The temperature evolution, stress evolution, and dendrite growth during the laser metal deposition (LMD) process of Inconel 718 alloy were simulated using a multi-scale model that combines the finite element method and the Cellular Automata (CA) method. This model established the relationship between the thermal behavior of the molten pool and the solidification characteristics, and predicted the morphological size of the solidified structure. Research indicates that under conditions of higher cooling rate and lower shape control factor, equiaxed grains are prone to form, with a lower degree of elemental segregation in this region, which is conducive to the precipitation of granular Laves phase. Conversely, under conditions of lower cooling rate and higher shape control factor, columnar grains are more likely to form, with a higher degree of elemental segregation in this region, leading to the precipitation of chain-like Laves phase. The results of thermo-mechanical coupled simulation suggest that the maximum transient thermal stress is located at the overlap position, and when the component cools down to room temperature, the maximum normal stress along the scanning direction is observed. Dendritic growth results demonstrate that during the equiaxed grain growth process in the top region of the melt pool, the released solute is constrained by time and space, leading to the accumulation of solute and the formation of interdendritic segregation. In areas with a higher grain density, the solute accumulation zones overlap, resulting in an increase in solute concentration, which inhibits dendrite growth. In the middle region of the melt pool, during the columnar grain growth process, the solute concentration in the gaps between adjacent grains continuously increases, resulting in micro-segregation phenomena. Experimental measurements have shown that the temperature field's melt pool morphology, stress values along the scanning path, and dendritic morphology are in good agreement with the experimental outcomes.
Read full abstract