Automatic FEM Mesh Generation of 3D Mid-Surface and Solid CAD Model with Shape Recognition Technique

2016-01-1382

04/05/2016

Event
SAE 2016 World Congress and Exhibition
Authors Abstract
Content
This paper describes how we recognize various specific features from the shape of input 3D CAD models and create adequate mesh automatically to meet the detailed mesh rules. It is important to create adequate mesh according to the mesh rules in order to assure the quality of simulation results since simulation results may vary depending on the input FEM mesh. Not only simple features such as fillets, flanges, cylinders, chamfers, and emboss but also other features and complex features such as thin planes, tapered tubes, ribs, and grooves can matter. However, it is time consuming to create adequate FEM mesh according to the mesh rules where the ideal mesh should be different for each analysis phenomena, such as crash, structural, and NVH phenomena. Recognizing shapes of 3D CAD models and controlling mesh for specific features automatically help to reduce time spent on manual mesh modifying work dramatically. To recognize specific features for the mesh rules, we used characteristics of each face of the input 3D CAD models. Characteristics we check are such as the maximum and minimum curvature, the length of perimeters, and inner angle of the face. Which characteristics we check varies on the features. In addition to simple features, we also developed feature recognition algorithm to recognize composite surface features such as a bolt seating surface. We recognize them by combination of faces and their cross sectional shape as well as characteristics of each face. Using the shape recognition technique described in this paper, we achieved to automate mesh generation procedures following the specific mesh rules.
Meta TagsDetails
DOI
https://doi.org/10.4271/2016-01-1382
Pages
4
Citation
Otani, K., "Automatic FEM Mesh Generation of 3D Mid-Surface and Solid CAD Model with Shape Recognition Technique," SAE Technical Paper 2016-01-1382, 2016, https://doi.org/10.4271/2016-01-1382.
Additional Details
Publisher
Published
Apr 5, 2016
Product Code
2016-01-1382
Content Type
Technical Paper
Language
English