基于MCM的多项式最小二乘法拟合的不确定度评定  

Uncertainty Assessment of Polynomial Least Squares Fitting Based on MCM

作　　者：李翔 LI Xiang(Guangzhou NEM Metrology&Test Technology Co.,Ltd.,Guangzhou 510070)

机构地区：[1]广州中广测计量检测技术有限公司,广州510380

出　　处：《环境技术》2024年第7期105-110,140,共7页Environmental Technology

摘　　要：通过利用仪器校准证书中的校准结果内容,可以从中得到测量值对应的修正值。为了获得任意测量值的修正数据,采用多项式最小二乘法进行拟合是一种常用手段。在研究拟合后修正值的不确定度的评定时发现,测量值的扩展不确定度会对其评定产生较大的影响。本文基于蒙特卡洛法(CMC法),对多项式最小二乘法拟合的修正值进行不确定度评定,通过利用Python编程实现MCM法评定其不确定度的步骤,获得不确定度结果。因此,MCM法在评定多项式最小二乘法修正值拟合的不确定度评定提供一定的技术参考。By utilizing the calibration results from instrument calibration certificates,correction values corresponding to measurement values can be obtained.To acquire correction data for any measured value,polynomial least squares fitting is a common method.In the study of uncertainty evaluation for the fitted correction values,it was found that the expanded uncertainty of the measurement values has a significant impact on its assessment.This paper evaluates the uncertainty of the correction values fitted by the polynomial least squares method based on the Monte Carlo Method(CMC).By using Python programming to implement the steps of evaluating its uncertainty with the MCM,the uncertainty results are obtained.Therefore,the MCM provides a technical reference for the uncertainty evaluation of correction value fitting by the polynomial least squares method.

关 键 词：修正值 多项式最小二乘法拟合 不确定度评定 蒙特卡洛法 Python软件 

分 类 号：TM932[电气工程—电力电子与电力传动]