摘要
在流变测控系统中,流变参数由扭矩值线性回归得到。为了弥补线性回归得到的斜率和截距对外界干扰容错性较差的缺陷,提出相关系数剔除法。该方法根据线性回归的相关系数抑制外界干扰,并在线性回归时应用加权最小二乘法。最后,利用现场测试的结果来评估数据处理的效果。结果表明,相关系数剔除法能够明显提高线性回归的相关系数,由加权最小二乘法得到的流变参数较普通最小二乘法更加稳定。
The rheological parameters are obtained by means of linear regression of torque values in the rheological measurement and control system.Since the slope and intercept obtained by linear regression has poor tolerance for external disturbances,the correlation coefficient elimination method is proposed.This method can restrain the external interference according to the correlation coefficient of linear regression.The weighted least square method is used for linear regression.The result of field test is used to evaluate the effect of data processing.The result shows that the correlation coefficient elimination method can improve the correlation coefficients of linear regression obviously,and the rheological parameters obtained by weighted least square method are more stable than those of ordinary least square method.
作者
吕国芳
常培晶
田正宏
LU Guofang;CHANG Peijing;TIAN Zhenghong(College of Energy and Electrical Engineering,Hohai University,Nanjing 211100,China;College of Water Conservancy and Hydropower Engineering,Hohai University,Nanjing 211100,China)
出处
《现代电子技术》
北大核心
2019年第3期173-176,共4页
Modern Electronics Technique
基金
国家自然科学基金资助项目(51279054)
江苏省优势学科建设工程项目(YS11001)~~
关键词
相关系数剔除法
加权最小二乘法
流变仪
测控系统
线性回归
粗大误差剔除法
correlation coefficient elimination method
weighted least square method
rheometer
measurement and control system
linear regression
coarse error elimination method
