High-throughput numerical exploration of preheating and sintering in electron beam powder bed fusion

In Electron Beam Powder Bed Fusion, the preheating step enables the processing of non-weldable materials by reducing thermal stresses. Additionally, the ongoing sintering creates a mechanically stable and conductive powder bed, preventing the formation of smoke and providing a beneficial basis for subsequent melting. The electron beam itself serves as an ideal heating device due to its high power and fast deflection velocities. The beam movement is typically performed according to the line order deflection pattern, with the aim of efficient, homogeneous and smoke-free heating and sintering. Given the multi-dimensionality of the parameters defining the preheating step and the deflection pattern, it is difficult to estimate the individual effects and dependencies, thus making it challenging to obtain the optimal setting. In this study, this issue is addressed by employing a numerical high-throughput approach to systematically explore how individual variations within the multi-dimensional parameter space influence the target quantities temperature and sinter degree. A total number of over 6000 relevant parameter combinations are evaluated to gather a comprehensive overview of the individual effects of the quantities line length, line offset, line order, beam focus, velocity and power. The results are presented in feature maps, facilitating clear identification of relationships and particular influences. Given the high fraction that preheating can take of the total layer time, the acquired understanding will have a significant impact on the efficiency of the process and the parameter selection. • High-throughput approach systematically exploring over 6000 parameter combinations. • Deployment of a highly GPU-parallelized model on a high-performance computing cluster. • Covering the effects of the preheating step on both temperature and sinter degree. • Creation of comprehensive process maps of the multi-dimensional parameter space. • Formulation of practical guidelines for optimal process parameter selection.