基于LWOA-εTSVR的转炉炼钢终点静态预测模型  

Static prediction model for end-point of BOF steelmaking based on LWOA-εTSVR

作　　者：汪淼 李胜利[2] 艾新港[2] 杨永辉[1] 高闯[1] WANG Miao;LI Shengli;AI Xingang;YANG Yonghui;GAO Chuang(School of Electronic and Information Engineering,University of Science and Technology Liaoning,Anshan 114051,China;School of Materials and Metallurgy,University of Science and Technology Liaoning,Anshan 114051,China)

机构地区：[1]辽宁科技大学电子信息与工程学院,辽宁鞍山114051 [2]辽宁科技大学材料与冶金学院,辽宁鞍山114051

出　　处：《冶金自动化》2025年第2期53-63,共11页Metallurgical Industry Automation

基　　金：十四五国家重点研发计划项目(2021YFB3702005);国家自然科学基金项目(52304352);辽宁省创新联合体科技重大专项(2023 JH1/11200012);辽宁省教育厅青年项目(JYTQN2023244);辽宁省高校基本科研业务费专项资金(LJ232410146036)。

摘　　要：碱性氧气转炉(basic oxygen furnace,BOF)的终点碳含量和温度是确保炼钢生产平稳运行的关键因素。故提出了一种新的基于莱维飞行鲸鱼优化算法(lévy-flying algorithm,LWOA)和ε-孪生支持向量回归机(ε-twin support vector machine for regression,εTSVR)终点预测模型。先采用箱线图对数据进行筛选,随后,利用冶金机理和Spearman相关分析,识别影响终点碳含量和温度的关键因素。最后,利用LWOA具有调节参数简单和收敛速度快的特点,对εTSVR算法中的参数进行自动优化。仿真结果显示,所提出的LWOA-εTSVR预测模型在终点碳含量和温度满足误差容限分别为±0.005%和±10℃下,命中率分别为89%和93%。同时,双命中率达到了83%。与其他四种预测模型相较而言,所提出的LWOA-εTSVR预测模型展现出了更为良好的优势。此外,通过设置不同的误差区间,也验证了所提出的预测模型性能的可靠性。并且所提出的预测模型比实际钢厂工艺具有更高的预测精度,为钢铁企业提供了有力的技术支持。The end-point carbon content and temperature of the basic oxygen furnace(BOF)are the crucial factors for ensuring the seamless operation of steelmaking production.Consequently,a novel end-point prediction model based on the Lévy flights whale optimization algorithm(LWOA)andε-twin support vector regression(εTSVR)has been established.Firstly,the box plot was employed to filter the data.Subsequently,the key factors influencing the end-point carbon content and temperature were identified through metallurgical mechanism and spearman correlation analysis.Finally,the parameters in theεTSVR algorithm were automatically optimized by utilizing LWOA,which possesses the characteristics of simple adjustment parameters and rapid convergence speed.The simulation results indicate that the proposed LWOA-εTSVR prediction model have hit rates of 89%and 93%at the end-point carbon content and temperature satisfying the error tolerance of±0.005%and±10℃,respectively.Meanwhile,the double hit rate reaches 83%.Compared with the other three prediction models,the proposed LWOA-εTSVR model demonstrates superior advantages.Furthermore,by setting different error intervals,the reliability of the performance of the proposed prediction model was also verified.Moreover,the proposed prediction model has higher prediction accuracy than the actual steel plant process,providing robust technical support for steel enterprises.

关 键 词：转炉炼钢 预报模型 Spearman相关分析 LWOA εTSVR 

分 类 号：TF713.3[冶金工程—钢铁冶金]