Mesh generation based on metal foam tomography data
(2025) FHLL01 20251Solid Mechanics
Department of Construction Sciences
- Abstract
- This thesis investigates the generation and evaluation of tetrahedral meshes based on tomographic data of open-cell aluminum metal foam. The work focuses on creating volume meshes suitable for finite element analysis (FEA), starting from a 3D reconstruction of the metal foam microstructure derived from a stack of over 1000 CT-like image slices, tomographs.
The work was based on, and extended a Python workflow that could process the data, segment the material, extract a surface mesh, and generate a tetrahedral mesh. Tetrahedral mesh generation was performed using TetGen, a Delaunay-based mesh generator. The mesh quality was evaluated using standard shape metrics such as aspect ratio, dihedral angle and volume deviation.
The work is... (More) - This thesis investigates the generation and evaluation of tetrahedral meshes based on tomographic data of open-cell aluminum metal foam. The work focuses on creating volume meshes suitable for finite element analysis (FEA), starting from a 3D reconstruction of the metal foam microstructure derived from a stack of over 1000 CT-like image slices, tomographs.
The work was based on, and extended a Python workflow that could process the data, segment the material, extract a surface mesh, and generate a tetrahedral mesh. Tetrahedral mesh generation was performed using TetGen, a Delaunay-based mesh generator. The mesh quality was evaluated using standard shape metrics such as aspect ratio, dihedral angle and volume deviation.
The work is structured to follow the iterative testing process, where each test different meshing parameters, geometry simplifications and resolutions. The impact of a developed surface decimation step is also evaluated with respect to meshing time, element count, and quality loss. It was found that mesh quality is sensitive to both voxel size and surface simplification, and that aggressive smoothing can result in topological defects that make meshing impossible.
The results show that careful tuning of both TetGen parameters and simplification steps are required to balance mesh quality and computational cost. The final pipeline allows for flexible mesh generation, with the additional step of decimating the surface mesh from tomographic metal foam data. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9203927
- author
- Olanders, Jesper LU
- supervisor
- organization
- course
- FHLL01 20251
- year
- 2025
- type
- M2 - Bachelor Degree
- subject
- keywords
- Mesh generation, Metal foam, Tomography, Python
- language
- English
- id
- 9203927
- date added to LUP
- 2025-07-02 09:42:11
- date last changed
- 2025-07-02 09:42:11
@misc{9203927, abstract = {{This thesis investigates the generation and evaluation of tetrahedral meshes based on tomographic data of open-cell aluminum metal foam. The work focuses on creating volume meshes suitable for finite element analysis (FEA), starting from a 3D reconstruction of the metal foam microstructure derived from a stack of over 1000 CT-like image slices, tomographs. The work was based on, and extended a Python workflow that could process the data, segment the material, extract a surface mesh, and generate a tetrahedral mesh. Tetrahedral mesh generation was performed using TetGen, a Delaunay-based mesh generator. The mesh quality was evaluated using standard shape metrics such as aspect ratio, dihedral angle and volume deviation. The work is structured to follow the iterative testing process, where each test different meshing parameters, geometry simplifications and resolutions. The impact of a developed surface decimation step is also evaluated with respect to meshing time, element count, and quality loss. It was found that mesh quality is sensitive to both voxel size and surface simplification, and that aggressive smoothing can result in topological defects that make meshing impossible. The results show that careful tuning of both TetGen parameters and simplification steps are required to balance mesh quality and computational cost. The final pipeline allows for flexible mesh generation, with the additional step of decimating the surface mesh from tomographic metal foam data.}}, author = {{Olanders, Jesper}}, language = {{eng}}, note = {{Student Paper}}, title = {{Mesh generation based on metal foam tomography data}}, year = {{2025}}, }