Please wait a minute...
Journal of ZheJiang University (Engineering Science)  2024, Vol. 58 Issue (2): 349-359    DOI: 10.3785/j.issn.1008-973X.2024.02.013
    
Recognition method of parts machining features based on graph neural network
Xinhua YAO(),Tao YU,Senwen FENG,Zijian MA,Congcong LUAN,Hongyao SHEN
School of Mechanical Engineering, Key Laboratory of 3D Printing Process and Equipment of Zhejiang Province, State Key Laboratory of Fluid Power and Mechatronic Systems, Zhejiang University, Hangzhou 310027, China
Download: HTML     PDF(2518KB) HTML
Export: BibTeX | EndNote (RIS)      

Abstract  

A method for recognizing machining features based on graph neural networks was proposed in order to address the difficulties in identifying intersecting features and accurately determining machining feature surfaces in existing deep learning-based approaches. Features of nodes and adjacent edges were extracted through a compression activation module, and a dual-layer attention network at the node and adjacent edge levels was constructed in order to segment the machining features corresponding to each node. The surface features and edge features of the part model were fully used combined with the topological structure of the part model. The recognition problem of non-face merged intersecting features was effectively addressed by employing attention mechanisms for deep learning on the feature information. The proposed method was experimentally compared with three other feature recognition methods on a dataset of parts with multiple machining features. The optimal results were obtained in terms of accuracy, average class accuracy and intersection-over-union metrics. The recognition accuracy exceeded 95%.



Key wordsmachining feature      attribute adjacency graph      graph neural network      attention mechanism      deep learning     
Received: 03 July 2023      Published: 23 January 2024
CLC:  TH?166  
Fund:  浙江省重点研发计划资助项目 (2021C01096);浙江省杰出青年科学基金资助项目(LR22E050002).
Cite this article:

Xinhua YAO,Tao YU,Senwen FENG,Zijian MA,Congcong LUAN,Hongyao SHEN. Recognition method of parts machining features based on graph neural network. Journal of ZheJiang University (Engineering Science), 2024, 58(2): 349-359.

URL:

https://www.zjujournals.com/eng/10.3785/j.issn.1008-973X.2024.02.013     OR     https://www.zjujournals.com/eng/Y2024/V58/I2/349


基于图神经网络的零件机加工特征识别方法

针对现有基于深度学习的方法存在的难以识别相交特征、无法精确确定加工特征面的问题,提出基于图神经网络的加工特征识别方法. 通过压缩激励模块提取节点与邻接边的特征,构建节点级与邻接边级的双层注意力网络,分割每个节点对应的加工特征. 该方法充分利用了零件模型的面特征与边特征,结合零件模型的拓扑结构,基于注意力机制对特征信息进行深度学习,可以有效地解决非面合并相交特征的识别问题. 在多加工特征零件数据集上,将该方法与其他3种特征识别方法进行实验对比,在准确率、平均类准确率和交并比3项指标上均取得最优结果,识别准确率高于95%.


