论文部分内容阅读
以热轧实测数据为基础,对不同钢种进行计算并选取基准变形抗力。以周纪华·管克智模型作为目标变形抗力数学模型,采用多元非线性回归的方法完成对碳素钢、合金钢共7个钢种变形抗力模型的优化,并选用特列齐亚科夫·久津模型和井上腾郎模型作为对比进行回归,确定以实测数据为基础的变形抗力模型,且采用对应模型预测除样本数据外钢卷轧制力,证明采用多元非线性回归的方法自学习系数效果较好。
Based on the measured data of hot rolling, different steel grades are calculated and the reference deformation resistance is selected. The model of Zhou Jihua and Guan Kezhi was taken as the mathematical model of target deformation resistance, the deformation resistance model of seven steel grades of carbon steel and alloy steel was optimized by multivariate nonlinear regression method, and the model of Tretyakov Kudzu The upwelling Tenglang model is used as a regression model to determine the deformation resistance model based on the measured data and the corresponding model is used to predict the rolling force of the coil except the sample data. It is proved that the self-learning coefficient is better using the multivariate nonlinear regression method.