摘 要

目前我国正处于城市轨道交通的高速发展阶段，随着建设规模的扩大和运营线路的增加，地铁为社会经济发展以及居民生活方式带来了巨大的改变。由于地铁利用地下空间发展，系统结构较为复杂，一旦发生突发事故，产生的影响不容小觑。因此，地铁运营安全问题一直是我国政府及相关部门单位关注的重点，也是该领域的研究热点。应急预案是预先制定的有效应对运营突发事件的行动指南，在应急管理各项工作中扮演着极为重要的角色。然而，目前大多数基于纸质文档或电子档的应急预案管理方式在知识存储、更新、应用以及集成等方面存在缺陷，严重阻碍了应急预案发挥应急指导作用的及时性、有效性。因此，实现应急预案知识的规范化与语义化表达，对提升应急预案的知识存储、管理与应用效率，促进更高效的应急管理工作的开展具有十分重大的意义。

本文利用本体在知识表达、信息集成方面的优势，首先提出了一种基于本体的地铁运营应急预案知识建模方法，构建了统一、规范化的地铁运营应急预案本体知识库。其次，结合自然语言处理技术、语义标注等方法对突发事件案例知识进行了语义化表达，完成了案例本体知识库的构建。最后，本文通过地铁运营应急预案知识检索及培训原型系统的开发，实现了预案知识推理及案例语义检索功能的应用展示，验证了所提出方法的可行性与有效性。本研究不仅显著提升了应急预案知识的管理和应用效率，展示了智能化的应急管理系统的发展前景，还为今后各类应急信息集成与互操作提供了知识共享平台，有利于促进更高效的各应急响应参与单位的协同合作。

关键词：地铁运营应急预案 本体 知识建模 语义相似度

Abstract

Urban rail transit is at a stage of rapid development in China. With the expansion of construction scale and operation lines, metro has brought huge changes to social economics and the lives of local residents. Because of the complexity of metro systems, the influence of metro operation emergencies can be massive. Hence, metro operation safety has always been our government's attention and research field’s focus. Emergency plans are efficient action guidance aimed to deal with all kinds of emergencies. It plays a key role in the emergency management of metro operation. However, most emergency plans remain as plain-text documents, which make it difficult to conduct efficient knowledge management, affecting the practicality and feasibility of emergency plans on site. Hence, it is important to organize and represent the plan knowledge in a standard and semantic way, which can improve the efficiency in both knowledge management and response work on site.

Based on the advantages of ontology in knowledge representation and information integration, this thesis has proposed an ontology-based knowledge modeling method of emergence plan. A unified and formalized ontology knowledge base of emergency plans has been built. Then, the knowledge of metro incident cases has been semantically represented with natural language processing technology and semantic annotation, achieving the construction of case knowledge base. A prototype of emergency plan retrieval and training system has been developed to demonstrate the feasibility and effectiveness of the method. This research improves the efficiency in both administrative and operational use of emergency plans, and shows a bright future of the development of intelligent emergency management system. It also provides a knowledge sharing platform for information integration and interoperability, which can promote efficient collaboration for different parties involved in the emergency response process.

Key words: Metro Operation Emergency Plan Ontology Knowledge Modeling Semantic Similarity