结合可视图的多状态交通流时间序列特性分析.docVIP

结合可视图的多状态交通流时间序列特 性分析 邢雪 于德新 田秀娟 王世广 吉林大学交通学院吉林化工学院信息与控制工程学 院吉林大学吉林省道路交通重点实验室 ： 交通流时间序列的研究主要采用数据挖掘和机器学习的方法，这些“黑箱”挖 掘方法很难直观反映序列特性.为增强交通流时间序列及其特征分析的可视化性, 结合可视图理论来构建交通流时间序列的关联网络，从复杂网络角度实现交通 流时间序列的特性分析?在网络构建的过程屮，考虑到不同交通状态下交通流表 征具有的差异性，首先利用交通流参量的相关性对交通流状态进行分类，然后 构建不同交通状态下的时间序列复杂网络，并对这些网络的特征属性给出统计 分析，如度分布、聚类系数、网络直径、模块化等?研究表明，可视图法可为交 通流时间序列映射到网络提供有效途径，并且不同状态下交通流时间序列构建 的复杂网络的模块化、聚类系数和度分布等统计特征呈现一定的变化规律，为交 通流运行态势的研究提供了可视化的分析角度. 关键词： 交通流时间序列;可视图;复杂网络;聚类分析; ：邢雪， : patricia_xx@126. com ：2017年3月19日 基金：国家自然科学基金（批准号 Analysis of multi-state traffic flow time series properties using visibility graph Xing Xue Yu De-Xin Tian Xiu-Juan Wang Shi-Guang College of Transportation, Jilin University; Abstract： A traffic flow time series is a sequence of traffic detection pareimeters in chronological order. This differs from a gencral quantitativc data sequence in that the time series includes a time attribute that contains not only the data with time characteristics, but also the distribution of the data itself. To date, studies of traffic time series have primarily adopted data mining met hods consist ing of data mining and mach ine 1 earn ing methods—similar sequcncc search, dimcnsion reduction, clustering， classification, pattern analysis, prediction, etc. In order to improve the visualization of traffic flow time series and feature analyses, a proposed method builds the association networks of traffic flow time series by using visibility graph theory. This approach differs from traditional traffic flow theory as it performs featurc amalysis of traffic flow time series from the perspective of complex networks, and then analyzes the relationship between the characteristics of the structure in the visual network and the state characteristics of the traffic flow. The proposed method al so takes into account the different traffic flow time seque nccs tha t correspond to differe nt t raffic stat cs. In the net work building process using the proposed method, the traffic flow is cla