MBD Based 3D CAD Model Automatic Feature Recognition and Similarity Evaluation

Automatic Feature Recognition (AFR) is considered as the key connection technique of integration Computer Aided Design (CAD) and Process Planning (CAPP). At present, there a lack systematic method to identify evaluate local features 3D CAD models. The process information such topological structure, shape size, tolerance surface roughness should be considered. Therefore, novel Model Based Definition (MBD) based on model AFR similarity evaluation are proposed in this paper. A Multi-Dimensional Attributed Adjacency Matrix (MDAAM) MBD established fully consideration roughness, other B-rep model. MDAAM, two-stage feature proposed, which combines methods optimal matching adjacency judgment. First, faces source target used independent sets construct bipartite graph. Secondly, supplement vertices set make number two equal. Thirdly, MDAAM data, weighted complete graph constructed with face between weight. Fourthly, Kuhn-Munkres algorithm calculate Fifthly, judged. Finally, models calculated, result. effectiveness verified by three applications.