大数据平台下动车组运维数据可视化系统的设计与实现

发布时间：2018-01-24 03:43

本文关键词： 动车组大数据数据可视化数据预处理数据分析 Hadoop　出处：《北京交通大学》2017年硕士论文　论文类型：学位论文

【摘要】：近年来,我国高速铁路发展迅速,给人们出行提供了便捷。但动车组在运行过程中难免会出现各种各样的故障,如何保障动车组安全高效地运行已经成为动车运营领域研究的重点。动车组运维数据量大且种类复杂,如何快速的在这些杂乱无章的数据中挖掘出有效的信息,并以可视化的方式展示出来,已成为一个重要的研究方向。随着现代化技术的发展,动车组产生的数据量越来越大,现有的数据处理分析方法对大规模数据的处理已越来越不适应,难以满足用户的需求。针对上述问题,本文设计并实现了一个基于Hadoop大数据平台的动车组运维数据的可视化系统。系统实现了对动车组实时运行数据的缺失值进行插补;根据业务的实际需求,通过对可视化技术应用场景、绘制方法以及交互技术的研究,实现多种数据可视化展示功能,为业务人员提供一种可视化分析工具。本文的主要内容包括:(1)通过对动车组运维数据可视化系统的需求分析,提出了系统的整体方案设计。根据功能不同,将模块划分为权限管理模块、数据导入导出模块、数据预处理模块以及数据可视化分析展示模块,对各模块进行设计并实现。重点实现了数据预处理模块以及数据可视化分析展示功能。(2)针对动车组实时运行中车-地数据传输存在的数据丢失问题,本文提出了加权贝叶斯算法,并将其迁移到大数据平台上实现了 MapReduce化,对动车组运维数据进行清洗,从而提高可视化的数据质量。(3)结合业务实际需求,系统引用可视化插件Echarts对动车组运维数据进行可视化展示,实现了监测数据展示、故障数据统计分析、性能趋势分析以及关联关系分析等可视化功能,从而验证了可视化系统的实用性。
[Abstract]:In recent years, the rapid development of high-speed railway in China has provided convenience for people to travel, but EMU will inevitably appear a variety of faults in the operation process. How to ensure the safe and efficient operation of EMU has become the focus of research in the field of motor car operation. EMU has a large amount of data and complex types. How to quickly mining out effective information in these chaotic data. Visual display has become an important research direction. With the development of modern technology, the amount of data generated by EMU is increasing. The existing methods of data processing and analysis have become more and more unsuitable for large-scale data processing, and it is difficult to meet the needs of users. This paper designs and implements a visualization system of EMU operation and maintenance data based on Hadoop big data platform, which can interpolate the missing value of EMU real-time operation data. According to the actual needs of the business, through the application of visualization technology scene, drawing methods and interactive technology research, the realization of a variety of data visualization display functions. This paper provides a visual analysis tool for business personnel. The main content of this paper includes: 1) the requirement analysis of EMU operation and maintenance data visualization system is carried out. According to the different functions, the module is divided into the authority management module, the data import and export module, the data preprocessing module and the data visualization analysis module. The data preprocessing module and data visualization analysis display function. 2) aiming at the data loss problem existing in the real time operation of EMU, the data loss of vehicle-ground data transmission is realized. This paper proposes a weighted Bayesian algorithm and migrates it to big data platform to realize MapReduce to clean the data of EMU operation and maintenance. In order to improve the visual data quality, combined with the actual needs of the business, the system uses the visual plug-in Echarts to visualize the EMU operation and maintenance data, and realizes the monitoring data display. The visualization functions such as statistical analysis of fault data, performance trend analysis and correlation analysis are used to verify the practicability of the visualization system.
【学位授予单位】：北京交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U269;TP311.13

【参考文献】