Abstract:

Automatic feature recognition (AFR) is one of the key technologies of intelligent manufacturing. Traditional rule-based recognition algorithms have poor scalability, while methods based on deep convolutional networks use discrete models as input, have low accuracy, and the recognition results are difficult to accurately map back to the original CAD model, causing inconvenience in application. In view of the above shortcomings, we proposed a feature recognition method based on graph neural network, which can directly analyze B-Rep models. Our method extracts effective characteristic information and geometric information from the B-Rep structures to form a feature descriptor, and then establishes an adjacency graph consist of high-level semantic information based on the topological structure of the CAD model. Taking the adjacency graph as input, an efficient graph neural network model is constructed. By introducing a differentiable generalized message aggregation function and a residual connection mechanism, the model has stronger information aggregation performance and multi-level feature capture capabilities. What is more, the message normalization strategy is used to ensure the stability of the training process and to accelerate the convergence of the model. After training, the network can directly classify and annotate all faces in the B-Rep model, thereby realizing feature recognition. Experimental results based on the public dataset MFCAD++ demonstrates that the proposed method achieved an accuracy of 99.53% and an average intersection-over-union ratio of 99.15%, which outperforms other similar studies. Further evaluations using more complex testing cases and typical CAD cases from real engineering applications show that the proposed method has better generalization ability and adaptability.

Key words: machining feature recognition, graph neural network, deep learning, computer aided design