Kármán Conference on Sustainable Computational Science & Engineering 2025
, which took place at the scenic Steinfeld Abbey in North Rhine-Westphalia. It was organized primarly by Julia Kowalski (RWTH Aachen University), and the HPSC Lab was a member of the organizer team and program committee. Our key goal for the conference was to bring together computational scientists from different domains, IT and HPC center representatives, as welll as software developers from industry, to discuss how to merge the three concepts of sustainable software, sustainable computing, and computing for sustainability. We thus had only a limited number of talks and ample times for discussion, putting a strong focus on interactions between the participants. The conference was a great opportunity to meet a lot of experienced and engaged researchers, with plenty of enlightening discussions during session breaks and in the evening.
Together with Erik Faulhaber, Sven Berger, Christian Wei?enfels und Gregor Gassner,?we have submitted our paper "Robust and efficient pre-processing techniques for particle-based methods including dynamic boundary generation".
?
?
?
Abstract
Obtaining high-quality particle distributions for stable and accurate particle-based simulations poses significant challenges, especially for complex geometries. We introduce a preprocessing technique for 2D and 3D geometries, optimized for smoothed particle hydrodynamics (SPH) and other particle-based methods. Our pipeline begins with the generation of a resolution-adaptive point cloud near the geometry's surface employing a face-based neighborhood search. This point cloud forms the basis for a signed distance field, enabling efficient, localized computations near surface regions. To create an initial particle configuration, we apply a hierarchical winding number method for fast and accurate inside-outside segmentation. Particle positions are then relaxed using an SPH-inspired scheme, which also serves to pack boundary particles. This ensures full kernel support and promotes isotropic distributions while preserving the geometry interface. By leveraging the meshless nature of particle-based methods, our approach does not require connectivity information and is thus straightforward to integrate into existing particle-based frameworks. It is robust to imperfect input geometries and memory-efficient without compromising performance. Moreover, our experiments demonstrate that with increasingly higher resolution, the resulting particle distribution converges to the exact geometry.