none

碩士 === 國立中央大學 === 機械工程學系 === 103 === Automatic feature recognition can be used extensively in industry both in CAE and CAM application. For CAE analysis, we always waste a lot of time to modify the mesh manually at specific features (Hole, Fillet, Rib Chamfer…) to improve the quality of meshes. Espe...

Full description

Bibliographic Details
Main Authors: Hsing-yu Chen, 陳星佑
Other Authors: Jiing-Yih Lai
Format: Others
Language:zh-TW
Published: 2015
Online Access:http://ndltd.ncl.edu.tw/handle/77119834199917101928
Description
Summary:碩士 === 國立中央大學 === 機械工程學系 === 103 === Automatic feature recognition can be used extensively in industry both in CAE and CAM application. For CAE analysis, we always waste a lot of time to modify the mesh manually at specific features (Hole, Fillet, Rib Chamfer…) to improve the quality of meshes. Especially, blend features always cause the narrow and long mesh in mesh generating, leading to the complexity of the topology structure and the difficulty to recognize other features. Therefore, this research uses the B-rep (Boundary representation) data structure in CAD models to develop automatic blend recognition technology, for improving the identification in the parts, and for providing the interrelated information for other feature recognition and CAE analysis to conduct the follow-up uses. In this program, users are able to modify the threshold, based on determining the smooth continuity or other conditions to recognize blend feature and classify the results. Finally, several real cases are presented and compared with those by CADdoctor, commercial software, to verify the accuracy of the proposed method.