基于知识图谱的工艺数据表达及提取  

Knowledge Graph-Based Representation and Extraction of Process Data

作　　者：张金龙 高琦[1,2] 翟健丰 吴春阳 吴晓晴 ZHANG Jinlong;GAO Qi;ZHAI Jianfeng;WU Chunyang;WU Xiaoqing(School of Mechanical Engineering,Ministry of Education,Shandong University,Jinan 250061,China;Key Laboratory of High Efficiency and Clean Mechanical Manufacture Shandong University,Ministry of Education,Shandong University,Jinan 250061,China)

机构地区：[1]山东大学机械工程学院,济南250061 [2]山东大学高效洁净机械制造教育部重点实验室,济南250061

出　　处：《组合机床与自动化加工技术》2024年第12期158-163,168,共7页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：山东省自然科学基金项目(ZR2020ME139)。

摘　　要：工艺数据是离散制造企业中最为关键的数据之一,如何将异构的工艺数据进行一体化管理成为目前企业面对的重要问题。对此提出一种基于知识图谱的工艺数据表达及提取方法,建立产品、工艺过程、工厂、资源为中心的工艺本体模型,作为知识图谱模式层规范知识图谱的架构,实现工艺数据的统一化表达。针对结构化工艺数据库以及半结构化工艺卡片的数据提取,提出基于本体对齐的数据提取到知识图谱的方法。对于非结构化工艺语句,使用一种实体识别模型与本体推理相结合的信息提取方法。构建了基于知识图谱的工艺数据表达及提取原型系统,验证了方法的有效性。Process data is one of the most critical types of data in discrete manufacturing enterprises,and how to integrate heterogeneous process data has become a crucial issue for current enterprises.A method based on a knowledge graph is proposed for the expression and extraction of process data.A process ontology model is established with a focus on products,process workflows,factories,and resources,serving as the central structure for the knowledge graph model to standardize the architecture and achieve unified expression of process data.For the extraction of data from structured process databases and semi-structured process cards,a method based on ontology alignment is proposed to extract data into the knowledge graph.In the case of unstructured process statements,an information extraction method combining entity recognition models and ontology reasoning is employed.A prototype system for process data modeling and extraction based on a knowledge graph is constructed,validating the effectiveness of the proposed method.

关 键 词：离散制造业 工艺数据 知识图谱 本体建模 数据提取 

分 类 号：TH164[机械工程—机械制造及自动化] TG659[金属学及工艺—金属切削加工及机床]