Experimental and numerical simulation analysis on establishment of a control model for rolled flatness

With the increasing demand for higher-quality flatness in downstream industries, the optimization of rolling processes and parameters has become a critical area of research. The effects of rolling force and front tension adjustments on flatness were examined systematically under various rolling process conditions. By embedding the Johnson–Cook constitutive model into the ABAQUS simulation platform through a user-defined subroutine, a series of three-dimensional finite element models for different rolling scenarios were developed. Simulation results indicate that, under all four rolling process conditions, edge strain consistently exceeds center strain, with forward-driven rolling exhibiting greater edge strain than reverse-driven rolling. Along the strip thickness direction, reverse-driven rolling results in higher strain compared to forward-driven rolling. Moreover, in single roll driven rolling, the upper surface of the strip experiences higher strain than the lower surface, while the reverse trend is observed in double roll driven rolling. As the rolling force increases from 1000 to 5000 kN, the strain difference in the width and thickness directions of the strip varies significantly under double roll driven rolling and double roll reverse-driven rolling, with change slopes of 5.74 × 10⁻⁶ and −2.85 × 10⁻⁶, respectively. Double roll driven rolling effectively prevents the deterioration of flatness along the rolling direction. Furthermore, as the front tension increases from 60 to 100 MPa, double roll reverse-driven rolling significantly suppresses strain differentials in the width, thickness, and rolling directions, with change slopes of −6.73 × 10⁻⁴, 1.22 × 10⁻⁵, and −1.29 × 10⁻⁵, respectively. Eventually, a predictive model is established, integrating rolling process, rolling force, and front tension, thereby providing a theoretical framework for advancing the precision and efficiency of strip rolling processes.