不同统计方法在磁粉检测能力验证结果分析中的应用  

Application of different statistical methods in the analysis of magnetic particle testing proficiency verification results

作　　者：宋飞 SONG Fei(Shanghai Research Institute of Materials,Shanghai 200437,China)

机构地区：[1]上海材料研究所,上海200437

出　　处：《无损检测》2022年第11期32-35,共4页Nondestructive Testing

基　　金：上海市2020年度科技创新行动技术标准项目(20DZ2203800)。

摘　　要：格拉布斯统计法、中位值和标准化四分位距法和迭代法是3种较为常用的能力验证结果统计分析方法,以两年的磁粉检测能力验证结果为例,对3种方法进行了比较。分析结果表明,当数据符合或基本符合正态分布时,选用迭代法会更为可靠;当数据不符合正态分布时,中位值和标准化四分位距法和迭代法均会失效,而选用格拉布斯统计法,并采用合适的剔除水平α~*,可以得到更合理的统计结果。The Grubbs statistical method, the median and normalized interquartile range method and the iterative method are three commonly used statistical analysis methods for proficiency testing results. A comparison was made for above-mentioned three methods and the analytical results showed that when the data conformed to or basically conformed to the normal distribution, the iterative method was more reliable. When the data did not conform to the normal distribution, the median and standardized interquartile range methods and the iterative method would be not applicable, whereas more reasonable statistical results could be obtained by using the Grubbs statistic method and adopting an appropriate elimination level α.

关 键 词：能力验证 格拉布斯统计法 中位值和标准化四分位距法 迭代法 

分 类 号：O17[理学—数学] TG115.28[理学—基础数学]