基于车牌识别数据的城市交通流时空特性分析

Analysis of Spatio-Temporal Characteristic in Urban Traffic Flow Based on License Plate Recognition Data

了解城市道路交通流的时空特性有助于提高交通流的预测精度。基于实测的车牌识别数据,运用相似系数、快速傅氏变换和混沌理论分析交通流相似性、周期性和混沌性的时间特性;运用相关系数及互相关函数作为度量标准对交通流空间特性进行描述,论证交通流检测断面之间存在空间相互作用并具有时滞性;根据交通流的空间相关性,使用多维标度法对实际路网中检测断面进行聚类和分组,为交通流预测中的多断面分组以及构建交通流时空预测模型提供理论基础。

To understand the spatial and temporal characteristics of urban road traffic flow will help to improve the prediction accuracy of traffic flow. The authors analyze spatial and temporal characteristics of urban arterial road traffic flow based on license plate recognition data. First, Similarity coefficient, FFT and chaos theory are adopted to analyze the temporal characteristics of traffic flow, such as similarity, periodicity and chaotic property. Second, the spatial characteristics are analyzed by using correlation coefficients and cross- correlation function. It detection sections, and it has time--delay. At proves that spatial interaction has existed between the last, according to the spatial correlation of traffic flow, multidimensional Scaling Method is proposed to divide an actual road network into several smaller ones, which can provide theoretical basis for the clustering analysis of traffic counting points in short--term traffic flow forecasting and building the time--space traffic flow predicting model.