GNN-based deep reinforcement learning for MBD product model recommendation

被引:6
作者
Hu, Yuying [1 ]
Sheng, Zewen [1 ]
Ye, Min [2 ]
Zhang, Meiyu [1 ]
Jian, Chengfeng [1 ]
机构
[1] Zhejiang Univ Technol, Comp Sci & Technol Coll, Hangzhou, Peoples R China
[2] Zhejiang Highway Technicians Coll, Dept Mech Equipment, Hangzhou, Peoples R China
基金
中国国家自然科学基金;
关键词
Model based definition; graph neural networks; deep reinforcement learning; reuse; recommendation; RECOGNITION;
D O I
10.1080/0951192X.2023.2258090
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
Digital twin is more and more widely used, and the delivery demand of digital twin is more and more prominent at the same time of product physical delivery. Research on the digital twin product model recommendation method is of great significance for the rapid construction and reuse of digital twins. The methods currently in use, however, principally concentrate on geometric reuse and pay little attention to functional or knowledge reuse. In this paper, a graph neural network (GNN)-based deep reinforcement learning (DRL) for product model recommendation is presented. First, an MBD (model-based definition)-based semantic feature attribute adjacency graph (MSFAAG) is introduced to structured MBD model as the carrier of the digital twin product model. The MSFAAG is then embedded into continuous vector spaces using a GNN to obtain the categorization of these MBD models. Finally, DRL is used to adaptively identify more important semantic features, including manufacturing semantics and functional semantics, to obtain more detailed model classification results. The experiment effectively improves the reuse efficiency of the non-geometric aspects of the digital twin product and MBD model. Compared with other traditional recommendation algorithms, the algorithm proposed in this paper has higher accuracy and can well meet the design requirements of users.
引用
收藏
页码:183 / 197
页数:15
相关论文
共 30 条
[1]   Deep reinforcement learning meets graph neural networks: Exploring a routing optimization use case [J].
Almasan, Paul ;
Suarez-Varela, Jose ;
Rusek, Krzysztof ;
Barlet-Ros, Pere ;
Cabellos-Aparicio, Albert .
COMPUTER COMMUNICATIONS, 2022, 196 :184-194
[2]   MVCNN plus plus : Computer-Aided Design Model Shape Classification and Retrieval Using Multi-View Convolutional Neural Networks [J].
Angrish, Atin ;
Bharadwaj, Akshay ;
Starly, Binil .
JOURNAL OF COMPUTING AND INFORMATION SCIENCE IN ENGINEERING, 2021, 21 (01)
[3]   Shape and Context-Based Recognition of Standard Mechanical Parts in CAD Models [J].
Bonino, Brigida ;
Giannini, Franca ;
Monti, Marina ;
Raffaeli, Roberto .
COMPUTER-AIDED DESIGN, 2023, 155
[4]   Graph-Based Shape Analysis for Heterogeneous Geometric Datasets: Similarity, Retrieval and Substructure Matching [J].
Chen, Jiangce ;
Ilies, Horea T. ;
Ding, Caiwen .
COMPUTER-AIDED DESIGN, 2022, 143
[5]   The Digital Twin Application for Micro-Tool Wear Monitoring with Open-Source CAD System [J].
Christiand ;
Kiswanto, Gandjar ;
Baskoro, Ario Sunar .
ADVANCES IN PRODUCTION MANAGEMENT SYSTEMS: SMART MANUFACTURING AND LOGISTICS SYSTEMS: TURNING IDEAS INTO ACTION, APMS 2022, PT II, 2022, 664 :11-18
[6]   Hierarchical CADNet: Learning from B-Reps for Machining Feature Recognition [J].
Colligan, Andrew R. ;
Robinson, Trevor T. ;
Nolan, Declan C. ;
Hua, Yang ;
Cao, Weijuan .
COMPUTER-AIDED DESIGN, 2022, 147
[7]   A Survey on Network Embedding [J].
Cui, Peng ;
Wang, Xiao ;
Pei, Jian ;
Zhu, Wenwu .
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, 2019, 31 (05) :833-852
[8]   A new approach for reusable 3D CAD objects detection, by similarity calculation based on models Bayesian network (MBN) [J].
Fradi, Ahmed ;
Louhichi, Borhen ;
Mahjoub, Mohamed Ali ;
Eynard, Benoit .
INTERNATIONAL JOURNAL OF COMPUTER INTEGRATED MANUFACTURING, 2021, 34 (12) :1285-1304
[9]  
[高雪瑶 Gao Xueyao], 2017, [西南交通大学学报, Journal of Southwest Jiaotong University], V52, P416
[10]   Quantitative reactive modeling and verification [J].
Henzinger, Thomas A. .
COMPUTER SCIENCE-RESEARCH AND DEVELOPMENT, 2013, 28 (04) :331-344