摘要
为快速预测双电极电弧炉在不同电极高度、间距、电压下的温度,基于电弧炉理论建立仿真模型,获得输入样本与响应之间的大量仿真数据。在此基础上,以确定性系数R2为主要评价指标,应用Kriging、Rbf、Svm代理模型进行拟合预测,得出Kriging代理模型的预测精度最高,通过加点优化得到精度更高的kriging代理模型。
In order to quickly predict the temperature of double-electrode electric arc furnace under different electrode heights,spacings and voltages,a simulation model is established based on the theory of electric arc furnace to obtain a large amount of simulation data between input samples and responses.On this basis,the kriging,Rbf and Svm agent models are applied for fitting prediction while taking the certainty coefficient R2 as the main evaluation index,it is found that kriging surrogate model has the highest prediction accuracy,and a kriging surrogate model with higher accuracy is obtained by adding points for optimization.
作者
高茹月
关丽荣
刘博林
Gao Ruyue;Guan Lirong;Liu Bolin
出处
《一重技术》
2024年第5期56-59,71,共5页
CFHI Technology
基金
辽宁省教育厅高校科研基金重点攻关项目(项目编号:LJKZZ20220037)。
关键词
kriging代理模型
电弧炉
温度预测
预测精度
Kriging surrogate model
electric arc furnace
temperature prediction
prediction accuracy
