Knowledge graph construction technology for provision of sewing process information

被引：0
作者：
Zheng X. [1 ,2 ,3 ]
Liu Z. [4 ]
Liu B. [5 ]
Zhang L. [1 ,2 ,3 ]
Xu X. [6 ]
Liu X. [7 ]
机构：
[1] Institute of Artificial Intelligence, Donghua University, Shanghai
[2] Engineering Research Center of Artificial Intelligence Technology in the Textile Industry, Ministry of Education, Shanghai
[3] Shanghai Industrial Big Data and Intelligent Systems, Engineering Technology Center, Shanghai
[4] College of Mechanical Engineering, Donghua University, Shanghai
[5] Hangzhou Zhongfu Technology & Innovation Research Institute Co., Ltd., Zhejiang, Hangzhou
[6] HIKARI (Shanghai) Precise Machinery Scientific & Technology Co., Ltd., Shanghai
[7] College of Information Science and Technology, Donghua University, Shanghai
来源：
Fangzhi Xuebao/Journal of Textile Research | 2024年 / 45卷 / 04期
关键词：
clothing; flat seam; knowledge graph; knowledge recommendation; process knowledge management; sewing; sewing parameter;
D O I：
10.13475/j.fzxb.20230204901
中图分类号：
学科分类号：
摘要：
Objective The sewing process is characterized by long processing chains, diverse production elements and scattered processing information. Using knowledge graph technology for the management of design, operation and maintenance data generated during the sewing process, this research proposed a knowledge graph construction method for sewing process information management to achieve standardized knowledge representation. Method Modelling methods for the organisation of sewing process information were investigated. The process information generated during the fabric sewing process was classified, and a sewing process knowledge ontology model was established based on the classification results to realise the construction of a knowledge graph. The process recommendation method was established based on the graph. Experiments were carried out on fabric structure, fabric mechanical parameters and fabric sewing process to establish a knowledge system and to analyse the mechanical properties of fabrics before and after sewing. Based on the analysis, a regression model of fabric mechanical properties and sewing flatness and a theoretical model of fabric sewing shrinkage were established. An ontology model of the sewing parameter knowledge system was created for sewing parameter recommendation based on knowledge graph. Results According to the requirements of sewing process corpus and knowledge graph, a process recommendation method based on knowledge graph was established by combining the characteristics of industry knowledge structure and knowledge management requirements. The developed ontology and knowledge graph contains a total of 2865 entities and 52 relations, with wide knowledge coverage and strong generalization, facilitating the standardized representation of unstructured knowledge. The relationship between mechanical parameters and sewing parameters were modelled for common fabrics in the flat sewing process, the flatness of the sewn fabric and the maximum sewing shrinkage were predicted and recommendations for sewing parameters, bonding parameters and processing instructions for the corresponding fabrics were achieved. The technical architecture for intelligent recommendation of sewing parameters was established. The knowledge system was interconnected with other sewing process knowledge and enabled integration of process information. Conclusion The established knowledge graph is characterized by strong integration and interconnection of sewing process knowledge, which enables data integration and facilitates the maintenance and expansion of knowledge at a later stage. The research provides a useful supplementary case for process information management paths in the sewing industry, showing that knowledge graph technology has good application prospects in the sewing industry and has a certain reference value. © 2024 China Textile Engineering Society. All rights reserved.
引用
收藏
页码：195 / 203
页数：8
相关论文
共 14 条
  • [1] WU Dan, Application of artilicial intelligence technology in clothing manulacturing, Textile Reports, 40, 10, pp. 35-36, (2021)
  • [2] CONG Honglian, ZHAO Boyu, DONG Zhijia, Development status and application prospect of intelligent knitting products [J], China Textile Leader, 5, pp. 20-24, (2020)
  • [3] ZHANG Yongjun, CHENG Xin, LI Yansheng, Et al., Research on land and resources management and retrieval using knowledge graph, Geomatics and Information Science of Wuhan University, 47, 8, pp. 1165-1175, (2022)
  • [4] ZHANG Yunzhong, ZHU Rui, The Construction of knowledge graph in the LIS academic field of knowledge question answering system
  • [5] a multi-source data integration perspective, Information Science, 39, 5, pp. 115-123, (2021)
  • [6] AMERI F, BERNSTEIN W., A Thesaurus-guided Iramework for visualization of unstructured manulacturing capability data [C], IFIP International Conlerence on Advances in Production Management Systems, pp. 202-212, (2017)
  • [7] HU Yang, DING Yiwen, XU Feng, Et al., Knowledge recommendation system for human-robot collaborative disassembly using knowledge graph [C], International Manulacturing Science and Engineering Conlerence, (2021)
  • [8] TAO Jiaqi, LI Xinyu, ZHENG Pai, Et al., State-ol-theart and Irontier of manulacturing knowledge graph application[J/OL], Computer Integrated Manulacturing Systems, pp. 1-32
  • [9] ZHANG Donghao, LIU Zhenyu, JIA Weiqiang, Et al., Review on knowledge graph and its application prospects to intelligent manulacturing [J], Journal of Mechanical Engineering, 57, 5, pp. 90-113, (2021)
  • [10] KIM M, SUL I H, KIM S., Development of a sewing machine controller for seam pucker reduction using online measurement leedback system, Journal of Engineered Fibers and Fabrics, 12, 2, pp. 67-72, (2017)
12