基于KULLBACK信息融合方法的系统可靠性评估

在产品具有多源验前信息的情形下，讨论冷贮备系统的可靠性评估问题，运用Kullback信息作为分布之间距离的度量，在Kullback信息的融合准则下对多个先验分布进行融合。并以融合后的先验分布作为产品的最终验前分布，对冷贮备系统的可靠性指标进行Bayes估计。最后进行的计算机随机模拟结果表明，文中所提出的方法合理且便于应用。

Reliability Evaluation of System Based on the Fusion of Kullback Information

Based on the Kullback information fusion method, the reliability evaluation of cold stand-by system is discussed. When product has the multiple prior sources, using the Kullback information as the distance measure of distribution, and combining the prior distribution under the fusion rule of Kullback information. The fused prior distribution is taken as the final prior distribution of the product, and using Bayes method to analyze the reliability performances for the cold stand-by system. Finally, the computer simulation is also carried on, the result shows that the estimator is reasonable and convenient for applying.