Local Symmetry Based Hint Extraction of B-Rep Model for Machining Feature Recognition

Abstract

Machining feature recognition is a key technology in design for manufacturing, inter-operation of neutral format solid models, and reuse of design models in down-streaming applications. Although many approaches have been proposed, the detection and processing of intersection features are still challenging. In this paper, we studied the local property of B-rep model and usage in intersection feature detection. To support hint based reasoning and processing, we proposed a symmetry property definition with relation to edges of solid models. This symmetry property is based on conventional edge convexity, which is an extension of the feature hint theory. The Transformation of global or partial graph matching to local property matching would reduce the complexity in feature matching stages. We implemented a prototype module to evaluate the feasibility of the proposed approach.