面向知识重用的挤压模具本体与知识图谱构建方法研究

doi:10.16731/j.cnki.1671-3133.2025.08.009

摘要/Abstract

摘要： 针对目前铝型材挤压模具领域生产设计严重依赖设计人员的知识经验、设计工艺知识重用困难及经验知识资源丢失等问题,提出了挤压模具本体知识表示模型,探索了挤压模具知识图谱(EDKG)的构建方法。将模具设计与铝型材挤压生产相关的规则经验归纳为一般性知识,通过本体构建七步法完成挤压模具知识图谱模式层的搭建。提出一种基于IA-DBSCAN的无监督知识抽取算法,通过融合词嵌入、信息熵与词频-逆文档频率的实体特征权重,将应用在大规模挤压模具领域文本数据中抽取的知识概念添加到知识图谱数据层,实现模式层与数据层的逻辑交互。最后,通过资源描述框架文件映射规则R2RML,以铝型材挤压模具为案例实现了Neo4j图数据库的数据存储与可视化,能够直观展示模具设计、材料属性和加工工艺等多维度信息间的关联,促进了模具加工领域知识的整合与标准化。

关键词: 知识图谱, 挤压模具, 本体构建, 知识抽取, Neo4j图数据库

Abstract: Aiming at the problems that the production design of aluminum profile extrusion die field relies heavily on the knowledge and experience of designers,the design process knowledge is difficult to reuse,and the experience knowledge resources are lost,the knowledge representation model of extrusion die ontology was proposed,and the construction method of Extrusion Die Knowledge Graph (EDKG) was explored. The rule experience related to die design and aluminum profile extrusion production was summarized as general knowledge,and the knowledge graph model layer of extrusion die was completed by seven-step method of ontology construction. An unsupervised knowledge extraction algorithm based on IA-DBSCAN was proposed. By fusing the entity feature weights of word embedding,information entropy and word frequency-inverse document frequency,the knowledge concepts extracted from the text data in the field of large-scale extrusion die were added to the knowledge graph data layer to realize the logical interaction between the model layer and the data layer. Finally,through the mapping rule R2RML of the resource description framework file,the data storage and visualization of the Neo4j graph database were realized by taking the aluminum profile extrusion die as a case,which can visually display the correlation between multi-dimensional information such as die design,material properties and processing technology,and promote the integration and standardization of knowledge in the field of die processing.

Key words: knowledge graph, extrusion die, ontology construction, knowledge extraction, Neo4j graph database

中图分类号:

TP391.1
TG76

王培龙, 秦昊, 郝晓曦, 张昱. 面向知识重用的挤压模具本体与知识图谱构建方法研究[J]. 现代制造工程, 2025, 539(8): 79-92.

WANG Peilong, QIN Hao, HAO Xiaoxi, ZHANG Yu. Research on the construction method of extrusion die ontology and knowledge graph for knowledge reuse[J]. Modern Manufacturing Engineering, 2025, 539(8): 79-92.

