蒙特卡洛法评定在线pH计示值误差的不确定度

以在线p H计为例,考察了在线酸度计示值误差不确定度的分布规律,利用蒙特卡洛法评定示值误差不确定度。对于0.01级的在线p H计,蒙特卡洛法与GUM法评定结果的差值为9.1%,小于不可靠性(20%);对于0.1级的在线p H计,蒙特卡洛法与GUM法评定结果的差值为3.8%,小于不可靠性(10%)。通过比较得出结论,采用GUM法验证了蒙特卡洛法(MCM)根据JJF 1547–2015评定在线p H计示值误差不确定度的方法是有效且适用的。尤其在测量模型非线性以及输出量的概率密度函数(PDF)较大程度地偏离正态分布或t分布等GUM法不适用的场合,蒙特卡洛法是评定在线分析监测仪器仪表示值误差不确定度的重要手段。

Uncertainty Evaluation of Indication of On-line pH Meters by MCM

Taking the on-line pH meter as an example, the distribution of uncertainty of indication error of the on-line pH meter was investigated, MCM method was used for the uncertainty analysis of the indication error. For 0.01 level on-line pH meter, the difference between two methods of MCM and GUM was 9.1%, which was less than the unreliability (20%);for 0.1 level on-line pH meter, the difference between the two methods was 3.8%, which was less than the unreliability (10%). A conclusion was gained that GUM method was suitable to verify the validity of MCM for evaluating the uncertainty of the indication error of on-line pH meter according to JJF 1547-2015. Especially in the situations which were not suitable for GUM method (the measurement model was nonlinear and the output PDF deviates from the normal distribution or t distribution), MCM was an important method to evaluate the uncertainty of indication error of on-line analytical and monitoring instruments.