| Author | Christopher J.R. Clutz |
|---|---|
| Title | Mesh Generation of SPG Based Articular Cartilage Models and a Rezoning System for 3D Multi-Region Models |
| Year | 1995 |
| School | Mechanical Engineering |
| Institution | RPI |
| Abstract | This thesis is in two parts. The first part addresses finite element mesh generation of solid models of human articular cartilage samples. These geometric models were defined using a collection of spatial points on the surfaces of the cartilage samples. Each data point was defined using an experimental procedure called stereophotogrammetry, or SPG. A finite element mesh of tetrahedral elements was created for three cartilage models using the SCOREC Finite Octree solid mesh generator. In order to provide the mesh generator with essential topological and geometric information about the models, a collection of interface operators were written. These subroutines are unique to this class of SPG based geometric models. Valid finite element meshes were successfully, and automatically, generated using these Finite Octree interface operators. In the second part of the thesis, a 3D rezoning system is presented which is capable of handling meshes of geometric models containing more than one region. The rezoning system can perform derivative recovery on a single mesh as well as solution interpolation between two meshes of the same geometric domain. Efficient techniques were used for derivative recovery and for the problem of locating the element containing a point inside a finite element mesh, which is necessary for solution interpolation. These techniques were compared with other existing methods to evaluate the relative effectiveness of each. Special data handling capacities were added to the rezoning system to track solution information at the boundaries between model regions. Results are presented for derivative recovery on meshes of multi-region models and for solution interpolation. |
| PDF File |  Download |