基于MFD的城市路网双层路径诱导策略研究

发布时间：2018-09-03 19:17

【摘要】：交通流诱导是缓解城市交通拥堵问题的有效途径,它可分为中心式诱导和分布式诱导。中心式诱导从交通管理者的角度出发,以系统最优为目标,以期达到路网整体运营效率的最大化,但大规模的路网、海量的数据、复杂的计算使中心端和通信带宽不堪重负,难以满足动态诱导实时性要求;分布式诱导从出行者的角度出发,以用户最优为准则,以期实现个体出行效益的最大化,然而当一定数量出行者为躲避拥堵集中驶向当前不拥挤路段时,往往又会诱发新的拥堵,产生拥挤转移现象。因此,本文将两者相结合,优势互补,提出一种双层路径诱导策略。上层以诱导子区为基本单元进行系统最优动态交通分配,生成诱导子区级的出行序列;下层在限定的诱导子区范围内搜索并生成用户最优条件下的诱导路径。该策略有效地弥补了中心式诱导和分布式诱导各自的不足,是完善交通流诱导系统的一种有益的尝试。MFD理论作为研究路网宏观交通特性的一种手段,贯穿全文的研究内容,它既提供了一种分析、评价路网交通运行水平的方法,又为本文双层路径诱导策略的提出奠定了理论基础。本文的研究工作主要体现在:(1)通过变量的重新选取,构建了一种表达路网运行能力和运行车辆数之间关系的MFD模型,分析了变量选取的合理性和实际意义,并基于浮动车数据提出了城市路网MFD的获取方法。(2)确立了基于MFD的城市路网诱导子区划分原则和依据,提出一种多步骤划分方法。该划分方法结合了初始划分和二次划分,分别应用特点不同的mean shift聚类算法和层次聚类算法实现其划分过程。(3)提出一种基于MFD的城市路网双层路径诱导策略,它同时考虑了交通管理者和交通出行者的意愿,是中心式诱导和分布式诱导的结合,是系统最优和用户最优的折中。在上层诱导中,基于MFD构建诱导子区交通流动态演变模型,并提出诱导子区级系统最优动态交通分配方法;在下层诱导中,通过求解用户最优路径问题实现车辆路径优化。
[Abstract]:Traffic flow guidance is an effective way to alleviate urban traffic congestion. It can be divided into central guidance and distributed guidance. From the point of view of traffic manager, the central guidance aims at the optimization of the system, in order to maximize the overall operational efficiency of the network, but the massive network, the massive data, the complex calculation make the center end and the communication bandwidth unbearable. It is difficult to meet the real-time requirement of dynamic induction. However, when a certain number of travelers concentrate to avoid congestion to the current uncongested road, it will often induce new congestion, resulting in congestion transfer phenomenon. Therefore, a two-layer path guidance strategy is proposed by combining the two methods and complementing each other. The upper layer uses the inductive sub-region as the basic unit to carry out the optimal dynamic traffic assignment and generates the trip sequence at the induced sub-area level; the lower layer searches and generates the induced path under the user optimal condition in the limited induced sub-area. This strategy effectively makes up for the shortcomings of central guidance and distributed guidance. It is a useful attempt to improve the traffic flow guidance system. As a means of studying the macroscopic traffic characteristics of road network, this strategy runs through the research content of this paper. It not only provides an analysis method to evaluate the traffic operation level of the road network, but also lays a theoretical foundation for the two-layer route guidance strategy proposed in this paper. The main works of this paper are as follows: (1) through the re-selection of variables, a MFD model is constructed to express the relationship between the network running capacity and the number of running vehicles, and the rationality and practical significance of variable selection are analyzed. Based on the floating vehicle data, a method of obtaining urban road network MFD is proposed. (2) the principle and basis of urban road network induced sub-area division based on MFD are established, and a multi-step partition method is proposed. This method combines the initial partition and the quadratic partition, and uses the mean shift clustering algorithm and hierarchical clustering algorithm with different characteristics to realize the partition process. (3) A two-layer route guidance strategy based on MFD is proposed for the urban road network. It takes into account the wishes of both traffic managers and travelers. It is a combination of central guidance and distributed guidance and a compromise between system optimization and user optimization. In the upper layer guidance, the traffic flow evolution model of the induced sub-area is constructed based on MFD, and the optimal dynamic traffic assignment method of the induced sub-area system is proposed. In the lower layer guidance, the vehicle routing optimization is realized by solving the user optimal routing problem.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U491

【相似文献】