Development of a soft sensor for determining the cooling rate for press hardening of medium manganese steel

Press-hardened ultra-high strength steel parts are extensively used in the automotive industry for their lightweight and safety benefits. Medium Manganese Steels offer further advantages due to their high strength and ductility, achieved through a multi-phase microstructure at lower annealing temperatures. However, their industrial application faces challenges due to stricter process control requirements. Data-driven approaches based on process parameters hold promise for advancing material-specific process development. This study establishes a digital material shadow to predict the influence of process parameters on final material mechanical properties. For this aim, a soft sensor is introduced to determine the crucial parameters, particularly cooling rate and forming temperature during press hardening, leveraging the measurement of tool temperature data. Linear regression is applied to correlate tool temperature increases with cooling rates and final temperatures. Comparisons of the soft sensor predictions with finite element simulations demonstrate high accuracy, with deviations below 1 %. The proposed soft sensor enhances the precision of data-driven predictions for final material properties, supporting the integration of Medium Manganese Steels into press-hardened applications while enabling sustainable manufacturing practices.