登录    |    注册

您好,欢迎来到中国测试科技资讯平台!

首页> 《中国测试》期刊 >本期导读>测量方程、观测方程与不确定度评估

测量方程、观测方程与不确定度评估

344    2020-09-17

¥0.50

全文售价

作者:胡红波, 季文晖

作者单位:中国计量科学研究院, 北京 100029


关键词:计量学;不确定度评估;测量方程;观测方程;贝叶斯统计


摘要:

GUM不确定度评估方法以测量方程为基础,通过标准不确定度传递的方法实现被测量标准不确定度的计算,是一种前向的不确定度评估方法;从观察方程入手的贝叶斯分析方法则是一种基于概率密度函数传递的后向的不确定度评估方法。该文详细说明两种方法评估测量不确定度的过程,解释两种方法的相同与差异之处。最后通过典型的不确定度评估实例,说明对于线性的测量模型,依据GUM准则评估的结果与利用被测量无信息先验的贝叶斯统计得到的结果是一致的,但在设定较强被测量先验信息或者非线性测量模型条件下,两种方法评估的结果有一定的差异。


Measurement equation, observation equation and uncertainty evaluation
HU Hongbo, JI Wenhui
National Institute of Metrology, China, Beijing 100029, China
Abstract: The GUM(Guide to the Expression of Uncertainty in Measurement) method is based on measurement equation, and the computational tool is the law of propagation of uncertainty, so it can be termed forward uncertainty evaluation. Bayesian statistics is alternative approach for uncertainty evaluation, which is based on observation equation, and it is an inverse uncertainty evaluation method using propagation of PDF(probability density function). We illustrate the two approached in detail in the paper and explain how and where they differ. Through the application of the proposed techniques, we explain that for linear models, the GUM and Bayesian method using noninformative prior distribution provide consistent results, but for nonlinear measurement model or with strong prior knowledge for Bayesian method, some practical differences of the results exist for the two approaches.
Keywords: metrology;uncertainty evaluation;measurement equation;observation equation;Bayesian statistics
2020, 46(8):7-12  收稿日期: 2020-06-19;收到修改稿日期: 2020-07-22
基金项目: 国家质量基础的共性技术研究与应用(2017YFF0206302)
作者简介: 胡红波(1980-),男,湖北荆州市人,副研究员,硕士,主要从事振动冲击加速度计量与测试技术研究
参考文献
[1] BIPM, IEC, IFCC, ISO, IUPAC, IUPAP and OIML. Guide to the Expression of Uncertainty in Measurement [S]. Genev: International Organization for Standardization,2008.
[2] BIPM, IEC, IFCC, ISO, IUPAC, IUPAP and OIML. Evaluation of measurement data-Supplement 1 to the “Guide to the expression of uncertainty in measurement”-, Propagation of distributions using a Monte Carlo method[S]. Geneva:International Organization for Standardization, 2008.
[3] POSSOLO, ANTONIO, BODNAR, et al. Approximate Bayesian evaluations of measurement uncertainty[J]. Metrologia, 2018, 55(2): 147-157
[4] 费业泰. 误差理论与数据处理[M]. 北京: 机械工业出版社, 2015:21-40.
[5] 倪育才. 实用测量不确定度评定[M]. 北京: 中国计量出版社, 2004:53-72.
[6] 高玉英. 基于贝叶斯理论的动态不确定度评定方法研究[D]. 合肥: 合肥理工大学, 2007.
[7] 胡红波, 孙桥, 杜磊. GUM S1与基于贝叶斯方法的不确定度评估比较[J]. 计量学报, 2017, 38(4): 517-520
[8] LIRA I, WÖGER W. Comparison between the conventional and Bayesian approaches to the evaluate measurement data[J]. Metrologia, 2006, 43(4): S249-259
[9] WALTER B, MAURICE G C, RENE D, et al. Revision of the ‘Guide to the Expression of Uncertainty in Measurement’[J]. Metrologia, 2012, 49(6): 702-705
[10] ELSTER C, TOMAN B. Bayesian uncertainty analysis for a regression model versus application of GUM Supplement 1 to the least-squares estimate[J]. Metrologia, 2011, 48(5): 233-240
[11] LIRA I, WOGER W. Bayesian evaluation of the standard uncertainty and coverage probability in a simple measurement model[J]. Meas. Sci.Technol, 2001, 12: 1172-1179
[12] 陈平, 吴诚鸥, 刘应安,等. 应用数理统计[M]. 北京: 机械工业出版社, 2008:6-9.
[13] 韦来生, 张伟平. 贝叶斯统计[M]. 合肥: 中国科学技术大学出版社, 2013: 200-220.
[14] 刘金山, 夏强. 基于MCMC算法的贝叶斯统计方法[M]. 北京: 科学技术出版社, 2016:54-80.
[15] 胡红波, 于梅. 激光干涉法冲击绝对校准的研究[J]. 计量学报, 2015, 36(4): 397-402
[16] WHITE D R. In pursuit of a fit-for-purpose uncertainty guide[J]. Metrologia, 2016, 53(4): S107-S124