摘要
地铁工程事故造成了严重的社会影响和经济损失,同时也为地铁施工提供宝贵的安全知识和工程经验。为了实现事故知识的有效集成和持续积累,为安全管理提供决策支持,本文提出基于知识图谱的地铁工程事故知识建模与事故分析方法。首先分析了知识图谱在地铁工程事故知识集成中的优势;设计领域知识概念和关系模式,从地铁工程事故数据中抽取关键事故知识要素,初步构建了地铁工程事故知识图谱,并存储在图数据库Neo4j中;在此基础上提出基于地铁工程事故知识图谱的事故分析流程,针对事故分类查询、统计分析、关联路径分析等任务进行实证分析。结果表明,知识图谱能够为地铁工程事故分析提供较全面的数据支撑,且事故分析结果可视化,能够为语义搜索和智能问答提供知识支持,从而为地铁工程安全管理提供决策支持。
Metro engineering accidents have caused serious social impact and economic losses,but also provided valuable safety knowledge and engineering experience for metro construction.In order to realize the effective integration and continuous accumulation of accident knowledge and provide decision support for the safety management,this paper proposed a knowledge graph-based method for metro engineering knowledge modeling and accident analysis.Firstly,it analyzed the advantages of knowledge graph in metro engineering accidents knowledge integration.Secondly,it designed the schema of concept and relationship,extracted key knowledge elements from the accidents data,and formed the knowledge graph of metro engineering accidents,which was stored in the graph database Neo4 j.On this basis,it presented the accident analysis process based on metro engineering accidents knowledge graph,experiments and analysis are carried out on accident classification query,statistical analysis,and correlation path analysis.The results show that the knowledge graph can provide comprehensive data support for various types of metro engineering accidents analysis,the analysis results can be displayed visually,it can also provide knowledge support for semantic search and intelligent question-and-answer,and then provide decision support for metro engineering safety management.
作者
王莉
王建平
许娜
邓勇亮
WANG Li;WANG Jian-ping;XU Na;DENG Yong-liang(School of Mechanics and Civil Engineering,China University of Mining and Technology,Xuzhou 221116,China)
出处
《土木工程与管理学报》
北大核心
2019年第5期109-114,122,共7页
Journal of Civil Engineering and Management
基金
国家自然科学基金(71801214)
关键词
地铁工程
事故分析
知识建模
知识图谱
图数据库
metro engineering
accident analysis
knowledge modeling
knowledge graph
graph database