Level-Set Method Based Predictive Modeling to Track the Evolution of Surface During Pulsed Laser Surface Melting

Abstract

The objective of this study is to investigate the evolution of surface geometry during pulsed laser surface melting (pLSM) via level-set method-based interface tracking numerical framework. Existing models to track surface geometry are inaccurate and computationally expensive. Therefore, they have limited use in gaining understanding of the surface evolution during pLSM. A numerical model, integrating the level-set approach, fluid flow, and heat transfer dynamics, is detailed in this paper. The multi-phase numerical model achieves accurate tracking of interface for a single pulse by implementing the volumetric laser heat source on the moving interface by modifying Beer–Lambert's law. The accuracy of the single pulse model is confirmed by comparing its peak-to-valley height (PVH) to the experimental data. The deviation in PVH is limited to about 15%, with a maximum root mean square error of ∼0.24 µm, highlighting the model's reliability. Additionally, the evolved surface of a single pulse from the model is replicated over an area with dedicated overlaps to generate the predicted textured surface with reasonable accuracy. Some inaccuracies in the predicted surface roughness values were observed because the textures were generated based on a single pulse geometry computed on an initially flat surface. Nonetheless, the results highlight a significant development in numerical frameworks for pLSM and can be used as a tool to gain deeper insights into the process and for process optimization.