An Approach to Recognize Machining Features for In-Process Models by Using a Hybrid Graph and Rule-Based Method
| Autor: | Xiao Liang Bai, Liang Li, Li Shao, Shu Sheng Zhang |
|---|---|
| Rok vydání: | 2014 |
| Předmět: |
business.industry
Computer science Voltage graph Quartic graph Pattern recognition General Medicine Directed graph Strength of a graph Graph Geometric graph theory Graph energy Graph bandwidth Graph power Clique-width String graph Adjacency list Artificial intelligence Adjacency matrix Null graph business Complement graph |
| Zdroj: | Applied Mechanics and Materials. 607:333-337 |
| ISSN: | 1662-7482 |
| DOI: | 10.4028/www.scientific.net/amm.607.333 |
| Popis: | A hybrid graph and rule-based method for recognizing machining features is proposed in this paper, which aims to get the machining feature in 3D in-process model construction system. Initially, the attributed adjacency graph (AAG) of the input MBD models is constructed and represented in matrix form. After that, by deleting the convex edges of the AAG, the concave sub-graphs are generated. For obtaining the volumetric features, their concave sub-graphs are matched with the feature template graphs in a pre-defined library; for those which can be generated by sweeping the edge loops, such as open pocket, closed pocket, through pocket and slot, their features are recognized by using rules. In experimental section, an example is given to illustrate the validity and effectiveness of the proposed method. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|