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 ›› 2024, Vol. 43 ›› Issue (2): 143-148.

• •    下一篇

融合知识图谱和案例推理的燃气应急辅助决策研究

胡玉玲1,2, 万雨瑞2, 李紫旋2, 齐子琛3   

  1. （1. 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；2. 北京建筑大学 电气与信息工程学院，北京 102016；3. 北京燃气集团，北京 100082）
  • 出版日期:2024-02-15 发布日期:2024-02-15
  • 作者简介:胡玉玲（1975- ），女，北京建筑大学电气与信息工程学院副教授，博士，主要从事应急管控与风险评估、融合知识图谱的燃气应急处置方法研究，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）；北京建筑大学基本科研业务基金（X20109）

Research on gas emergency assistance decision-making by integrating knowledge graph and case-based reasoning

Hu Yuling1,2, Wan Yurui2, Li Zixuan2, Qi Zichen3   

  1. (1. Beijing Key Laboratory of Intelligent Processing for Building Big Data, Beijing University of Civil Engineering and Architecture, Beijing 100044, China; 2. Institute of Electrical and Information Engineering, Beijing University of Civil Engineering and Architecture, Beijing 102016, China; 3. Beijing Gas Group, Beijing 100082, China)
  • Online:2024-02-15 Published:2024-02-15

摘要： 燃气事故发生时，应急处置人员基于文本资料查询应急处置方案难以满足处置科学性与快速性的需求。为此，提出了一种融合知识图谱和案例推理的燃气应急辅助决策方法。将已有燃气事故案例文本资料以知识图谱的形式存储与表示，利用案例推理的属性相似度与关系相似度加权计算方法，根据事故目标案例与源案例的综合相似度，检索出最佳相似源案例作为现场处置人员应急处置决策的重要参考，并将新的案例与处置措施存储于燃气应急处置知识图谱库中，实现处置措施的知识更新。通过案例相似度权重分配、属性权重分配、属性相似度方法选取，以及案例验证等，验证了本文方法的合理性与有效性。

关键词: 燃气事故, 知识图谱, 案例推理, 辅助决策

Abstract: A gas emergency decision-making method that integrates knowledge graph and case-based reasoning is proposed to address the difficulty of emergency response personnel searching for emergency response plans based on text data to meet the needs of scientific and rapid response in the current occurrence of gas accidents. Store and represent existing text data of gas accident cases in the form of a knowledge graph, propose a weighted calculation method for attribute similarity and relationship similarity in case reasoning, and based on the comprehensive similarity between the accident target case and the source case, retrieve the best similar source case as an important reference for emergency response decisions of on-site disposal personnel. Store new cases and disposal measures in the gas emergency response knowledge graph database, updated knowledge on implementation of disposal measures. The rationality and effectiveness of our method were verified through experiments such as case similarity weight allocation, attribute weight allocation, selection of attribute similarity methods, and case validation.

Key words: gas emergency, knowledge graph, case-based-reasoning, assistant decision