煤矿安全事件本体中案例推理的研究与应用

发布时间：2018-06-08 17:58

本文选题：事件本体 + 描述逻辑　；参考：《北京工业大学》2015年硕士论文

【摘要】：在我国,煤炭资源仍然在一次性能源结构中占据很大的比重,我国对煤炭资源的需求仍然很高。作为世界最大的煤炭生产国和消费国,因煤矿事故造成的死亡人数竟然是其他产煤国家死亡人数总数的3倍,这说明我国在煤矿安全事故的处理上还有很大的不足。事故发生的时候,会因为难以采取及时、科学的应急处理决策,而致使事态扩大或引起次生事故。本文采用面向事件本体的模型来表示煤矿安全事件,在事故发生后,通过案例推理,将事故处理的解决方案及时提供给决策人员,从而减少事故损失。首先,介绍了事件本体的概念、本体与本体语言、事件本体的相关定义以及事件间的关系。介绍了描述逻辑以及描述逻辑与本体语言的关系,还介绍了本体检测的基础Tableau算法。针对案例推理,从案例推理概述、案例推理相关技术以及案例推理应用领域进行了阐述。其次,对事件本体相关技术进行了大量的研究,通过分析其他案例表示模型在表示煤矿安全事件方面的不足,提出了面向事件本体的案例表示模型。并根据煤矿安全事件的特点、事件的构成及事件之间的关系,利用本体开发工具Protege构建了煤矿安全事件本体模型,并对数据属性和对象属性进行了定义。还对构建好的事件本体进行了检测,以确定事件本体是否存在冲突。再次,案例检索和案例的相似度计算是案例推理相关技术的核心部分,还对这两个方面进行了研究,并依据煤矿安全事件的特点,提出了改进的案例检索方法和相似度计算方法。最后,将以上研究成果应用于煤矿安全事件本体,并设计和实现了煤矿安全事件本体的案例推理系统,选取一个煤矿事故进行了案例推理,以展示案例推理工作流程。最后用实验的方式,将提出的相似度计算方法与向量空间模型相似度计算方法进行比较,验证了该相似度计算方法的有效性和准确性。
[Abstract]:In China, coal resources still occupy a large proportion in the disposable energy structure, and the demand for coal resources in China is still very high. As the world's largest coal producer and consumer, the death toll caused by coal mine accidents is three times that of other coal-producing countries, which indicates that China still has a lot of shortcomings in dealing with coal mine safety accidents. When an accident occurs, it is difficult to take timely and scientific emergency decision, which causes the situation to expand or cause secondary accident. In this paper, an event-oriented model is used to represent coal mine safety events. After an accident occurs, the solution of accident processing is provided to decision makers in time through case-based reasoning, so as to reduce the loss of accidents. Firstly, the concept of event ontology, ontology and ontology language, the definition of event ontology and the relationship between events are introduced. The description logic and the relation between description logic and ontology language are introduced, and the tableau algorithm based on ontology detection is also introduced. In view of the case-based reasoning (CBR), this paper expounds the general situation of CBR, the related technology of CBR and the application field of CBR. Secondly, a large number of research on event ontology related technology is carried out. By analyzing the shortcomings of other case representation models in the representation of coal mine safety events, an event ontology oriented case representation model is proposed. According to the characteristics of coal mine safety events, the composition of the events and the relationship between the events, the ontology model of coal mine safety events is constructed by using protege, and the data attributes and object attributes are defined. The event ontology was also detected to determine whether the event ontology was in conflict. Thirdly, case retrieval and case similarity calculation are the core of Case-Based reasoning (CBR) technology, and the two aspects are also studied, and according to the characteristics of coal mine safety events, An improved case retrieval method and a similarity calculation method are proposed. Finally, the above research results are applied to the coal mine safety event ontology, and the case-based reasoning system of the coal mine safety event ontology is designed and implemented. A coal mine accident is selected for case-based reasoning to demonstrate the work flow of CBR. Finally, by comparing the proposed similarity calculation method with the vector space model similarity calculation method, the validity and accuracy of the proposed similarity calculation method are verified.
【学位授予单位】：北京工业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TD76;TP391.1

【参考文献】