Feature recognition technology for aircraft structural parts based on a holistic attribute adjacency graph

Abstract

A feature-recognition method based on an holistic attribute adjacency graph is put forward to solve the problems that complex features of aircraft structural parts find difficult to recognize by traditional feature-recognition methods. Extending the attribute's information and adding node types based on a traditional attribute adjacency graph, the method not only represents freeform surfaces and edge features, but also describes geometric information of topology elements precisely and completely. Combining feature recognition based on a graph with that based on hint, the method can deal with feature recognition of freeform surfaces, edge features, intersecting features, and convex features by a uniform algorithm, which virtually performs hint search, hint extension, and feature combination with hints of seed faces. According to the research, an original system has been used in a numerically controlled machining process of integer frame parts in certain large aviation enterprises.