关键词: 加工特征,  属性邻接图,  图神经网络,  注意力机制,  深度学习 
Fig.1 Diagram of intersection edge convexity algorithm
Fig.2 Comparison of curve sampling results
Fig.3 Part model and its sampling point distribution
Fig.4 Framework of machining feature recognition network
Fig.5 Feature extraction based on SENet
节点特征提取网络各层输入特征尺寸输出特征尺寸
Conv2d(7,64,3,1,1)(Nv,7,10,10)(Nv,64,10,10)
SE(16)(Nv,64,10,10)(Nv,64,10,10)
Conv2d(64,128,3,1,2)(Nv,64,10,10)(Nv,128,5,5)
SE(16)(Nv,128,5,5)(Nv,128,5,5)
Conv2d(128,256,3,1,2)(Nv,128,5,5)(Nv,256,3,3)
SE(16)(Nv,256,3,3)(Nv,256,3,3)
MixPool(Nv,256,3,3)(Nv,256,1,1)
Reshape(Nv,256,1,1)(Nv,256)
Linear(Nv,256)(Nv,64)
Tab.1 Parameters of node feature extraction network
边特征提取网络各层输入特征尺寸输出特征尺寸
Conv1d(7,64,3,1,1)(Ne,7,10)(Ne,64,10)
SE(16)(Ne,64,10)(Ne,64,10)
Conv1d(64,128,3,1,2)(Ne,64,10)(Ne,128,5)
SE(16)(Ne,128,5)(Ne,128,5)
Conv1d(128,256,3,1,2)(Ne,128,5)(Ne,256,3)
SE(16)(Ne,256,3)(Ne,256,3)
MixPool(Ne,256,3)(Ne,256,1)
Reshape(Ne,256,1)(Ne,256)
Linear(Ne,256)(Ne,64)
Tab.2 Parameters of edge feature extraction network
Fig.6 Node attention module
Fig.7 Feature type of multi-machining feature dataset
Fig.8 Example of models in multi-machining feature dataset
Fig.9 Feature and feature surface distribution of multi-machining feature dataset
Fig.10 Accuracy and IoU curves of multi-machining feature validation set
模型AApcIoU
基于面基于点基于面基于点基于面基于点
本文
模型
97.3595.1592.16
PointNet38.8575.5232.0435.4920.4527.53
UV-Net94.6591.2586.31
DGCNN51.4485.6546.0654.0831.9940.95
Tab.3 Comparison of experimental results of different models in multi-machining feature dataset %
Fig.11 Recognition results of part model feature
模型AApcIoU
完整网络模型97.3595.1592.16
去除邻接边特征96.3693.4689.60
去除图注意力网络85.7186.6578.52
只保留拓扑结构40.2527.1117.12
Tab.4 Comparison of different parameters in ablation experiment %
[1]   VANDENBRANDE J H, REQUICHA A G Spatial reasoning for the automatic recognition of machinable features in solid models[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1993, 15 (12): 1269- 1285
doi: 10.1109/34.250845
[2]   VERMA A K, RAJOTIA S A hint-based machining feature recognition system for 2.5 D parts[J]. International Journal of Production Research, 2008, 46 (6): 1515- 1537
doi: 10.1080/00207540600919373
[3]   WU J, LEI R, PENG Y. Manufacturing feature recognition method based on subgraph decomposition [C]// International Conference on Frontiers of Artificial Intelligence and Machine Learning . Hangzhou: IEEE, 2022: 120–126.
[4]   YAN H, YAN C, YAN P, et al Manufacturing feature recognition method based on graph and minimum non-intersection feature volume suppression[J]. The International Journal of Advanced Manufacturing Technology, 2023, 125 (11/12): 5713- 5732
[5]   易晗. 基于图和规则的钣金件特征识别方法研究[D]. 武汉: 华中科技大学, 2019.
YI Han. Research on feature recognition method of sheet metal parts based on graphs and rules [D]. Wuhan: Huazhong University of Science and Technology, 2019.
[6]   WOO Y, SAKURAI H Recognition of maximal features by volume decomposition[J]. Computer-Aided Design, 2002, 34 (3): 195- 207
doi: 10.1016/S0010-4485(01)00080-X
[7]   ZHANG Z, JAISWAL P, RAI R Featurenet: machining feature recognition based on 3d convolution neural network[J]. Computer-Aided Design, 2018, 101: 12- 22
doi: 10.1016/j.cad.2018.03.006
[8]   SHI P, QI Q, QIN Y, et al A novel learning-based feature recognition method using multiple sectional view representation[J]. Journal of Intelligent Manufacturing, 2020, (31): 1291- 1309
[9]   MA Y, ZHANG Y, LUO X. Automatic recognition of machining features based on point cloud data using convolution neural networks [C]// Proceedings of the 2019 International Conference on Artificial Intelligence and Computer Science . Wuhan: ACM, 2019: 229-235.
[10]   QI C R, SU H, MO K, et al. Pointnet: deep learning on point sets for 3d classification and segmentation [C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition . Hawaii: IEEE, 2017: 652-660.
[11]   高玉龙. 基于加工面点云数据深度学习的加工特征自动识别[D]. 大连: 大连理工大学, 2021.
GAO Yulong. Automatic recognition of machining features based on deep learning for machining surface point cloud data [D]. Dalian: Dalian University of Technology, 2021.
[12]   YAO X, WANG D, YU T, et al A machining feature recognition approach based on hierarchical neural network for multi-feature point cloud models[J]. Journal of Intelligent Manufacturing, 2023, 34 (6): 2599- 2610
doi: 10.1007/s10845-022-01939-8
