The Research Of Manufacturing Resource Based Machining Feature Recognition Approach

Feature technologies are widely regarded as the foundation for the integration of CAD and CAPP. It is always a hot research topic in the world during the last three decades. Many different feature recognition methods have been developed. However these methods still suffer from some drawbacks such as the difficulty in the recognition of interacting features, the poor machinability of generated features and the insufficient utilization of part information. These drawbacks prevent the existing feature recognition methods from practical applications in industries.To deal with the problems, after a survey of existing methods, this thesis focuses on the research of machining feature recognition approach based on manufacturing resources:Firstly, manufacturing resources such as machines, fixtures and cutters are defined parametrically. Geometrical data and technical requirements of machining parts are extracted from the CAD models and restructured using an enhanced winged-edge data structure and surface & tolerance relationship graph respectively to support the analysis processes in the feature recognition.Secondly, a machining capability model of manufacturing resources is established and conducted in two parts: the cutting capability of cutters and the clamping capability of fixtures. The cutting capability of cutters aims at providing mapping relationships between machining surfaces and cutters. The clamping capability of fixtures aims at providing knowledge about which surfaces can be used to realize the fixturing of a part.Thirdly, a new manufacturing resources based feature recognition approach is proposed. The process of feature recognition is carried out in two stages. First, feasible cutters, fixitures and setup planning are derived for each machining surface based on the machining capability model. Second, suitable manufacturing resources are selected using heuristic algorithms with the objective of improving the machining accuracy and efficiency. Surfaces that can be machined by a single cutter and fixture in the same setup are clustered into a machining feature.Finally, the popular testing part ANC-101 is used to test the validity and effectiveness of developed approach, and the final results illustrate that the approach can effectively improve the reliability of derived manufacturing resources by comprehensive utilization of geometric and technical requirements of machining parts, recognize and express the interacting relationships between machining features clearly as well as guarantee the machinability of generated features.