Data-driven prediction and control of roll marks in galvanizing line temper mills
YANG Yu
ZHANG Kangwu
HUANG Guan
QU Ge
QIN Luyu
XU Zhaoyang
XU Yanghuan
Abstract：To address the issue of SKP roll mark defects in the galvanizing line,this paper proposes a collaborative control method that deeply integrates data-driven approaches with process optimization.First,big data analysis was employed to quantify the correlation between prolonged production runs of exterior panels and the occurrence of roll marks.Subsequently,a deep neural network(DNN)-based model was developed to predict roll surface condition,enabling accurate early warnings of potential defects.Furthermore,an intelligent production rhythm optimization strategy,termed"periodic roll surface refreshing,"was introduced.A refined lifecycle management system for work rolls was also established,incorporating high-resolution scanning and progressive loading protocols.Application results demonstrate that the proposed method significantly reduces the rate of abnormal roll changes,thereby ensuring the stable production of high-grade exterior panels.This research presents a paradigm shift for quality control in rolling processes,moving from an"experience-driven"to a"data-and-intelligence-driven"approach.
Keywords：galvanizing linetemper millroll markssurface qualityabnormal roll change rate
Publication Date：2025-11-20
Online Publishing Date：2026-04-01(First online date of this platform, not the publication date of the document)
Pages：6( 22-27 )