基于机理和XGBoost算法的LF精炼钢水成分预测模型  

Prediction Model for LF Refined Steel Composition Based on Mechanism and XGBoost Algorithm

作　　者：杨黔 程斯祥 周鹏 彭翼军 彭其春[2,3] 姚建华[4] YANG Qian;CHENG Sixiang;ZHOU Peng;PENG Yijun;PENG Qichun;YAO Jianhua(Hunan VALIN V-CLOUD Technology Co.,Ltd.,Changsha,Hunan 410000,China;The State Key Laboratory of Refractories and Metallurgy,Wuhan University of Science and Technology,Wuhan,Hubei 430081,China;Key Laboratory for Ferrous Metallurgy and Resources Utilization of Ministry of Education,Wuhan University of Science and Technology,Wuhan,Hubei 430081,China;Hunan Valin Xiangtan Iron&Steel Co.,Ltd.,Changsha,Hunan 411100,China)

机构地区：[1]湖南华联云创信息科技有限公司,湖南长沙410000 [2]武汉科技大学省部共建耐火材料与冶金国家重点实验室,湖北武汉430081 [3]武汉科技大学钢铁冶金及资源利用省部共建教育部重点实验室,湖北武汉430081 [4]湖南华菱湘潭钢铁有限公司,湖南长沙411100

出　　处：《自动化应用》2024年第2期5-7,共3页Automation Application

摘　　要：以我国某钢厂120 t LF精炼炉为研究对象,通过建立由冶炼机理模型和XGBoost模型相结合的混合模型,预测LF精炼过程中的钢水成分并进行实际应用。结果表明,模型预测终点碳、硅、锰、铝等元素均处于内控范围内,并平均减少了每炉钢取样工序0.8次,提高了生产效率。Taking 120 t LF refining furnace in a domestic steel mill as the research object,a mixture model combining smelting mechanism model and XGBoost model was established to predict the composition of molten steel in LF refining process,and the actual application was carried out.The results show that the predicted end-point carbon,silicon,manganese,aluminum and other elements are all within the internal control range,and the production of each furnace steel is reduced by 0.8 sampling procedures on average,and the production efficiency is improved.

关 键 词：LF精炼钢 成分预测 机理模型 XGBoost模型 

分 类 号：TU991.33[建筑科学—市政工程]