Prediction of grain refinement using multiscale crystal plasticity-integrated dislocation density-based model in multiphase steel alloys

This study investigates an approach to simulating grain refinement during macroscopic deformation considering microstructural evolution in each constituent phase within the steel using a physics -based model integrating crystal plasticity with a dislocation density -based model. The constitutive model for each constituent phase is first predicted using crystal plasticity and validated. Based on the crystal plasticity data, the parameters of the dislocation density model are determined to define microstructure -sensitive material plasticity behavior and to store evolving microstructure as state variables in the simulation of large structures with heterogeneous microstructure. The validity of the numerical approach is evaluated through simulations at room temperature and elevated temperatures for steels of varying phase percentages with experimental results. It is shown that the proposed approach well captures the mechanical behavior during tensile tests and the final grain size of the Equal Channel Angular Pressed steels with high accuracy (average error = 5.91 %) in cases of varying initial microstructures. This proposed approach could be a powerful tool to explore and predict the mechanical properties of novel multiphase materials with reduced experimentation costs.