疲劳寿命分布模型的统一描述

A General Model of Fatigue Life

在大量仿真试验的基础上，发现：①正态分布和3参数Weibull分布之间存在解析关系；②在工程中常见的情况下，3参数Weibull分布可有效地拟合正态和对数正态数据；③当一组数据可同时用2参数Weibull分布（B_1），正态分布（B_2），对数正态分布（B_3）和3参数Weibull分布（B4）描述时，则其B基值间存在以下关系B_1＜B_2＜B_3＜B_4；④在工程中常见的情况下，以3参数Weibull分布拟合正态和对数正态数据，其B基值的相对误差｜ε｜＜5％。

The following results are discovered from many of data of simulation experiments: ①A ana-lytic relation exists between normal distribution and 3 parameter Weibull distribution. ②Normal or lognor-mal data can be well fit by 3 parameter Weibull distrbution in the case of general engineering application.③The following relationship exists:B_1 <B_2<B_3<B_4,when a set of data can be well fit by 2 parameterWeibull distribution, normal distribution, lognormal distribution and 3 parameter Weibull distribution re-spectively, among which, B_1,B_2,B_3 and B_4 is their B value. ④ The relative error |ε| of B value is lowerthan 5 percent when 3 parameter Weibull distribution is used to fit normal or lognormal data with respect'to general engineering applications.