Abstract
This paper proposes a method for predicting the complexity of meshing computer aided design (CAD) geometries with unstructured, hexahedral, finite elements. Meshing complexity refers to the relative level of effort required to generate a valid finite element mesh on a given CAD geometry. A function is proposed to approximate the meshing complexity for single part CAD models. The function is dependent on a user defined element size as well as on data extracted from the geometry and topology of the CAD part. Several geometry and topology measures are proposed, which both characterize the shape of the CAD part and detect configurations that complicate mesh generation. Based on a test suite of CAD models, the function is demonstrated to be accurate within a certain range of error. The solution proposed here is intended to provide managers and users of meshing software a method of predicting the difficulty in meshing a CAD model. This will enable them to make decisions about model simplification and analysis approaches prior to mesh generation.
Original language | English (US) |
---|---|
Pages (from-to) | 76-90 |
Number of pages | 15 |
Journal | Engineering with Computers |
Volume | 21 |
Issue number | 1 |
DOIs | |
State | Published - Nov 2005 |
Externally published | Yes |
All Science Journal Classification (ASJC) codes
- Software
- Modeling and Simulation
- General Engineering
- Computer Science Applications
Keywords
- Blend detection
- Geometry clean-up
- Mesh complexity
- Time to mesh