高级检索
张忠典, 李冬青, 尹孝辉. 基于回归分析理论的点焊质量监测模型[J]. 焊接学报, 2001, (4): 31-35.
引用本文: 张忠典, 李冬青, 尹孝辉. 基于回归分析理论的点焊质量监测模型[J]. 焊接学报, 2001, (4): 31-35.
ZHANG Zhong-dian, LI Dong-qing, YIN Xiao-hui. Study on Spot Welding Quality Monitoring Models by Linear Regression Theory[J]. TRANSACTIONS OF THE CHINA WELDING INSTITUTION, 2001, (4): 31-35.
Citation: ZHANG Zhong-dian, LI Dong-qing, YIN Xiao-hui. Study on Spot Welding Quality Monitoring Models by Linear Regression Theory[J]. TRANSACTIONS OF THE CHINA WELDING INSTITUTION, 2001, (4): 31-35.

基于回归分析理论的点焊质量监测模型

Study on Spot Welding Quality Monitoring Models by Linear Regression Theory

  • 摘要: 以动态电阻与接头质量参数之间关系为例,用线性回归分析技术,系统地研究了点焊加热过程中的监测信息与点焊接头质量参数之间的一元和多元回归模型关系。结果表明:(1)各种回归模型总体线性关系经统计检验均为高显著,可以用于点焊接头质量监测;(2)一元回归模型的决定系数仅为0.15~0.33,多元回归模型的决定系数则可以提高到0.59~0.79,说明发展多参量综合监测技术是提高点焊质量监控精度的有效途径;(3)点焊过程监测信息与质量参数之间关系也含有较大的非线性,用线性模型去描述这样的关系将导致模型误差的增加。

     

    Abstract: By means of linear regression analysis technique,the single regression model and multiple regression model between dynamic resistance information in spot welding process of mild steel and quality parameters are studied systematically.Results show as follows: (1) The total linear relations of all models are obvious by statistic test,which shows that the regression models can be used for monitoring and controlling spot weld quality.(2) The determination coefficients of the single regression model are only from 0.15 to 0.33,and the determination coefficients of the multiple regression model can be raised from 0.59 to 0.79,which show that developing multi-variable synthetical monitor is an effective way to raise monitoring accuracy.(3) The relations between monitoring information and quality factors are substantially non-linear,it will make model more error to describe such relations by linear model.

     

/

返回文章
返回