基于DDDAS的高速公路异常事件影响范围仿真分析

发布时间：2018-01-11 04:18

本文关键词：基于DDDAS的高速公路异常事件影响范围仿真分析　出处：《重庆大学》2016年硕士论文　论文类型：学位论文

【摘要】：高速公路异常事件(如车辆故障、交通事故等)会降低路段通行效率,在车流量较大的情况下,可能会引发道路交通阻塞和车辆排队的问题。异常事件的影响范围和发展趋势的可靠估计是制定针对性交通管控策略的前提和基础,对保障高速公路的畅通运行和提高高速公路的管理服务水平具有重要的现实意义。目前高速公路异常事件的影响范围主要是通过交通流理论建立预测模型来进行估计,由于现有交通参数检测精度无法满足模型的输入要求尚难以在工程中进行应用。针对此问题论文引入仿真分析技术,对高速公路交通流时间关联特性进行分析,并结合历史车检器数据特性提出了基于VISSIM仿真系统的交通流参数标定方法和驾驶行为参数校正方法。在此基础上,结合对粒子滤波算法的深入分析,研究了基于DDDAS的高速公路异常事件影响范围仿真分析方法。论文主要内容包括:(1)仿真模型交通流参数标定和驾驶行为参数校正。在对高速公路交通流时间关联特性分析的基础上,结合历史车检器数据对仿真模型交通流参数进行了标定;针对仿真模型驾驶行为参数默认值标定不准确的情况,结合单因素差方法进行敏感性分析确定用于校正的核心参数,研究了基于遗传算法的仿真模型驾驶行为参数校正方法;最后利用实际车检器数据进行了模型验证。结果表明建立的仿真模型能准确的对道路上的交通流运行趋势进行仿真。(2)研究基于粒子滤波算法的交通仿真模型数据同化方法。结合交通波理论和阈值理论,建立高速公路车检器数据预处理方法。在此基础上结合DDDAS范式和粒子滤波理论,研究了基于DDDAS的高速公路异常事件仿真分析方法,最后对模型的有效性进行了算例验证。结果表明,基于粒子滤波的交通仿真模型能够不断地同化实时数据,实现对道路上堵塞事件位置和实时排队长度的精确估计。最后介绍了基于粒子滤波算法的交通仿真系统的设计与实现,并结合G75高速北碚隧道段车检器数据,选取典型真实交通异常事件构建相应的仿真场景,验证了基于DDDAS的高速公路异常事件影响范围仿真系统的有效性。结果表明:本文方法可以准确地对异常事件引起的排队长度进行估计。
[Abstract]:Expressway abnormal events (such as vehicle failures, traffic accidents, etc.) will reduce the efficiency of road sections, in the case of large traffic flow. The problem of road traffic jam and vehicle queuing may be caused. The reliable estimation of the influence range and development trend of abnormal events is the premise and foundation of formulating targeted traffic control strategy. It is of great practical significance to ensure the smooth operation of expressway and to improve the level of management and service of expressway. At present, the influence of abnormal events on expressway is mainly carried out through the establishment of prediction model based on traffic flow theory. Estimate. Because the existing precision of traffic parameter detection can not meet the input requirements of the model, it is difficult to be applied in engineering. In order to solve this problem, the paper introduces simulation analysis technology to analyze the characteristics of time correlation of expressway traffic flow. The calibration method of traffic flow parameters and the method of correcting driving behavior parameters based on VISSIM simulation system are proposed based on the data characteristics of historical vehicle detector. On this basis, the particle filter algorithm is deeply analyzed. This paper studies the simulation and analysis method of the influence range of expressway abnormal events based on DDDAS. The main contents of this paper are as follows: 1). The traffic flow parameters calibration and driving behavior parameters calibration of the simulation model. Based on the analysis of the time correlation characteristics of expressway traffic flow. The traffic flow parameters of the simulation model are calibrated with the historical vehicle detector data. In view of the inaccurate calibration of the default values of driving behavior parameters in the simulation model, the core parameters for correction are determined by sensitivity analysis combined with the single factor difference method. The driving behavior parameters correction method of simulation model based on genetic algorithm is studied. Finally, the model is verified by using the actual vehicle detector data. The results show that the established simulation model can accurately simulate the traffic flow running trend on the road. The data assimilation method of traffic simulation model based on particle filter algorithm is studied. The traffic wave theory and threshold theory are combined. Based on the DDDAS normal form and particle filter theory, the simulation analysis method of highway abnormal events based on DDDAS is studied. Finally, the validity of the model is verified by an example. The results show that the traffic simulation model based on particle filter can assimilate real-time data continuously. Finally, the design and implementation of traffic simulation system based on particle filter algorithm are introduced. And combined with the G75 high-speed Beibei tunnel section vehicle detector data, select typical real traffic anomalies to build the corresponding simulation scene. The effectiveness of the simulation system based on DDDAS is verified. The results show that the proposed method can accurately estimate the queue length caused by abnormal events.
【学位授予单位】：重庆大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：U491

【相似文献】