一元线性回归分析方法在压力传感器数据拟合中的应用  

Application of univariate linear regression analysis in data fitting of pressure sensor

作　　者：李久龙[1] 李玲 刘瑞敏[1] 孙旭 薛炜[1] 许腾[1] 杨懿[1] LI Jiulong;LI Ling;LIU Ruimin;SUN Xu;XUE Wei;XU Teng;YANG Yi(Beijing Institute of Aerospace Test Technology,Beijing 100074,China)

机构地区：[1]北京航天试验技术研究所,北京100074

出　　处：《计测技术》2022年第2期40-49,共10页Metrology & Measurement Technology

基　　金：国家重点研发计划(2017YFB0603702);国家自然科学基金(51976178)。

摘　　要：针对压力传感器在工程测量中需要根据计量检定数据获取压力传感器输出值和所加载标准压力值之间关系的问题,采用正态检验和方差分析的方法分析了压力传感器输出电压值和标准压力值之间的特征关系。根据相关系数值的计算结果,建立一元线性回归模型。分析了一元线性回归模型的拟合优度、模型线性关系、回归系数和模型的残差。结果表明:一元线性回归分析方法能够应用于计量检定工作中的压力传感器检定数据拟合分析,这对计量检定人员研究其他类似检定数据的拟合方法具有重要的参考价值。Aiming at the necessity to obtain the relationship between the output value of the pressure sensor and the loaded standard pressure value according to the metrological verification data when a pressure sensor is used in engineering measurement,the characteristic relationship between the output voltage of the pressure sensor and the standard pressure is analyzed by using the methods of normal test and variance analysis.According to the calculation results of correlation coefficient,a univariate linear regression model is established.The goodness of fit,linear relationship,regression coefficient and residual error of the univariate linear regression model are analyzed.The results show that the univariate linear regression analysis method can be applied to the fitting analysis of pressure sensor verification data in metrological verification,which has important reference value for metrological verification personnel to study other similar verification data fitting methods.

关 键 词：压力传感器 检定数据 一元线性回归 

分 类 号：TB9[一般工业技术—计量学] V438.4[机械工程—测试计量技术及仪器]