基于二项式展开与D_n-检验的复杂疲劳概率模型分析  

Complex Fatigue Probability Analysis Based on Binominal Formula Expansion and D_n-Test

作　　者：徐楠[1] 王威强[2] 陈举华[2] 

机构地区：[1]山东建筑大学,济南250101 [2]山东大学,济南250061

出　　处：《中国机械工程》2007年第4期396-398,共3页China Mechanical Engineering

基　　金：山东省自然科学基金资助项目(Y2002F19);山东省科学技术发展计划资助项目(012050107);山东省技术创新项目(2003[182])

摘　　要：疲劳试验数据的概率模型设计是疲劳可靠性分析的基础之一。以解析法和图形估计法进行复杂概型参数估计有时难以得到优良的拟合结果,为此提出复杂疲劳概型参数估计方法。该方法将估计的疲劳参量代入概型中,并以牛顿二项式展开求点估计值。引入非参数统计的Dn-检验法,发展了广泛适用于各种常用疲劳可靠性分布概型的假设检验方法。疲劳试件失效寿命数据验证了方法的有效性。One of the base on fatigue reliability analysis is probability design for fatigue data. It is difficult to obtain the fitted results by using the analysis method and graph estimation when making complex probability parametric estimation. Therefore, a complex probability parametric estimation method was put forwards, in which the value of fatigue parameter was substituted into the probability model, and the point estimation value was solved by expanding binominal formula of Newton. A hypotheses verification method that widely used for various commonly fatigue reliability distribution probability model was developed by introducing nonparametric statistical Dn- testing procedure. Availability of this method was verified by the failure life data of fatigue sample.

关 键 词：疲劳概型 参数估计 假设检验 Dn-检验 

分 类 号：TB114.3[理学—概率论与数理统计]