Mechanism and Data‐Driven Prediction of Warpage Behavior in High‐Strength Thin Strip in S6‐High Cold Rolling Mill

Warpage caused by uneven stress distribution is a critical issue limiting shape quality in rolling high‐strength thin strips with an S6‐high cold rolling mill. Owing to high yield strength and small thickness, such strips are sensitive to strain heterogeneity, while conventional methods cannot reveal the complex warpage mechanism. To address this, an elastoplastic finite element (FE) model of the mill‐strip system is developed and validated. Mechanistic analysis clarifies the effects of material and process parameters. A simulation database is established, and a synthetic warpage factor is proposed to quantify global and local trends. An intelligent prediction model is built using the White Shark Optimizer (WSO) optimized Light Gradient Boosting Machine (LightGBM). Results demonstrate that strip thickness and yield strength are intrinsic sources of warpage sensitivity, while work roll misalignment and side support roll position are key inducing factors. Compared with conventional methods, WSO‐LightGBM achieves superior accuracy ( R 2 = 0.963, RMSE = 0.00109 mm, MAE = 0.00191 mm). This study confirms the applicability of mechanism‐data fusion for accurate prediction and process optimization.