锻压技术

基于随机森林的热轧带钢终轧温度预报

英文标题：Prediction on finish rolling temperature for hot-rolled strip based on random forest

作者：马更生1 樊涛2 马晓宝3 路平1 李炯1 陆伟1 韩东序1

单位：1. 常州工业职业技术学院机械与交通学院 2. 中国船舶重工集团应急预警与救援装备股份有限公司 3. 太原理工大学机械工程学院

分类号：TG335.5

出版年,卷(期):页码：2025,50(1):134-139

摘要：

复杂的精轧区边界条件和难以观测的参数导致传统在线带钢热终轧温度的预报精度受限。为了提高终轧温度预报精度，采用随机森林进行数据驱动方式的建模。选取影响终轧温度的43个特征因子作为数据驱动终轧温度预测模型的输入值，采用NCL和SMOTE混合算法处理换规格等情况的非平衡数据集，决策树的随机特征选取包括与目标变量高、低相关的特征。结果表明：构建的热轧带钢终轧温度随机森林预报模型在测试集上预测值的最大误差在15 ℃以内，具有较好的回归效果和泛化能力，满足热轧现场带钢终轧温度预报精度的要求。

The complex boundary conditions and difficulty in parameter prediction in the precision rolling zone limit the accuracy of traditional online hot rolling temperature prediction for strip steel. Therefore, in order to improve the accuracy of final rolling temperature prediction, a data-driven modeling approach using random forest was adopted, and taking forty-three characteristic factors that affect the final rolling temperature as the input values for the data-driven final rolling temperature prediction model. Then, the imbalanced datasets such as changing specifications were processed by a hybrid algorithm of NCL and SMOTE, and the random feature selection of the decision tree included features that were highly or lowly correlated with the target variable. The results show that the constructed random forest prediction model for the final rolling temperature of hot-rolled strip has a maximum prediction value error of within 15 ℃ on the test set and good regression effect and generalization ability, meeting the accuracy requirements for the final rolling temperature prediction of hot-rolled strip on site.

基金项目：

国家自然科学基金资助项目（52305405）;海安太原理工大学先进制造与智能装备产业研究院开放研发项目（2023HA-TYUTKFYF011）；常州工业职业技术学院高层次人才项目（GCC202413101006）

作者简介：马更生（1984-），男，博士，讲师 E-mail：gengshengma@163.com

参考文献：

[1] 龚殿尧, 徐建忠, 刘相华, 等. 热连轧带钢终轧温度控制样本跟踪策略[J]. 东北大学学报(自然科学版), 2006, 27(8):883-886.

Gong D Y, Xu J Z, Liu X H, et al. Sample tracking strategy for finish rolling temperature control of hot-rolled strip[J]. Journal of Northeastern University (Natural Science), 2006, 27(8): 883-886.

[2] 周进, 沈丙振, 韩志强, 等. 精轧区热轧带钢温度场的数值模拟[J]. 钢铁研究学报, 2003, 15(2): 14-18.

Zhou J, Shen B Z, Han Z Q, et al. Numerical simulation on temperature field of hot strip in finish rolling process[J]. Journal of Iron and Steel Research, 2003, 15(2): 14-18.

[3] 刘玠, 杨卫东, 刘文仲. 热轧生产自动化技术[M]. 2版. 北京: 冶金工业出版社, 2017.

Liu J, Yang W D, Liu W Z. Automation Technology for Hot Rolling Production[M]. 2nd Edition. Beijing: Metallurgical Industry Press, 2017.

[4] 龚殿尧, 徐建忠, 薛文颖, 等. 热连轧带钢终轧温度预报模拟软件开发[J]. 钢铁研究学报, 2007(1): 60-62.

Gong D Y, Xu J Z, Xue W Y, et al. A simulation software for forecasting finish rolling temperature for hot strip mill[J]. Journal of Iron and Steel Research, 2007(1): 60-62.

[5] Laasraoui A, Jonas J J. Prediction of temperature distribution, flow stress and microstructure during the multipass hot rolling of steel plate and strip[J]. ISIJ International, 1991, 31(1): 95-105.

[6] Lee J H, Kwak W J, Sun C G, et al. Precision online model for prediction of strip temperature in hot strip rolling[J]. Ironmaking & Steelmaking, 2004, 31(2): 153-168.

[7] 邬晓方, 陈剑, 钱涛. 基于大数据及人工智能技术的电力行业技能等级评价方法[J]. 自动化技术与应用, 2024, 43(3): 165-168.

Wu X F, Chen J, Qian T. Skill level evaluation method of power industry based on big data and artificial intelligence technology[J]. Techniques of Automation and Applications, 2024, 43(3): 165-168.

[8] Breiman L. Random forests[J]. Machine Learning, 2001, 45(1): 5-32.

[9] Svetnik V, Liaw A, Tong C, et a1. Random forest: A classification and regression tool for compound classification and QSAR modeling[J]. Journal of Chemical Information ＆ Computer Sciences, 2003, 43(6): 19-47.

[10]陶新民, 刘福荣, 杜宝祥. 不均衡数据SVM分类算法及其应用[M]. 哈尔滨: 黑龙江科学技术出版社, 2011.

Tao X M, Liu F R, Du B X. SVM Classification Algorithm and Application for Imbalanced Data[M]. Harbin: Heilongjiang Science and Technology Press, 2011.

[11]王青天, 孔越. Python金融大数据风控建模实战: 基于机器学习[M]. 北京: 机械工业出版社, 2020.

Wang Q T, Kong Y. Modeling Practical of Python Financial Big Data Risk Control: Based on Machine Learning[M]. Beijing: China Machine Press, 2020.

[12]梅子行. 智能风控：原理、算法与工程实践[M]. 北京: 机械工业出版社, 2020.

Mei Z X. Intelligent Risk Control：Principles, Algorithms and Practice[M]. Beijing: China Machine Press, 2020.

[13]Chawla N V, Bowyer K W, Hall L O, et al. SMOTE: Synthetic minority over-sampling technique[J]. Journal of Artificial Intelligence Research, 2002, 16(1): 321-357.

[14]纪英俊, 勇晓玥, 刘英林, 等. 基于随机森林的热轧带钢质量分析与预测方法[J]. 东北大学学报（自然科学版）, 2019, 40(1): 11-15.

Ji Y J, Yong X Y, Liu Y L, et al. Random forest based quality analysis and prediction method for hot-rolled strip[J]. Journal of Northeastern University (Natural Science), 2019, 40(1): 11-15.

服务与反馈：

【文章下载】【加入收藏】