基于KPCA-IDBO-LSSVM预测转炉终点磷含量  

Prediction of terminal phosphorus contentin converter based on KPCA-IDBO-LSSVM

作　　者：牛玉梁 李爱莲[1] 解韶峰 Niu Yuliang;Li Ailian;Xie Shaofeng(Inner Mongolia University of Science and Technology)

机构地区：[1]内蒙古科技大学自动化与电气工程学院 [2]内蒙古科技大学基建处

出　　处：《冶金能源》2024年第3期55-58,64,共5页Energy For Metallurgical Industry

基　　金：内蒙古自治区自然科学基金(2022MS06003)。

摘　　要：为实现转炉终点钢水磷含量的精准预报,首先采用核主成分分析(KPCA)对数据进行降维处理,再提出改进的蜣螂优化算法(IDBO)来优化最小二乘支持向量机(LSSVM),最后建立KPCA-IDBO-LSSVM预测模型来进行转炉终点钢水磷含量预测。将KPCA-IDBO-LSSVM终点磷含量预测结果与LSSVM、IDBO-LSSVM以及多种其他模型进行对比,结果表明,KPCA与IDBO的加入均明显地提升了预测效果,KPCA-IDBO-LSSVM的终点磷含量预测误差在±0.003%内的命中率达到了90%,为冶炼带来了具有实际意义的帮助。In order to achieve accurate prediction of phosphorus content in molten steel at the end point of converter,kernel principal component analysis(KPCA)was used to reduce dimensionality of data,and then an improved dung beetle optimization(IDBO)algorithm was proposed to optimize least square support vector machine(LSSVM).Finally,KPCA-IDBO-LSSVM prediction model is established to predict the phosphorus content of steel at the end of converter.The prediction results of terminal phosphorus content of KPCA-IDBO-LSSVM were compared with those of LSSVM,IDBO-LSSVM and other models.The results showed that the addition of KPCA and IDBO significantly improved the prediction effect.The prediction error of terminal phosphorus content of KPCA-IDBO-LSSVM reaches 90%within±0.003%,which brings practical help for smelting.

关 键 词：转炉炼钢 磷含量预测 蜣螂算法 核主成分分析 

分 类 号：TF713[冶金工程—钢铁冶金] TP18[自动化与计算机技术—控制理论与控制工程]