考虑排队动态演变过程的多相位信号交叉口最小交通延迟研究

发布时间：2017-12-28 12:24

本文关键词：考虑排队动态演变过程的多相位信号交叉口最小交通延迟研究　出处：《中国科学技术大学》2017年博士论文　论文类型：学位论文

【摘要】：交通流理论的研究现象主要是车辆交通及其引起的交通堵塞现象。当很多人以各种出行目并以各种出行方式同时出现在某个交叉口时,交通堵塞就会随之出现。为了改进道路安全并缓解交通堵塞,交通管理部门引入了交通信号灯,而信号灯的概念是来自铁路管理技术中的控制方法。由于未能很好地理解汽车交通的内部运行方式,信号灯控制的效果并不明显。在20世纪中叶,一些著名科学家的研究工作奠定了汽车交通理论研究的基础。交通流模型应该建立在对交通流物理特性的理性考虑之上,因为模型的主旨就是要描述真实交通流现象。交通流理论,不仅需要模拟交通运动,而且需要模拟城市交叉口的运动特性。通过模拟城市交叉口,人们可以通过某些装置(如交通信号灯)或者某些方法(如管理通过控制点的交通流量)来改进交叉口的通行能力。本文在前人研究的基础上,解析分析了多相位信号交叉口的最小交通延迟,并提出了考虑排队动态演变过程的两相位十字信号交叉口的元胞自动机模型,本文的主要工作如下:在考虑了排队动态演变的基础上,解析研究了两相位信号交叉口的最小交通延迟。延迟函数受到两个约束条件的影响,当两个约束条件至少有一个取等号时,交通延迟最低。我们给出了最小延迟及相应的交通信号周期的解析表达式,解析结果表明,在两种情况下,系统平均延迟最低:第一种情况是一个相位需要额外的绿灯时间,另一种情况是两个相位都不需要额外的绿灯时间。我们的研究工作说明,虽然很多实验结果都显示Newell的"清除策略"的交通延迟最低,但或许还有更好的方法。在两相位信号交叉口解析研究的基础上,我们考虑了排队动态演变过程,通过建立连续方程,解析研究了三相位T型信号交叉口的最小交通延迟。我们得到了三相位T型信号交叉口的延迟函数,并考虑了三个约束条件。三个约束条件组成了一个三维空间,该三维空间由三个平面构成,通过研宄延迟函数的性质,我们首先确定了最小延迟一定在三个平面上得到。三个平面彼此两两相交,因此有三条边界,我们接下来证明了最小延迟一定在三条边界上取得。我们通过分析延迟函数在三条边界上的性质,最后得知最小延迟一定在两个位置取得:其中一个位置是三条边界的交点处,另一个位置是三条边界中的其中一条边界上·。如果最小延迟在第一个位置取得,那么在这种情况中,三条入流车道都不需要额外的绿灯时间。如果最小延迟在第二个位置得到,那么三条入流车道中入流量最大的车流需要额外的绿灯时间。我们考虑到现实交通应用的需要,用元胞自动机模型模拟了两单向单车道两相位十字信号交叉口,模型分别采用了固定周期策略与排队完全疏散自适应策略。固定周期策略模拟结果表明,当两车道入流量相等时,模拟结果与解析结果吻合度较高,当分配给两条车道的绿灯时间相等时,系统平均延迟最低。当两条车道入流量不相等时,模拟得到的平均延迟高于解析结果。当系统采用自适应策略时,无论两条车道的入流量是相等还是不相等,模拟得到的平均延迟均低于解析结果。模拟结果说明,两单向单车道两相位十字信号交叉口中,系统流量越高,交通延迟越高,同时自适应策略优于固定周期策略。我们接下来用元胞自动机模拟了两双向双车道两相位十字信号交叉口,信号控制策略同样采用了固定周期策略与交通自适应策略。模拟结果表明,当采用固定周期策略时,模拟得到的平均延迟与解析结果之间的大小关系与系统流量有关,如果系统流量较低,模拟得到的平均延迟低于解析结果,如果系统流量较高,解析得到的平均延迟低于模拟结果。当系统采用自适应策略时,如果东西向与西东向的入流量相等,或者南北向与北南向的入流量相等,模拟得到的排队长度大于解析得到的排队长度,解析得到的平均延迟低于模拟结果。但当东西向与西东向的入流量不等并且南北向与北南向的入流量不等时,模拟得到的平均延迟低于解析结果。本文解析分析了两相位与三相位信号交叉口的最小交通延迟,得到了最小延迟与相应的信号周期的表达式,同时用元胞自动机模型模拟了两单向单车道与两双向双车道十字信号交叉口,定性比较了固定周期策略与排队完全疏散自适应策略之间的优劣。我们的研究工作不仅能对缓解城市交叉口交通拥堵现象提供理论指导,而且还能推动交通信号控制系统的发展。
[Abstract]:Traffic flow theory is mainly the phenomenon of traffic congestion and traffic congestion. Traffic jams will occur when many people travel at a certain intersection with a variety of trips and travel in a variety of ways. In order to improve road safety and relieve traffic jam, traffic management department introduced traffic lights, and the concept of signal lamp came from the control method in railway management technology. Because of the failure to understand the internal operation of car traffic, the effect of signal control is not obvious. In the middle of the twentieth Century, the research work of some famous scientists laid the foundation of the research on the theory of automobile transportation. The traffic flow model should be based on the rational consideration of the physical characteristics of the traffic flow, because the main purpose of the model is to describe the real traffic flow phenomenon. The theory of traffic flow not only needs to simulate the traffic movement, but also needs to simulate the motion characteristics of the urban intersection. Through simulating the urban intersection, people can improve the capacity of intersection by means of some devices, such as traffic lights or some methods, such as traffic flow through control points. Based on the previous research, the analysis of the minimum traffic delay multi-phase signal intersection, and proposed two phase signalized intersection of cellular automaton model queuing the dynamic evolution process of consideration, the main work is as follows: Based on the dynamic evolution of the queue, we study the minimum traffic two phase signalized intersection delay. The delay function is affected by two constraints. When the two constraints have at least one equal number, the traffic delay is the lowest. We derive the analytical expressions of the minimum delay and traffic signal cycle corresponding, the results show that in the two case, the minimum average delay system: the first one is an additional phase green time, another is the two phase without the need for extra green time. Our research work shows that although many of the experimental results show that the "clear strategy" of Newell has the lowest traffic delay, there may be