In this study, we develop a microstructure prediction model that bridges a multi-phase-field method incorporating a nonlinear preconditioning with the temperature field obtained from thermal-fluid simulations, in order to accurately predict microstructural changes under various scanning strategies in metal additive manufacturing. The introduction of the nonlinear preconditioning improves the anisotropy of the grid and enables application to a wide range of scanning strategies. Furthermore, coupling with thermal-fluid simulations allows accurate reproduction of the complex temperature distribution induced by the laser heat source. Using this method, we perform high-accuracy microstructure prediction in metal additive manufacturing.
Takahashi et al. (2025) studied this question.