使用GUM法简化实验室测量不确定度评定程序

Uncertainty assessment procedure simplifying of laboratory measurement using GUM method

介绍了使用《测量不确定度表示指南》(GUM法)评定测量不确定度的使用范围及流程,分析测量不确定度的来源,建立合适的数学模型,明确不同项目中导致测量不确定度因素的概率分布,结合标准不确定度的评定、合成标准不确定度的计算、扩展不确定度的确定及报告不确定度评定结果在评定过程中合理利用数学方法进行变换,为简化实验室不确定度评定程序提出了相应的建议。在评定过程中遇到测量模型明显非线性、输入量的概率密度函数(PDF)明显非对称、输入量的PDF较大程度地偏离正态分布或t分布时,可采用蒙特卡洛法(MCM法)进行不确定度评定。

This paper introduced the usage scope and process of using GUM method to evaluate the measurement uncertainty, the uncertainty resources was analyzed, the suitable mathematical model was established, the probability distribution of factors which lead to uncertainty in different projects was clear and definite, combining with the evaluation of standard uncertainty, component standard uncertainty calculation, determination of extended uncertainty and report uncertainty evaluation results, and in the evaluation process, the reasonable mathematical method was used to vary, .so that some suggestions to simplify the laboratory uncertainty evaluation process could be proposed. In the evaluation process, if obvious nonlinear of measurement model, obvious asymmetric situation of probability density function （PDF） of input quantity, PDF of input quantity deviating from the normal distribution or T distribution to a great degree occurred, Monte carlo method （MCM method） could be used to do the uncertainty evaluation.