论文部分内容阅读
针对产品研制过程存在多阶段的试验信息和产品性能参数呈现动态变化的特点,提出一种Bayes框架下成败型系统可靠性的动态评估方法.该方法依据产品研制过程中不同研制阶段的试验数据,利用离散AMSAA可靠性增长模型对产品可靠性的变化趋势进行描述,进而对新批次的产品可靠性进行预测,并由最大熵方法给出产品可靠性验前分布的确定方法,再结合产品少量的现场试验数据,对新批次的产品可靠性做出Bayes评估.最后通过实例对比分析,表明该方法很好地解决了小子样产品的可靠性试验评估问题.
According to the characteristics of multi-stage test information and dynamic performance parameters of product development process, this paper proposes a dynamic evaluation method of reliability of the system based on Bayes framework.This method is based on the test data of different development stages in the product development process, Using the discrete AMSAA reliability growth model to describe the trend of product reliability, and then predict the product reliability of the new batch, and by the maximum entropy method to determine the pre-test distribution of product reliability, combined with a small amount of products Field test data to make a Bayes assessment of the product reliability of the new batch.At last, the comparative analysis of the examples shows that the method solves the reliability test evaluation problem of the small sample products well.