In computer-aided design (CAD) and process planning (CAPP), feature recognition is an essential task which identifies the feature type of a 3D model for computer-aided manufacturing (CAM). In general, traditional rule-based feature recognition methods are computationally expensive, and dependent on surface or feature types. In addition, it is quite challenging to design proper rules to recognise intersecting features. Recently, a learning-based method, named FeatureNet, has been proposed for both single and multi-feature recognition. This is a general purpose algorithm which is capable of dealing with any type of features and surfaces. However, thousands of annotated training samples for each feature are required for training to achieve a high single feature recognition accuracy, which makes this technique difficult to use in practice. In addition, experimental results suggest that multi-feature recognition part in this approach works very well on intersecting features with small overlapping areas, but may fail when recognising highly intersecting features. To address the above issues, a deep learning framework based on multiple sectional view (MSV) representation named MsvNet is proposed for feature recognition. In the MsvNet, MSVs of a 3D model are collected as the input of the deep network, and the information achieved from different views are combined via the neural network for recognition. In addition to MSV representation, some advanced learning strategies (e.g. transfer learning, data augmentation) are also employed to minimise the number of training samples and training time. For multi-feature recognition, a novel view-based feature segmentation and recognition algorithm is presented. Experimental results demonstrate that the proposed approach can achieve the state-of-the-art single feature performance on the FeatureNet dataset with only a very small number of training samples (e.g. 8-32 samples for each feature), and outperforms the state-of-the-art learning-based multi-feature recognition method in terms of recognition performances.

In this paper, an automatic determination method of part build orientation for laser powder bed fusion is presented. This method includes two steps. First, an existing facet clustering-based approach is applied to automatically generate the alternative orientations of a laser powder bed fusion part. Second, support volume, volumetric error, surface roughness, build time and build cost are considered. Their values in each alternative orientation are estimated by certain estimation models. The weights of these factors are determined via pairwise comparison. The weighted sum model is used to calculate a summary value of the factors in each alternative orientation. According to the calculated summary values, an optimal orientation to build the part is generated. To demonstrate the method, a set of part orientation cases are tested, and effectiveness analysis and efficiency and characteristic comparisons are reported. The demonstration results suggest that the proposed method can work for both regular and freeform surface models, produce stable and reasonable results and provide desired efficiency.

In this paper, a novel facet clustering based method is proposed to generate alternative build orientations for laser powder bed fusion. This method consists of two steps. First, a hierarchical clustering algorithm is applied to divide facets of the design model in standard tessellation language format into different clusters, each of which shares a similar normal vector. Second, alternatives of each cluster are computed and the final set of alternative build orientations is generated by combining and refining the alternatives from all clusters. To illustrate and validate the method, a set of examples including both regular and freeform surface models are tested, and qualitative and quantitative comparisons between the method and the existing methods are reported. The results suggest that the proposed method is feasible and effective for both regular and freeform surface parts. It is evident that the proposed method can output stable results, provide satisfying efficiency, and work well with facet clusters of varying probability density.