主管：中华人民共和国应急管理部
主办：应急管理部天津消防研究所
ISSN 1009-0029  CN 12-1311/TU
消防科学与技术

消防科学与技术 ›› 2022, Vol. 41 ›› Issue (6): 812-817.

• • 上一篇    下一篇

燃气输配突发事件应急处置的知识图谱构建

李思洁1,2，王亚慧1,2，张子豪1,2   

  1. （1.北京建筑大学 电气与信息工程学院，北京 100044；2.北京建筑大学 建筑大数据智能处理方法研究北京市重点实验室，北京 100044）
  • 出版日期:2022-06-15 发布日期:2022-06-15
  • 作者简介:李思洁（1996-），女，北京人，北京建筑大学电气与信息工程学院硕士研究生，主要从事基于知识图谱的燃气突发事件应急管理研究，北京市西城区展览馆路1号，100044。
  • 基金资助:
    国家重点研发项目(2018YFC0807806)；城镇燃气管网突发事件分析及应急处置标准化研究项目（0036201601）

Construction of knowledge graph for emergency disposal of gas transmission and distribution emergencies

LI Si-jie1,2， WANG Ya-hui1,2， ZHANG Zi-hao1,2   

  1. （1. School of Electrical and Information Engineering, Beijing University of Civil Engineering and Architecture, Beijing 100044, China; 2. Beijing Key Laboratory of Intelligent Processing for Building Big Data, Beijing University of Civil Engineering and Architecture, Beijing 100044, China）
  • Online:2022-06-15 Published:2022-06-15

摘要： 摘 要：基于知识图谱（Knowledge Graph）技术对燃气突发事件及其应急处置信息进行知识抽取、知识表示，提出了燃气突发事件应急处置的知识图谱构建方法，为后期进行完整的应急处置管理提供技术支撑。为解决燃气突发事件应急处置案例样本少的问题，采取网络爬虫技术爬取网络燃气突发事件应急处置案例，并以爬取的燃气突发事件应急处置案例及某燃气公司提供的燃气突发事件应急处置案例为研究对象，自顶向下建立燃气突发事件应急处置层次结构模型。利用BiLSTM-CRF模型和拓扑网络图进行案例中的实体抽取，通过实验验证实体抽取的有效性；定义实体间的关系并采用半监督协同训练的方法进行分类；利用Neo4j图数据库将燃气突发事件及应急处置方法知识图谱可视化。通过3个实际案例进行验证分析，证明上述方法的可靠性。

关键词: 关键词：燃气, 突发事件, 应急处置, 知识图谱, Neo4j

Abstract: Abstract: Based on the knowledge graph technology, the knowledge of gas emergencies and their emergency disposal information was extracted and represented, and the construction method of knowledge graph for gas emergency disposal was put forward, so as to provide technical support for complete emergency disposal management in the later stage. In order to solve the problem of few samples of gas emergency cases, this paper adopts Web crawler technology to crawl online gas emergency cases. Taking the crawling gas emergency case and the gas emergency case provided by a gas company as the research object, the top-down hierarchical model of gas emergency disposal is established. The BiLSTM-CRF model and topology network diagram are used to extract entities in the case, and the effectiveness of entity extraction is verified by experiments. The relationship between entities is defined and classified by semi supervised collaborative training method. The neo4j graph database is used to visualize the knowledge graph of gas emergencies and emergency disposal methods. Through the verification and analysis of three practical cases, it is proved that the above method has good effect.

Key words: Key words: gas, emergency accident, emergency management, knowledge graph, Neo4j