An extended cellular automaton finite volume method for grain nucleation–growth–coarsening during the wire-based additive manufacturing process

Among various metal additive manufacturing (AM) processes, the wire-based AM process has gained increasing attention in heavy industries due to its potential to overcome size and weight limitations. However, this process commonly results in increased microstructural heterogeneity and significant grain coarsening in the heat-affected zone (HAZ), necessitating appropriate solutions. This paper proposes an extended cellular automaton finite volume method (xCAFVM) for predicting the melt pool flow and grain structure evolution in Ti6Al4V during the wire-based AM process. In xCAFVM, the cellular automaton model for grain evolution within the mushy zone is extended to predict the grain coarsening in the HAZ by coupling it with an improved Monte Carlo (MC) model. The improved MC model considers the influence of grain crystallographic orientation on grain coarsening and the temperature-dependent grain growth rate in the cell selection probability. A two-way coupling scheme is proposed for the xCA, with the CA model providing the grain crystallographic orientations for the improved MC model and the latter determining the accurate size of partially melted grains for epitaxial grain growth in the mushy zone. A one-way coupling scheme is applied to integrate the grain structure prediction method and the finite volume method for the heat and fluid flow in the process. A set of numerical examples is presented, and the simulation results are in good agreement with experimental data from the literature. It is identified that grain coarsening plays a crucial role in determining grain size and shape. The proposed method could be a powerful tool for gaining insights into the relationship between the process and the microstructure and guiding the parameter optimization of the wire-based AM.