基于偏最小二乘回归模型的带钢热镀锌质量监控方法
Quality monitoring method of strip hot-dip galvanizing based on partial least squares regression
-
摘要: 提出了基于偏最小二乘回归模型的带钢热镀锌质量监控方法.以带钢热镀锌生产中带钢力学性能和锌层质量的质量监控为研究对象,用偏最小二乘方法建立了生产过程参数与质量结果之间的回归模型,对生产过程控制能力进行了分析,并给出了产品质量的预测方法.用鞍钢股份有限公司带钢热镀锌的实际生产数据进行验证.结果表明,偏最小二乘法比传统的多元线性回归方法具有更好的预测精度,基于偏最小二乘回归的锌层质量预测模型,其相对预测误差可达到5.93%.Abstract: A quality monitoring method for strip hot-dip galvanizing based on partial least square regression was proposed. Taking the quality monitoring of mechanical properties and zinc coating mass in strip hot-dip galvanizing as the investigated subject, a regression model between process parameters and quality results was constructed through partial least square method. With the regression model, the capability of production process control was analyzed and a production quality prediction method was presented, Real field data from strip hot-dip galvanizing production in Angang Steel Company Limited were used for validation, The results show that partial least square regression has a better predicting precision than traditional multiple linear regression, and that the zinc coating mass prediction model based on partial least square regression has a relative prediction error of 5.93%.