基于时延赋色Petri网的车辆旅行时间预测与分析-软件工程专业论文.docx

Modeling and analyses of travel time estimation using timed colored Petri nets A Thesis Submitted for the Degree of Master Candidate：Lei Tian Supervisor：Prof. An YiSheng Chang’an University, Xi’an, China i i PAGE PAGE iv 摘 要 先进的出行者信息系统已经成为国内外交通运输领域的研究热点，它不但能够有效 地控制和协调路网上的车流，还能准确的预测车辆旅行时间，为人们的出行提供了极大 的方便。本文在充分考虑城市主干线路网及交通流特性的基础上，通过分析并对比目前 国内外研究现状，利用时延赋色 Petri 网对车辆旅行时间预测进行建模与分析，主要内 容有： 1.通过分析已有车辆旅行时间预测研究的成果，对传统的车辆旅行时间预测计算方 法进行了介绍。针对城市交通系统的分布、并发特征，采用时延赋色 Petri 网对城市交 通系统进行模块化、分层次的建模，分别建立了路段、交叉口和信号控制三个子模型。 2.针对城市路网中交叉口车辆通行效率低下，交通信号控制策略难于满足输入路段 上车流量变化的问题，提出了一种基于 TCPN 的交叉口信号控制优化方法，该方法以交 叉口各输入路段车辆数最小化为目标，通过检测路段总车辆数，实现自适应信号控制， 并结合 Matlab 软件对优化算法进行了求解。 3.在城市路网 TCPN 模型和信号控制优化方法的基础上，利用 CPN ML 语言编写 TCPN 模型的数据、状态监视器获取模型运行数据，提出了基于车辆紧迫程度和车流密 度的模糊推理方法，实现了对自由旅行路段上车辆平均速度的估计，从而对车辆旅行时 间进行预测，并结合 Matlab 软件对该过程进行仿真和分析。 关键词：出行者信息系统，车辆旅行时间，时延赋色 Petri 网，模糊推理，信号控 制 Abstract Advanced Traveler Information System, ATIS, has become a research focus at home and abroad in the field of transportation. It can not only predict the travel time of the vehicle accurately, but also control and coordinate the traffic flow on the road network efficiently. As a result, it brings much convenience to the travelers. The thesis has taken a full consideration of the features that the urban trunk road and the traffic flow show. On the basis of it, the thesis built a model to predict the travel time of the vehicle using the time colored Petri nets by analyzing and comparing the current domestic and overseas research status. The main contents are: 1．The thesis gave an introduction of the traditional algorithms on the prediction of the travel time of the vehicle through analyzing the existing research achievements. Aiming at the distribution and concurrence features of the urban traffic system, the thesis choose the time colored Petri nets to build the model of it modularly hierarchically, including the road submodel, intersection submodel and the signal control submodel. 2．In the urban road network, the eff