参考文献

[1] LIU M,LI X,LI J,et al.A knowledge graph-based data representation approach for IIoT-enabled cognitive manufacturing[J].Advanced Engineering Informatics,2022,51:101515.
[2] 姜巨福,李明星,王迎.铝合金挤压铸造技术研究进展[J].中国有色金属学报,2021,31(9):2313-2329.
[3] SAHU R K,DAS R,ROUTARA B C,et al.Multihole extrusion process:A review[J].Materials Today:Proceedings,2022,62:3522-3527.
[4] 李世龙,庄新村,黄少东,等.基于知识的模锻工艺与模具设计系统开发[J].锻压技术,2012,37(3):140-144.
[5] 李柏林,王洪申,黄忠金,等.面向工艺规划重用的复杂曲面精加工知识图谱构建及可视化研究[J].现代制造工程,2022(3):6-12.
[6] 肖博文,杨晓英,张志文.基于神经网络的工艺路线知识库建立及优化[J].现代制造工程,2021(8):107-112.
[7] 李博.设计重用研究综述[J].计算机集成制造系统,2014,20(3):453-463.
[8] PENG C,XIA F,NASERIPARSA M,et al.Knowledge graphs:Opportunities and challenges[J].Artificial Intelligence Review,2023,56(11):13071-13102.
[9] HAN J,PEI J,TONG H.Data mining:concepts and techniques[M].Burlington:Morgan Kaufmann,2011.
[10] JIA J,ZHANG Y,SAAD M.An approach to capturing and reusing tacit design knowledge using relational learning for knowledge graphs[J].Advanced Engineering Informatics,2022,51:101505.
[11] BESTA M,GERSTENBERGER R,PETER E,et al.Demystifying graph databases:Analysis and taxonomy of data organization,system designs,and graph queries[J].ACM Computing Surveys,2023,56(2):1-40.
[12] MILLER J J.Graph database applications and concepts with Neo4j[C]//Proceedings of the Southern Association for Information Systems Conference.Atlanta:SAIS, 2013:141-147.
[13] SHIMIZU C,HAMMAR K,HITZLER P.Modular ontology modeling[J].Semantic Web,2023,14(3):459-489.
[14] 李宝清,王运平,曹立林,等.模具设计知识表示方法[J].锻压技术,2023,48(6):176-184.
[15] 苏宁,刘敏,洪楚扬,等.限定领域的本体自动构建方法[J].起重运输机械,2023(8):49-57.
[16] LI Z,LIU X,WANG X,et al.Transo:a knowledge-driven representation learning method with ontology information constraints[J].World Wide Web,2023,26(1):297-319.
[17] 张栋豪,刘振宇,郏维强,等.知识图谱在智能制造领域的研究现状及其应用前景综述[J].机械工程学报,2021,57(5):90-113.
[18] 方喜峰,柳大坤,龚婵媛,等.面向CAM数控编程领域的知识图谱构建方法[J].中国机械工程,2023,34(12):1486-1494.
[19] DONG J,JING X,LU X,et al.Process knowledge graph modeling techniques and application methods for ship heterogeneous models[J].Scientific Reports,2022,12(1):2911.
[20] 陈治宇,鲍劲松,郑小虎,等.基于长短期记忆网络的装配工艺语义识别方法[J].计算机集成制造系统,2021,27(6):1582- 1593.
[21] 冯锦丹,何楠,刘金山,等.基于知识驱动的航天复杂产品装配工艺重构方法[J].计算机集成制造系统,2024,30(12):4314-4327.DOI:10.13196/j.cims.2023.0561.
[22] 邱凌,张安思,张羽,等.面向无人机故障诊断的知识图谱构建应用方法[J].计算机工程与应用,2023,59(9):280-288.
[23] GUO L,YAN F,LI T,et al.An automatic method for constructing machining process knowledge base from knowledge graph[J].Robotics and Computer-Integrated Manufacturing,2022,73:102222.
[24] 刘洪波,陈越,卢记仓,等.面向知识图谱的规则挖掘研究综述[J].计算机工程与应用,2023,59(14):30-38.
[25] JIANG W,WANG Y,HU J,et al.Construction of substation engineering design knowledge graph based on “ontology seven-step method”[C]//2021 4th International Conference on Energy,Electrical and Power Engineering (CEEPE).Chongqing: IEEE,2021:957-962.
[26] RODRIGUEZ P L,SPIRLING A.Word embeddings:What works,what doesn’t,and how to tell the difference for applied research[J].The Journal of Politics,2022,84(1):101-115.
[27] PANDEY K,MISHRA A,RANI P,et al.Selecting features by utilizing intuitionistic fuzzy Entropy method[J].Decision Making:Applications in Management and Engineering,2023,6(1):111-133.
[28] AIZAWA A.An information-theoretic perspective of tf-idf measures[J].Information Processing & Management,2003,39(1):45-65.
[29] BHARADWAJ A G,STARLY B.Knowledge graph construction for product designs from large CAD model repositories[J].Advanced Engineering Informatics,2022,53:101680.
[30] HE B,CHEN W,LI F,et al.Directed acyclic graphs-based diagnosis approach using small data sets for sustainability[J].Computers & Industrial Engineering,2023,176:108944.
[31] MORWAL S,JAHAN N,CHOPRA D.Named entity recognition using Hidden Markov Model (HMM)[J].International Journal on Natural Language Computing (IJNLC),2012,1(4):15-23.
[32] LIU P,QIU X,HUANG X.Learning Context-Sensitive Word Embeddings with Neural Tensor Skip-Gram Model[C]//IJCAI.[S.l.]:AAAI Press,2015:1284-1290.
[33] GOLDBERG Y,LEVY O.Word2vec Explained:deriving Mikolov et al.'s negative-sampling word-embedding method[J/OL].ArXiv,2014[2024-06-12].DOI:10.48550/arXiv.1402.3722.
[34] TSCHANNEN M,BACHEM O,LUCIC M.Recent advances in autoencoder-based representation learning[J/OL].2018[2024-06-13].DOI:10.48550/arXiv.1812.05069.
[35] DINH D T,FUJINAMI T,HUYNH V N.Estimating the optimal number of clusters in categorical data clustering by silhouette coefficient[C]//Knowledge and Systems Sciences:20th International Symposium.Da Nang:Springer Singapore,2019:1-17.
[36] WANG X,XU Y.An improved index for clustering validation based on Silhouette index and Calinski-Harabasz index[C]//IOP Conference Series:Materials Science and Engineering.Zhuhai: IOP Publishing,2019:052024.
[37] ROS F,RIAD R,GUILLAUME S.PDBI:A partitioning Davies-Bouldin index for clustering evaluation[J].Neurocomputing,2023,528:178-199.
[38] MEDEIROS L F,PRIYATNA F,CORCHO O.MIRROR:Automatic R2RML mapping generation from relational databases[C]//Engineering the Web in the Big Data Era:15th International Conference.Rotterdam:Springer International Publishing,2015:326-343.