基于DEI-RBF算法的电脉冲热轧机轧制力预测  

Rolling Force Prediction of Electric Pulse Hot Rolling Mill Based on DEI-RBF Algorithm

作　　者：李静[1] Li Jing(College of Intelligent Manufacturing,Xinxiang Vocational and Technical College,Xinxiang Henan 453006,China)

机构地区：[1]新乡职业技术学院智能制造学院,河南新乡453006

出　　处：《山西冶金》2025年第1期88-90,共3页Shanxi Metallurgy

摘　　要：电脉冲热轧机的轧制力预测精度直接影响到轧制设备的运行质量。设计了一种通过差分进化改进支持向量机模型(DEI-RBF),以RBF核函数支持向量机构建初始模型。研究结果表明,逐渐提高核函数,测试集拟合性能下降,而训练集的拟合能力提升。加入差分进化算法后,实现了支持向量机回归模型性能的显著提升,获得了比传统轧制力模型更准确的预测结果,可有效指导生产过程。该研究有助于提高锻压设备的机电控制效果。The rolling force prediction accuracy of electric pulse hot rolling mill directly affects the running quality of rolling equipment.An improved differential evolution support vector machine(DEI-RBF)model is designed,and the initial model is built with the support vector mechanism of RBF kernel function.The results show that with the gradual increase of kernel function,the test set fitting performance decreases.The fitting ability of training set is improved.After adding differential evolution algorithm,the performance of support vector machine regression model is significantly improved.More accurate prediction results are obtained than the traditional rolling force model,which can effectively guide the production process.The research is helpful to provide the mechanical and electrical control effect of forging equipment.

关 键 词：电脉冲热轧 轧制力预测 支持向量机 差分进化 

分 类 号：TG333[金属学及工艺—金属压力加工]