【作者】 陈曦；

【导师】 李润梅；

【作者基本信息】 北京交通大学 ， 控制工程（专业学位）， 2017， 硕士

【摘要】 列车运行控制系统(简称列控系统,Chinese Train Control System,CTCS),由地面设备系统和车载设备系统构成,车载设备系统对高铁的运行起着至关重要的作用。车载计算机运行中形成的Log日志文件和工作人员撰写的交接班记录中包含着大量的、反映车载设备运行状态的信息,是车载设备故障发现和诊断的重要依据。然而对这两类文本数据的利用目前仍局限于工作人员人工记录、查询和解读,存在效率低、主观性强、缺乏理论指导意义等一系列问题。本文针对以上问题,通过对上述两类文本进行信息抽取和整理,建立了较为完备的故障字典;基于正则表达式方法,结合Matlab仿真,实现了 Log日志文件语句的自动定位,开发了车载设备故障发现的自动化处理平台;从而提高了工作人员交接班故障发现、记录和分析流程的工作效率;避免了以往人工处理的主观性;所构建的故障字典对故障发现和进一步研究具有指导意义。本文的工作主要有以下几个方面。首先,本文介绍了高铁车载设备的基本结构及本文研究需要的两类数据源,即:车载计算机Log日志文本及工作人员交接班记录文本;分析了两类文本文件的数据特点,总结了信息抽取算法等研究成果;采用数据挖掘工具WEKA与SQL Server数据库相结合的方式进行处理探索,发现了常规故障文本数据挖掘的问题。结合两类文本数据特点,确定了故障发现与自动定位的研究框架。其次,确定了最大匹配分词算法和TFIDF算法对工作人员交接班记录和Log日志文件进行信息抽取,构建了标准化的故障字典,并且针对Log文件非结构化的特点利用正则表达式算法实现了故障语句的自动定位。最后,利用Matlab编程实现了基于两类文本文件的故障自动发现与定位,搭建了适用于Windows系统的故障发现人机交互平台,并基于已有的Log日志文件对平台进行了实用性展示。更多还原

【Abstract】 Chinese Train Control System(CTCS)consists of a ground equipment system and an on-board vehicle equipment system.As an important part of CTCS,the stability of on-board vehicle equipment(referred to as "vehicle equipment")plays a crucial role in the operation of high-speed railway.Log files formed in the process of the on-board computer and staff shift records include a large amount of information reflecting the operation status of vehicle equipment,which are important data source for vehicle equipment fault discovery.However,the existing way of utilization of these data still depends on the staff experience and manual record and query,which is inefficient,subjective,and lack of theoretical basis and guidance for fault discovery.In this thesis,based on extracting the Log files and staff shift records,a fault dictionary about vehicle equipment of high speed railway is made,the automatic positioning of faults in Log files statement is achieved,and the automatic fault discovery platform is established based on the Matlab simulation tools,which have guiding significance for further study,reduces the manual operation of staff and improves the work efficiency greatly.The whole research work is mainly from the following aspects.Firstly,the structure of vehicle equipment and information extraction algorithms are mentioned in the thesis.Then the characteristics of Log files and the staff shift records are analyzed.Based on WEKA data mining tool and SQL Server database,the text data mining is studied.The results show that WEKA and SQL Server are not suitable for the data.Based on the characteristics of the data,a research framework of fault discovery automatic location is established.Secondly,the Maximum Matching(MM)algorithm and Term Frequency&Inverse Documentation Frequency(TFIDF)algorithm are selected to extract information from two kinds of data and a standardized fault dictionary is built.The automatic positioning of fault statement in Log file is realized by using regular expression algorithm.Finally,based on the Matlab simulation tool,a human-computer interaction platform,suitable for Windows system,is created with the help of established framework.By dealing with Log files formed during April,the usage of platform process,module function and results are shown in this thesis.更多还原