An artificial intelligence approach for the in-line evaluation of steels mechanical properties in rolling

The large demand of high-performance steels to improve the safety and energetic performances in automotive is driving increasing need higher uniformity mechanical properties rolled products, both within a single coil batch productions multiple coils. To achieve these targets, we are witnessing an evolution rolling towards introduction sensors Artificial Intelligence (AI) algorithms allow real-time monitoring strip tensile microstructure during annealing process. work described paper part research that aimed at developing AI framework could full control entire mill plant thanks (i) innovative applied different steps processing, (ii) mixed analytical numerical for data computation analyses, (iii) IT infrastructure collect elaborate data. focus on tension-levelling operation which one first performed can have relevant influence characteristics. reach this aim, predictive model applicable process was developed, as used minimize flatness imperfections residual stresses by means plastic deformation. During process, material subjected cyclic tension-compression stresses, obtained with repeated alternate bends under superposed tension. Consequently, desired product standards, terms thickness flatness, achieved, while mechanism deformation changes final material, yield strength decreasing maximum elongation. Finally, predict characteristics, using sampled from placed inside machine.