Predicting melt track geometry and part density in laser powder bed fusion of metals using machine learning

Abstract Laser powder bed fusion of metals (PBF-LB/M) is a process widely used in additive manufacturing (AM). It highly sensitive to its parameters directly determining the quality components. Hence, optimal are needed ensure highest part quality. However, current approaches such as experimental investigation and numerical simulation time-consuming costly, requiring more efficient ways for parameter optimization. In this work, use machine learning (ML) search investigated based on influence laser power speed simulated melt pool dimensions experimentally determined density. total, four algorithms considered. The models trained predict size density parameters. accuracy evaluated deviation prediction from actual value. implemented python using scikit-learn library. results show that ML provide generalized predictions with small errors both density, demonstrating potential AM. main limitation data collection, which still done or simulatively. provides an opportunity optimization PBF-LB/M.