Adiabatic efficiency and mechanical properties optimization for the laser-based powder bed fusion process of 316L stainless steel alloy

Abstract

This study presents a multi-objective optimization framework for the laser-based powder bed fusion process (L-PBF) of 316L stainless steel, focusing on enhancing adiabatic efficiency and mechanical performance while minimizing surface roughness. A data-driven approach, integrating machine learning models with Non-Dominated Sorting Genetic Algorithm II (NSGA-II), was employed to identify optimal printing conditions. The framework considers key process parameters to balance conflicting objectives, including laser power, scanning speed, hatch space, and layer thickness. The experimental validation of the optimal parameters selected demonstrated a high correlation between the predicted and measured values, with minor deviations attributed to manual handling, sample preparation, and measurement uncertainties. The results confirm the robustness and reliability of the proposed optimization methodology, providing valuable insights for improving the process efficiency and quality of the component in additive manufacturing. The findings contribute to the advancement of L-PBF technology, offering a scalable and adaptable optimization strategy that is applicable to various industrial applications.