Prediction and control for head warpage of hot-rolled slab based on machine vision detection and BP neural network
HUANG Hua-gui1, LIU Ying-gang1, HUANG Hai-lin1, CHEN Guang2, WANG Xiao-dong2
1. National Engineering Research Center for Equipment and Technology of Cold Strip Rolling, Yanshan University, Qinhuangdao 066004, Hebei, China; 2. Technology Center,Shougang Jingtang Iron and Steel United Co., Ltd., Tangshan 063009, Hebei, China
Abstract:In order to solve the problem for automatic control of slab head warpage in the roughing stage of hot rolling, the principle of machine vision inspection and its characterization of inspection position and inspection results on the hot rolling line were introduced taking a 2 250 mm hot rolling mill roughing unit as the object. A 9-10-6-1 four-layer BP neural network prediction model was established by analyzing factors affecting the head warpage of slab, and the network was trained and validated using the field inspection data as samples. The results show that the prediction accuracy of the neural network model meets the control requirements in the field, which lays the foundation for the automatic control of head and tail warpage in the hot rolling process and the unattended production.
黄华贵, 刘迎港, 黄海林, 陈光, 王晓东. 热轧板坯头部翘曲机器视觉检测与BP神经网络预测控制[J]. 中国冶金, 2022, 32(9): 85-89.
HUANG Hua-gui, LIU Ying-gang, HUANG Hai-lin, CHEN Guang, WANG Xiao-dong. Prediction and control for head warpage of hot-rolled slab based on machine vision detection and BP neural network[J]. China Metallurgy, 2022, 32(9): 85-89.