Optimized shearing strategy for heavy plate based on contour recognition
作者机构：School of Computer Science and EngineeringShenyang Jianzhu UniversityShenyang 110168LiaoningChina Shandong Iron and Steel Co.Ltd.Jinan 271105ShandongChina School of Electrical and Control EngineeringShenyang Jianzhu UniversityShenyang 110168LiaoningChina
出 版 物：《Journal of Iron and Steel Research International》 (国际钢铁研究杂志)
年 卷 期：2023年第30卷第9期
页 面：1821-1833页
核心收录：
学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 081104[工学-模式识别与智能系统] 08[工学] 0835[工学-软件工程] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）]
基 金：The paper was prepared under the support of the Natural Science Foundation of Liaoning Province(Grant No.2022-MS-277) This research was also financially supported by the Youth Project of Foundation of Liaoning Province Education Administration(Grant No.lnqn202016)
主 题：Heavy plate Contour detection Intelligent shearing strategy Length prediction Neural network model
摘 要：The shearing line is the key to improve the quality and efficiency of heavy plates.A model of contour recognition and intelligent shearing strategy for the heavy plate was ***,multi-array binocular vision linear cameras were used to complete the image ***,the total length of the steel plate after cooling was predicted by back propagation neural network algorithm according to the contour ***,using the scanning line and a new camber description method,the shearing strategy including head/tail irregular shape length and rough dividing strategy was *** practical application shows that the model and strategy can effectively solve the problems existing in the shearing process and can effectively improve the yield of steel *** maximum error of detection width,length,camber,and the length of the irregular deformation area at the head/tail of the plate are all less than 5 *** correlation coefficient of the length prediction model based on the back propagation neural network is very *** reverse ratio result of edge cutting failure using the proposed rough dividing strategy is 1/401=0.2%,which is 2%higher than that by human.
维普期刊数据库