【作者】 杨洪；

【导师】 刘澜；

【作者基本信息】 西南交通大学 ， 交通工程， 2013， 硕士

【摘要】 随着社会经济高速发展,机动车保有量不断增加,城市道路交通需求增长迅速,道路的交通拥挤问题日益凸显。在此背景下,公安部对全国交警明确提出了利用智能交通系统来解决城市道路拥挤的要求,而城市道路交通状态的研判正是实现智能交通管理与控制的基础。目前我国各大城市的交通信息数据采集系统以及交通诱导信息发布系统建设均已初具规模,但在交通状态判别过程中还存在一些问题亟需完善。基于上述背景,本文结合国内外相关课题研究现状,从交通动态研判概念诠释交通状态研判过程,以基础数据的准确性和研判过程的动态性为目标展开城市交通动态研判应用技术展开系统的研究,对提高交通状态研判精度,完善智能交通实用性有一定的借鉴意义。本文从城市道路交通拥挤成因、度量标准、指标体系构成出发,选择反映交通状态的指标和参数；针对利用时间序列对交通信息数据修复方法的不足提出了基于交通信息数据时间序列与空间位置相似关系的数据修复方法,确保了基础数据的合理性和准确性；利用交通流向的时空关系将预测交通流数据与实测交通流数据的差值作为偏差控制形成动态反馈,提出了基于控制论与时空关系的交通流动态预测模型,实现了交通流预测的动态化,提高了交通流预测的精度；利用历史交通信息数据进行模糊聚类分析确定各交通状态的聚类中心,根据实时交通数据与聚类中心隶属度关系确定实时交通状态,在此过程本文提出了以交通状态聚类中心作为研判阈值的方法,解决了交通状态判别阈值预设的局限性。最后,本文采用VISSIM仿真软件以及Matlab软件完成了案例分析。更多还原

【Abstract】 The traffic congestion has become a increasingly prominent problem with the rapid development of social economy, vehicle ownership increasing continuously and rapid growth of urban road traffic demand. In this context, Ministry of Public Security clearly suggest traffic police to solve the urban traffic congestion by using ITS, while it’s just the basis for intelligent traffic management and control that urban road traffic conditions judgement. Nowadays, traffic information data acquisition system and traffic guidance information release system construction have begun to take shape in major cities of China,however, some problems still exist urgent need to improve traffic conditions identification in process.Based on the above background and the situation on relative subject at home and abroad, it’s starting from the concept to explain traffic conditions judgement process, to research the urban traffic dynamic judgement application technology that it’s aim at the accuracy of basic data and dynamic of judgement process, and it’s certain referential significance to which improve traffic conditions judgement accuracy and Intelligent Transportation practicality.This paper starts from urban road traffic congestion causes, metric standard and constitution of index system to choice indicators and parameters which reflect the traffic conditions; proposed a data recovery method based on the data in the time series and spatial location proportional relationship to improve the lack of only by using time series, ensured the reasonableness and accuracy of basic data; it has proposed a traffic flow dynamic prediction model based on control theory and space-time relationship,which utilizes difference of predicted and actual collection traffic flow data as deviation control to form dynamic feedback through traffic flow time-space relationship, in order to achieve traffic conditions judgement’s dynamic and improve the accuracy of traffic flow forecasting; It has determined the cluster centers of each traffic condition utilizing the historical traffic information data through fuzzy clustering analysis, according to the membership relationship of real-time traffic data and cluster centers to determine the real-time traffic conditions, and this paper propose a method which use traffic conditions cluster centers as the judgement threshold, to solve the limitations of traffic conditions identification default threshold. Finally, simulation software VISSIM and software MATLAB complete case analysis.更多还原