Computer aided design and manufacturing (CAD/CAM) has been a focal research area for the manufacturing industry. Genuine CAD/CAM integration is necessary to make products of higher quality with lower cost and shorter lead times. Although CAD and CAM have been extensively used in industry, effective CAD/CAM integration has not been implemented. The major obstacles of CAD/CAM integration are the representation of design and process knowledge and the adaptive ability of computer aided process planning (CAPP). This research is aimed to develop a feature-based CAD/CAM integration methodology. Artificial intelligent techniques such as neural networks, heuristic algorithms, genetic algorithms and fuzzy logics are used to tackle problems. The activities considered include: 1) Component design based on a number of standard feature classes with validity check. A feature classification for machining application is defined adopting ISO 10303-STEP AP224 from a multi-viewpoint of design and manufacture. 2) Search of interacting features and identification of features relationships. A heuristic algorithm has been proposed in order to resolve interacting features. The algorithm analyses the interacting entity between each feature pair, making the process simpler and more efficient. 3) Recognition of new features formed by interacting features. A novel neural network-based technique for feature recognition has been designed, which solves the problems of ambiguity and overlaps. 4) Production of a feature based model for the component. 5) Generation of a suitable process plan covering selection of machining operations, grouping of machining operations and process sequencing. A hybrid feature-based CAPP has been developed using neural network, genetic algorithm and fuzzy evaluating techniques.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:410540 |
| Date | January 2003 |
| Creators | Ding, Lian |
| Publisher | University of Bedfordshire |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://hdl.handle.net/10547/622150 |
Page generated in 0.0158 seconds