Research on knowledge extraction of semi-structured assembly data and construction method of machine tool assembly knowledge graph
-
摘要: 如果机床当中的零部件信息以及部件之间的关系以图谱的形式展示,能够更为有效获悉零部件周围相连的部件信息以及当前零件定位细节,又可以辅助工厂数控机床装配,解决装配数据分散、无规范以及装配效率低下等机床现状问题。由于当前表述机床三维信息软件有很多,因此相应文件格式也众多,导致复杂装配产品难以数字化管理,同时传统的单一连接关系难以全面描绘三维图形,以至于这方面深度学习模型效率不高,无法真正理解机床模型知识,知识推理能力几乎没有。而知识图谱以图的形式,可以表示多种模型部件连接关系,对知识的理解能力远超单一连接关系,可以提高知识推理能力。通过机床装配数据准备、装配知识抽取、知识推理、知识存储和知识可视化表示环节来完成从三维模型到知识图谱构建。Abstract: If the information of parts and components in the machine tool and the relationship between parts are displayed in the form of graph, which can more effectively learn the information of parts connected around the parts and the positioning details of current parts. It can also assist the assembly of factory CNC machine tools, and solve the current problems of machine tools such as scattered assembly data, non-standard and low assembly efficiency. At present, there are many software describing 3D information of machine tools, so there are many corresponding file formats, which makes it difficult to digitally manage complex assembly products. At the same time, the traditional single connection relationship is difficult to comprehensively depict 3D graphics, so that the efficiency of in-depth learning model in this respect is not high, the knowledge of machine tool model cannot be truly understood, and the knowledge reasoning ability is almost not available. The knowledge graph, in the form of graph, can represent the connection relationship of various model components. The ability to understand knowledge is far beyond the single connection relationship, which can improve the ability of knowledge reasoning. The construction from three-dimensional model to knowledge graph is completed through the steps of machine tool assembly data preparation, assembly knowledge extraction, knowledge reasoning, knowledge storage and knowledge visualization.
-
Key words:
- knowledge graph /
- assembly knowledge extraction /
- knowledge reasoning
-
表 1 SAX解析主要重载函数表
函数名称 主要参数 功能 startElement 元素标签和属性信息 获取标签信息,标志
新的标签characters 字符串和长度 获取具体文本 endElement 元素标签 标志当前标签访问结束 表 2 部分关系提取结果
Start End Relation 5 1 AggregatedBy 5 2 InstanceOf 9 1 AggregatedBy 9 6 InstanceOf 13 1 AggregatedBy 13 10 InstanceOf 表 3 部分实体属性提取结果
Id Name Type 1 0632 Reference3DType 2 垫圈1 Reference3DType 5 垫圈1-1 Instance3DType 14 前端盖 Reference3DType 23 后端盖 Reference3DType -
[1] Singhal A. Introducing the knowledge graphs: things, not strings[J]. Official Google Blog, 2012, 16: 1-10. [2] 徐增林, 盛泳潘, 贺丽荣, 等. 知识图谱技术综述[J]. 电子科技大学学报, 2016, 45(4): 589-606. doi: 10.3969/j.issn.1001-0548.2016.04.012 [3] Bizer C, Lehmann J, Kobilarov G, et al. DBpedia - a crystallization point for the web of data[J]. Web Semantics Science Services & Agents on the World Wide Web, 2009, 7(3): 154-165. [4] Suchanek F M, Kasneci G, Weikum G. YAGO: a large ontology from wikipedia and wordnet[J]. Journal of Web Semantics, 2008, 6(3): 203-217. doi: 10.1016/j.websem.2008.06.001 [5] 王晓斌, 宁涛, 王可. 3DXML文件格式解析及应用[J]. 工程图学学报, 2010, 31(2): 33-37. [6] 顾星海, 鲍劲松, 吕超凡. 基于知识图谱的装配语义信息建模[J]. 航空制造技术, 2021, 64(4): 74-81. doi: 10.16080/j.issn1671-833x.2021.04.074 [7] 宋浩楠, 赵刚, 孙若莹. 基于深度强化学习的知识推理研究进展综述[J]. 计算机工程与应用, 2022, 58(1): 12-25. [8] 梁鸿翔, 吴肇良, 杨帅. 面向司法案件判定的知识引导智能分析系统[J]. 数据通信, 2021(1): 28-32, 47. [9] Lao N, Mitchell T M, Cohen W W. Random walk inference and learning in a large scale knowledge base[C] //Proceedings of the 2011 Conference on Empirical Methods in Natural Language Processing, EMNLP 2011, 27-31 July 2011, John McIntyre Conference Centre, Edinburgh, UK, A meeting of SIGDAT, a Special Interest Group of the ACL, F, 2011 .