基于网格的铁路轨道状态大数据可视化模型  

A grid-based visualization model for big data of railway track condition

作　　者：李擎[1] 刘仍奎[1] 白磊[1,2] 王福田 陈云峰[4] LI Qing;LIU Rengkui;BAI Lei;WANG Futian;CHEN Yunfeng(School of Traffic and Transportation,Beijing Jiaotong University,Beijing 100044,China;Beijing E-Hualu Information Technology Co.,Ltd.,Beijing 100043,China;State Key Laboratory of Rail Traffic Control and Safety,Beijing Jiaotong University,Beijing 100044,China;Permanent Way Department,Lanzhou Railway Bureau,Lanzhou 730000,China)

机构地区：[1]北京交通大学交通运输学院,北京100044 [2]北京易华录信息技术股份有限公司,北京100043 [3]北京交通大学轨道交通控制与安全国家重点实验室,北京100044 [4]兰州铁路局工务处,甘肃兰州730000

出　　处：《铁道科学与工程学报》2018年第7期1879-1885,共7页Journal of Railway Science and Engineering

基　　金：国家自然科学基金资助项目(51578057);轨道交通控制与安全国家重点实验室(北京交通大学)自主研究课题资助项目(RCS2016ZT007)

摘　　要：为全面直观地把握轨道健康状态,基于网格化管理思想,提出一种基于网格的铁路轨道状态大数据可视化模型。将铁路线路划分为连续的等长的轨道网格,采用多维尺度分析算法和混合层次K均值聚类算法,对轨道网格健康状态分布进行可视化展现,直观反映各轨道网格健康状态特征的相似性或差异性。选取兰新线1 447个轨道网格的状态数据,验证了模型的有效性。It is of great significance for railway managers to have a comprehensive and visual assessment of the track health for the guarantee of operational safety, and reasonable allocation of M R resources. A grid-based visualization model for big data of railway track condition was proposed. The model divided a continuous railway track line into adjacent ＂segments＂ of the same specific length. These ＂segments＂ are called track grids. The multidimensional scaling and hybrid hierarchical k-means clustering method were employed to present health distribution of track grids visually, and directly reflect the similarities and differences of track grids health features. The real condition data of 1 447 track grids in the Lanxin railway was collected to verify the proposed model.

关 键 词：铁路轨道 轨道状态 网格化 大数据 可视化分析 

分 类 号：U216.4[交通运输工程—道路与铁道工程]