半导体器件并联系统可靠性评估的经验Bayes方法

Empirical Bayes Method of Reliability Evaluation forSemiconductor Parallel System

为了提高机载开关电源中半导体器件并联系统可靠性评估的准确性,运用经验 Bayes 法和经典的统计方法,研究了该系统的可靠性评估问题。分别给出了系统可靠性指标的经验 Bayes 估计,极大似然估计。利用 Monte-Carlo方法,对两种估计结果进行了比较,结果表明,经验 Bayes 估计的最大绝对误差为 0.07,它小于极大似然估计的最大绝对误差 0.368。

In order to improve the accuracy of the reliability evaluation for semiconductor parallel system, evaluated are the reliability for this system by using empirical Bayes and classical statistical methods. Empirical Bayes estimation (EBE) and maximum likelihood estimation (MLE) of reliability index are given, respectively. MLE and EBE results were also compared by Monte-Carlo method. Obtained results show that the maximun absolute error of EBE is 0.07, and it is less than the maximum absolute error of MLE which is 0.368.