| 高可靠性软件的极值统计分析 |
| |
| 引用本文： | 陈云,张志华.高可靠性软件的极值统计分析[J].海军工程大学学报,2005,17(2):95-99. |
| |
| 作者姓名： | 陈云 张志华 |
| |
| 作者单位： | 海军工程大学,理学院,湖北,武汉,430033 |
| |
| 摘 要： | 提出了极值统计分析的方法.对于失效数据极值分布函数的拟合检验,采用了相关系数法;对于母体分布的参数估计,则分别采用了最小二乘估计(LSE)和极大似然估计(MLE);最后通过一个实例说明了统计过程,并用模拟的方法证明了最小二乘估计较好.
|
| 关 键 词： | 软件可靠性 高可靠性软件 软件可靠性增长模型 极值统计 失效时间 |
| 文章编号： | 1009-3486(2005)02-0095-05 |
| 修稿时间： | 2004年10月8日 |
Using statistics of extremes for highly reliable software reliability analysis |
| |
| CHEN Yun,ZHANG Zhi-hua.Using statistics of extremes for highly reliable software reliability analysis[J].Journal of Naval University of Engineering,2005,17(2):95-99. |
| |
| Authors: | CHEN Yun ZHANG Zhi-hua |
| |
| Abstract: | The reliability evaluating of highly reliable software is important .So the analytical method of statistics of extremes is presented. In the fit test of the failure data extreme value distribution function, a covariate coefficient method is used. For the parameter estimator of parent distribution, this paper introduces least square estimation (LSE)and maximum like-hood estimation (MLE),and then an example is given to show the whole statistic process and it is provedthat the least square estimator is better through the simulation method. |
| |
| Keywords: | softwarereliability highly reliable software software reliability growth model statistics of extremes time to failure |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
