statistical characteristics to functions of time. We can assess the reliability of products with the relationship between reliability and the distribution of degradation amount. Finally, this paper gives a case to apply the two methods whose results demonstrate that the methods are correct and valid.