多元回归模型评价天然气流量计算机的测量误差

Multivariate regression model to evaluate measurement errors in computerized gas flow testing

流量计算机作为天然气贸易交接计量的重要计量器具,其现场测试中测量误差的确定十分重要,而测试点的选择又是其决定因素。因此,有必要开展流量计算机测试点的回归分析,合理确定测试点。开展回归模型的建立,利用建立的回归模型编制测量程序,确定测量公式,对流量计算机的测量误差进行线性分析,确定标准误差。通过线性分析,提出减少现场测试点的建议,并对减少测量点后的标准误差进行计算对比。最后给出流量计算机测量误差的合成不确定度。开展的研究工作建立完善的多元回归模型,为流量计算机现场测试过程中测量点数的选择与分布提供理论依据与实践基础。

As an important measuring instrument and basis for natural gas metering trade handover,flow computer is very important to determine the measurement error in its field test,and the selection of test point is its determining factor.Therefore,it is necessary to carry out the regression analysis of the flow computer test points to reasonably determine the test points.Carried out the establishment of the regression model,using the establishment of the regression model to prepare the measurement procedure,determine the measurement formula,the measurement error of the flow computer linear analysis,to determine the standard error.Through the linear analysis,recommendations were made to reduce the test points in the field,and the standard errors after reducing the measurement points were calculated and compared.Finally,the synthetic uncertainty of the measurement error of the flow computer is given.The research work established a perfect multivariate regression model,which provides the theoretical and practical basis for the selection and distribution of measurement points in the field test of flow computer.