Cell agglomeration strategy for cut cells in eXtended discontinuous Galerkin methods

Muhammed Toprak,Matthias Rieckmann,Florian Kummer
2024-03-14
Abstract:In this work, a cell agglomeration strategy for the cut cells arising in the extended discontinuous Galerkin (XDG) method is presented. Cut cells are a fundamental aspect of unfitted mesh approaches where complex geometries or interfaces separating sub-domains are embedded into Cartesian background grids to facilitate the mesh generation process. In such methods, arbitrary small cells occur due to the intersections of background cells with embedded geometries and lead to discretization difficulties due to their diminutive sizes. Furthermore, temporal evolutions of these geometries may lead to topological changes across different time steps. Both of these issues, i.e., small-cut cells and topological changes, can be addressed with a cell agglomeration technique. In this work, a comprehensive strategy for the typical issues associated with cell agglomeration in three-dimensional and multiprocessor simulations is provided. The proposed strategy is implemented into the open-source software package BoSSS and tested with 2- and 3-dimensional simulations of immersed boundary flows.
Numerical Analysis,Computational Physics,Fluid Dynamics
What problem does this paper attempt to address?
The problems that this paper attempts to solve are two main issues related to cut cells in the extended discontinuous Galerkin (XDG) method: the small - cut - cell problem and the topological change problem. ### 1. Small - cut - cell problem When a geometry or interface is embedded into the background grid, so - called cut cells are generated. These cut cells may become very small due to the intersection with the embedded geometry, leading to difficulties in discretization. Specifically: - **Extremely small size**: The size of cut cells may be several orders of magnitude smaller than that of typical cells, which not only affects the condition number of discretization but also limits the time step of explicit schemes. - **Irregular shape**: The irregular shape of cut cells may lead to additional errors and complexity in integration. ### 2. Topological change problem When dealing with geometries that evolve over time, the topological structure may change between different time steps. Such changes will cause the discretization to be mismatched between different time steps, thus introducing computational challenges. ### Solution: Cell Agglomeration Strategy To solve the above problems, the paper proposes a cell agglomeration strategy, with the following main features: - **Aggregating small cut cells**: By merging small cut cells with their appropriate neighboring cells into larger cells, the numerical problems brought by small cut cells can be effectively alleviated. - **Handling topological changes**: The agglomeration strategy can also be used to adjust the topological changes caused by evolving interfaces, ensuring the consistency of discretization between different time steps. ### Method Implementation This strategy has been implemented in the open - source software package BoSSS and has been tested through two - dimensional and three - dimensional immersed - boundary - flow simulations. In addition, the paper also details how to efficiently parallelize the cell agglomeration algorithm in a multi - processor environment to meet the requirements of large - scale computing. ### Summary Through this method, the paper aims to improve the stability and accuracy of the XDG method when dealing with complex geometries and dynamic interface problems, while reducing the consumption of computational resources.