[13]   CAO W, ROBINSON T, HUA Y, et al. Graph representation of 3d CAD models for machining feature recognition with deep learning [C]// International Design Engineering Technical Conferences and Computers and Information in Engineering Conference . [S. 1. ]: ASME, 2020, 84003: 202-212.
[14]   PAGANI L, SCOTT P J Curvature based sampling of curves and surfaces[J]. Computer Aided Geometric Design, 2018, (59): 32- 48
[15]   胡盼旺. 复杂回转类零件自动特征识别算法研究[D]. 武汉: 华中科技大学, 2019.
HU Panwang. Feature recognition based on graphs and rules for rotational parts [D]. Wuhan: Huazhong University of Science and Technology, 2019.
[16]   MEYER M, DESBRUN M, SCHRÖDER P, et al. Discrete differential-geometry operators for triangulated 2-manifolds [C]// Visualization and Mathematics III . [S. 1. ]: Springer-Verlag, 2003: 35-57.
[17]   PAULY M, GROSS M, KOBBELT L P. Efficient simplification of point-sampled surfaces [C]// I EEE Visualization . Boston: IEEE, 2002: 163-170.
[18]   HU J, SHEN L, SUN G. Squeeze-and-excitation networks [C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Beijing: IEEE, 2018: 7132-7141.
[19]   PAVIOT T. Pythonocc (7.7. 0) [EB/OL]. (2022-10-05)[2023-04-05].https://doi.org/10.5281/zenodo.3605364.
[20]   JAYARAMAN P K, SANGHI A, LAMBOURNE J G, et al. Uv-net: learning from boundary representations [C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. [S. 1. ]: IEEE, 2021: 11703-11712.
[1] Siyi QIN,Shaoyan GAI,Feipeng DA. Video object detection algorithm based on multi-level feature aggregation under mixed sampler[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(1): 10-19.
[2] Xuefei SUN,Ruifeng ZHANG,Xin GUAN,Qiang LI. Lightweight and efficient human pose estimation with enhanced priori skeleton structure[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(1): 50-60.
[3] Chao-hao ZHENG,Zhi-wei YIN,Gang-feng ZENG,Yue-ping XU,Peng ZHOU,Li LIU. Post-processing of numerical precipitation forecast based on spatial-temporal deep learning model[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(9): 1756-1765.
[4] Hai-feng LI,Xue-ying ZHANG,Shu-fei DUAN,Hai-rong JIA,Hui-zhi LIANG. Fusing generative adversarial network and temporal convolutional network for Mandarin emotion recognition[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(9): 1865-1875.
[5] Xiao-qiang ZHAO,Ze WANG,Zhao-yang SONG,Hong-mei JIANG. Image super-resolution reconstruction based on dynamic attention network[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(8): 1487-1494.
[6] Hui-xin WANG,Xiang-rong TONG. Research progress of recommendation system based on knowledge graph[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(8): 1527-1540.
[7] Xiu-lan SONG,Zhao-hang DONG,Hang-guan SHAN,Wei-jie LU. Vehicle trajectory prediction based on temporal-spatial multi-head attention mechanism[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(8): 1636-1643.
[8] Xiao-yan LI,Peng WANG,Jia GUO,Xue LI,Meng-yu SUN. Multi branch Siamese network target tracking based on double attention mechanism[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(7): 1307-1316.
[9] Zhe YANG,Hong-wei GE,Ting LI. Framework of feature fusion and distribution with mixture of experts for parallel recommendation algorithm[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(7): 1317-1325.
[10] Yun-hong LI,Jiao-jiao DUAN,Xue-ping SU,Lei-tao ZHANG,Hui-kang YU,Xing-rui LIU. Calligraphy generation algorithm based on improved generative adversarial network[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(7): 1326-1334.
[11] Wei QUAN,Yong-qing CAI,Chao WANG,Jia SONG,Hong-kai SUN,Lin-xuan LI. VR sickness estimation model based on 3D-ResNet two-stream network[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(7): 1345-1353.
[12] Xin-lei ZHOU,Hai-ting GU,Jing LIU,Yue-ping XU,Fang GENG,Chong WANG. Daily water supply prediction method based on integrated learning and deep learning[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(6): 1120-1127.
[13] Pei-feng LIU,Lu QIAN,Xing-wei ZHAO,Bo TAO. Continual learning framework of named entity recognition in aviation assembly domain[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(6): 1186-1194.
[14] Jun HAN,Xiao-ping YUAN,Zhun WANG,Ye CHEN. UAV dense small target detection algorithm based on YOLOv5s[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(6): 1224-1233.
[15] Jia-chi ZHAO,Tian-qi WANG,Li-fang ZENG,Xue-ming SHAO. Rapid prediction of unsteady aerodynamic characteristics of flapping wing based on GRU[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(6): 1251-1256.