基于热轧工艺过程数据的硅钢同板差预测模型  

Same Plate Difference Forecasts Model for Silicon Steel Based on Hot Rolling Process Parameters

作　　者：成印明[1] CHENG Yinming(Maanshan Iron and Steel Co.,Ltd.,Maanshan 243000,China)

机构地区：[1]马鞍山钢铁股份有限公司,安徽马鞍山243000

出　　处：《山东冶金》2020年第3期36-40,共5页Shandong Metallurgy

摘　　要：为建立冷轧硅钢同板差预测模型,运用岭回归分析法对同板差影响因素进行分析,验证热轧断面参数到冷轧同板差的遗传性,并得到各特征参数的特征重要度.根据特征参数与热轧板形数据,以冷轧同板差为目标,利用BP神经网络算法建立热轧断面到冷轧同板差的遗传模型,并对冷热轧生产中各工艺特征的特征参数进行重要度关联分析,量化各工艺特征参数对带钢冷轧同板差指标的重要程度;并依据重要度参数进行模型特征选择,从而建立带钢同板差预报模型.最终将预报模型应用于工业现场,通过热轧板形数据,定量的来预测冷轧成品质量,获得较好的应用效果.In order to establish the prediction model of the same plate difference of cold-rolled silicon steel,the influence factors on the same plate difference are parameters analyzed by ridge regression analysis in this paper.The heredity of the hot rolling section parameters to the same plate difference of cold-rolled silicon steel is verified,and the characteristic importance of each characteristic parameter is obtained.According to the characteristic parameters and hot rolling shape data,aiming at the same plate difference of cold rolling,the genetic model from the hot rolling section to the same plate difference of cold rolling is established by using BP neural network algorithm,and the correlation analysis of the characteristic parameters of each process characteristic in the cold rolling production is carried out to quantify the importance of each process characteristic parameter to the same plate difference index of cold rolling;and the model is carried out according to the important parameters feature selection,so as to establish the prediction model of the same plate difference.Finally,the prediction model is applied to the industrial field,through the hot rolling shape data,the quality of cold-rolled products can be predicted quantitatively,and good application results can be obtained.

关 键 词：冷轧同板差 热轧板形 重要度 BP神经网络算法 预测模型 

分 类 号：TG337.3[金属学及工艺—金属压力加工]