GUM评定数字式气压计不确定度的适用性验证

Evaluation of Measurement Uncertainty for Digital Barometer Based on Monte Carlo Method

摘要:

为确保数字式气压计测量不确定度的可靠性，需对现有的测量不确定度表示指南（GUM）评定数字式气压计不确定度的方法进行适用性验证。本文根据国际标准ISO/IEC Guide 98-3:2008 附件1给出的蒙特卡洛法(Monte Corlo Method,简称MCM)验证方案，以PTB220型数字式气压计为研究对象，分别采用自适应MCM对GUM方法进行不确定度的评定；将评定结果对比求偏差；根据偏差值是否超限来判定GUM方法适用性。结果表明：所采用的自适应MCM通过迭代增加试验次数，实时判断试验统计意义上的稳定，能有效确保评定结果的可靠性，且仅通过2×10⁴样本量，计算出对应的评定结果。通过将GUM评定结果与其对比求偏差，分析得出GUM在进行数字式气压计的不确定度评定时偏离程度超过规定范围，说明GUM未能通过验证，即GUM可能不宜适用于该仪器不确定度的评定。

Abstract:

The International Joint Committee for Guides in Metrology (JCGM) puts forward “Evaluation of measurement data-propagation of distributions using the Monte Carlo method” (MCM) in ISO/IEC Guide 98-3:2008 supplement 1. MCM is a method based on the law of propagation of distribution (LPD). Therefore, compared with the GUM (Guide to the Expression Uncertainty in Measurement) evaluation method, MCM has higher accuracy and reliability when applied to the evaluation of measurement uncertainty. There are two schemes of MCM: one is single batch MCM and the other is adaptive MCM. Since adaptive MCM can effectively save computer resources and is more efficient in evaluation, it is widely used in the evaluation of measurement uncertainty in various fields. To ensure the accuracy of the detection data of the digital barometer, it needs to be calibrated periodically, and to characterise the reliability of the calibration results, and needs to be analysed and evaluated in terms of the measurement uncertainty (referred to as “uncertainty”) of the calibration results. To effectively improve the accuracy of the calibration results of the digital barometer and provide technical support for the credibility and reliability of meteorological observation data, taking the calibration results of PTB220 digital barometer as the research object, the adaptive MCM method is used to evaluate the uncertainty of the measurement results of the digital barometer. The adaptive MCM method is used to compare and verify the GUM method in accordance with the validation scheme given in the ISO/IEC Guide 98-3: 2008 supplement 1. The results show that the deviation of GUM in the uncertainty of the digital barometer exceeds the specified range, i.e., the absolute deviation of the left and right endpoints of the 95% inclusion interval between GUM and MCM uncertainty exceeds the numerical tolerance of 0.005 hPa, which is determined by the effective number of digits of the standard uncertainty, so the GUM method fails to pass the validation. Through the above results, it is indicated that the GUM method may be unsuitable for the verification of the uncertainty of this instrument. The above results indicate that the GUM method may not be suitable for the evaluation of the uncertainty of this instrument.