可靠性试验数据分布类型的综合判别法

Synthetical Discrimination Method of Reliability Test Data Distribution Model

由于目前常用的数据分布类型判别方法如图分析法、树叶图法等都有一定的局限性,笔者在综合前人研究成果的基础上,提出了一种较为全面的判别方法——综合判别法。这种方法建立在相关系数法,似然比检验法以及峰度和偏度法三种方法的基础上,充分利用了试验数据本身所包含的信息,考虑较为全面,该法具有较好的判别效果。

Since the commonly used data distribution model discrimination methods, suchas diagram analysis method and treeleaf graph method, have some certain limitation, the au-thors, synthesizing the existing study results, puts forward a relatively over-all discriminationmethod-synthetical discrimination method. This method, built upon the basis of corelationcoefficient and kurtosis method, using completely information implied by the test data and tak-ing all the conditions into regard, has a better discrimination result.