a better way to do it. Based on the analytical study of the two phase signalized intersection, we consider the dynamic evolution process of queues, and establish the continuous equation to analyze the minimum traffic delay of the three phase T signalized intersection. We get the delay function of the three phase T type signal intersection and consider three constraints. The three constraint conditions to form a three-dimensional space, the three-dimensional space composed of three plane structure, through the research on the nature of the delay function, we must first determine the minimum delay in the three plane. The three planes intersected each other 22, so there are three boundaries. Then we prove that the minimum delay must be obtained on the three boundary. By analyzing the properties of the delay function on the three boundaries, we know that the minimum delay must be obtained in two locations: one is the intersection of three boundaries, the other is one of three boundaries. If the minimum delay is obtained at the first position, in this case, no additional green light time is required for the three entry lanes. If the minimum delay is obtained in second locations, the maximum traffic flow in the three entry lanes requires extra green time. Considering the need of real traffic applications, we simulate the two one-way single lane, two phase cross signalized intersection with cellular automata model. The model adopts fixed cycle strategy and queuing full evacuation adaptive strategy respectively. The simulation results of fixed cycle strategy show that when the two lane traffic flow is equal, the simulation results coincide with the analytical results. When the green light time allocated to two lanes is equal, the average delay of the system is the lowest. When the inflow of two lanes is not equal, the average delay of the simulation is higher than that of the analytical result. When the adaptive strategy is adopted, no matter whether the flow of two lanes is equal or unequal, the average delay of the simulation is lower than that of the analytical result. The simulation results show that the higher the system traffic and the higher the traffic delay, the better the adaptive strategy is better than the fixed period strategy in the two one-way single lane two phase cross signalized intersection. Next, we use cellular automata to simulate the two-way two-way intersection with two phases and two phases. The signal control strategy also adopts the fixed period strategy and the traffic adaptive strategy. The simulation results show that, when using a fixed cycle strategy, and the relationship between the size of the average delay analysis of the simulation results and system flow, if the system flow is low, lower than the average delay of analytical results obtained by simulation, if the higher flow, lower than the average delay of the simulation results. When using the adaptive strategy system, if east and East and west to flow into the same, or from north to South and North South flow is equal, the simulated queueing length greater than the length of the analysis, the simulation results is lower than the average delay of the. But when things to flow into the north and South and East and west to the range and to the South and North into the flow range, lower than the average delay of analytical results obtained by simulation. In this paper, the analysis of the minimum traffic delay of two phase and three phase signalized intersection, the minimum delay expression and corresponding signal cycle, and the cellular automaton model is used to simulate two and two lane one-way two lane cross intersections, qualitative comparison between fixed cycle strategy and adaptive strategies between the complete evacuation line the advantages and disadvantages. Our research work can not only provide theoretical guidance to alleviate traffic congestion in urban intersection, but also promote the development of traffic signal control system.
【学位授予单位】：中国科学技术大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：U491

【参考文献】