SIM-AM 2025

Keynote

Bridging simulation and experimental research: An easy-to-use, open-source PBF-LB/M simulation library

  • Kopp, Philipp (Bauhaus-Universität Weimar)
  • Carraturo, Massimo (University of Pavia)
  • Holla, Vijaya (Technical University of Munich)
  • Kollmannsberger, Stefan (Bauhaus-Universität Weimar)

Please login to view abstract download link

Conduction-based thermal models are useful to experimentalists as they can predict the general characteristics of the temperature evolution for process states without excessive evaporation. This includes narrowing the range of feasible process parameters before conducting the first experiments, transferringprocess parameters to a similar setup (e.g. changing the material), or manual and algorithmic optimization of process parameters [1]. Although thermal simulations are quite simple (especially compared to more complicated multi-physics models), they are not widely accessible to experimentalists. Some contributing factors are that despite their simplicity, thermal models require careful calibration and are only reasonably accurate when the quantities of interest are primarily driven by heat conduction. The numerical approximation using (for example) finite elements in space and a finite-difference-based time-stepping scheme adds further complexity for a user unfamiliar with such methods. Finally, research codes are often complicated to set up and use, requiring the installation of build tools for source code compilation with dependencies on several third-party libraries. To bridge the gap between simulation experts and experimentalists, we are developing a code framework that packages our experience in thermal and thermomechanical PBF-LB/M simulations into an easy-to-use Python library. The goal of this project is to reduce the complexity of PBF-LB/M simulations as much as possible, enabling their use by people who are not experts in numerical simulations. This talk will outline the project philosophy, show some examples, and present recent results of experimentally validating the presented models [2]. We also discuss indicators for assessing the model error (as feedback to the user) and the discretization error (to drive mesh adaptivity). REFERENCES [1] V. Holla, P. Kopp, J. Gr¨unewald, K. Wudy, and S. Kollmannsberger: Laser beam shape optimization in powder bed fusion of metals. Additive Manufacturing. (2023) 72:103609. https://doi.org/10.1016/j.addma.2023.103609. [2] V. Holla, J. Gr¨unewald, P. Kopp, P. Praegla, C. Meier, K. Wudy, and S. Kollmannsberger: Validity of Thermal Simulation Models for Different Laser Beam Shapes in Bead-on-Plate Melting. Integrating Materials and Manufacturing Innovation (2024) 13:969-985. https://doi.org/10.1007/s40192-024-00382-2.