M2P 2023

PADME-AM: Partition of Unity Methods for Additive Manufacturing

  • Elsweijer, Sandro (German Aerospace Center, University of Bonn)
  • Holke, Johannes (German Aerospace Center)
  • Ponnusamy, Prasanna (German Aerospace Center)
  • Schweitzer, Marc Alexander (University of Bonn, Fraunhofer SCAI)
  • Troska, Lukas (University of Bonn, Fraunhofer SCAI)

Please login to view abstract download link

This research abstract focuses on the rapid growth of additive manufacturing (AM), particularly in the metal AM sector in recent years. Laser powder bed fusion (L-PBF) manufacturing processes offer exceptional options for high-quality, complex parts in industries such as medicine and aerospace, due to their ability to realize complex parts in highly integrated designs while implementing lightweight concepts. Equipment manufacturers have long recognized that expanding the material portfolio, increasing plant capacity, and improving the quality of components can bring clear competitive advantages. To achieve these goals, new and efficient simulation methods need to be created to consider the complex physical processes of interaction between the laser, powder, and solid, considering different material states, phase transitions, and interfaces. We focus on the challenges of simulating the meso- and macroscopic effects of the L-PBF process. One of the major challenges is accurately capturing the thermal-fluid dynamics and the associated changes in material properties at the meso-scale, which is the level of the powder particles. Another challenge is the simulation of the macroscopic effects, such as the thermal distortion and residual stresses that occur at the component scale. Therefore, it is necessary to couple the meso-scale phenomena with the macroscopic phenomena, which is a major challenge in the field of AM. Currently, there is no software environment that economically enables a multiscale simulation of the process. In PADME-AM, we will therefore provide an easy and efficient method for thermo-mechanical computation on a component scale, combined with high-resolution thermal analysis in areas of particular interest. This will be implemented with the meshfree Partition of Unity Method (PUM), a generalization of the Finite Element Method, which allows users without extensive simulation expertise to solve multi-scale partial differential equations on complex geometries, as no meshing is necessary. Through the integration of advanced numerical methods and our proposed partition of unity approach, PADME-AM aims to significantly advance the simulation of L-PBF processes and pave the way for its widespread industrial implementation.