城市交通干线的协调优化控制方法及仿真研究

发布时间：2018-04-27 07:24

本文选题：交通工程 + 干线协调　；参考：《上海应用技术大学》2017年硕士论文

【摘要】：城市交通拥堵已经成为困扰城市发展与人民生活的难题,“治堵”已经成为城市工作中重要内容。城市交通信号控制的合理配时,是更好的利用城市道路时空资源的重要手段,随着交通拥堵的加剧,对信号配时的精细化、合理化要求更加提高。城市交通中,由于相邻交叉口之间往往相互关联、相互影响,只关注某一个交叉口的交通控制是不能够解决城市主干道的交通问题的。合理分配交叉口的信号周期、绿信比、相位差是提高交通信号控制系统效益的方法。由于受限于交通流组成、道路行驶条件、交通信号配时方案等限制,想要众多车流得到良好的效果并不是一件容易的事情。协调控制的效果是由交通流组成、周期、绿信比、行程时间以及相位相序安排众多因素联合起作用的。以前大部分的研究都是基于各交叉口相位相序不变的前提下进行优化,因为将相位相序的优化引入到协调控制中,会较大的增加优化与求解难度。不变的相位相序限制了协调控制方案的解空间范围,限制了交通信号控制系统最大发挥作用。为了提高更多的干线协调控制系统效益,本文设计了基于信号相位相序调整的干线协调控制策略,通过调整干线协调控制交叉口的相位相序,扩大协调控制基本参数的求解范围,从而发挥协调控制模型的最大效益。首先,本文重点设计了干线协调控制系统结构,提出了系统的控制原理;其次,对协调控制基本参数信号周期、绿信比、相位差的优化方法分析研究,在此基础上分析了相位相序对协调控制的制约关系;再次,本文提出了协调交叉口冲突策略,采用二进制形式对相位序列进行组合编码,建立基于车辆平均延误最小的相序模型,并采用多种群遗传算法求解最优的相位相序组合方案,基于相位相序,优化调整各交叉口的基本参数设置。最后,对石河子市北三路这一协调控制干线的交叉口进行实例研究。与组合遍历求解相位相序法对比,仿真结果表明,多种群遗传算法能够提高相位相序优化的有效性;采用VISSIM软件建立路网模型进行仿真,比较现状干线交叉口的控制方案,对车辆的平均延误时间进行对比,例证了研究方法的可行性。
[Abstract]:Urban traffic congestion has become a difficult problem for urban development and people's life, and "governance and plugging" has become an important part of urban work. The reasonable timing of urban traffic signal control is an important means to make better use of the space-time resources of urban roads. With the aggravation of traffic congestion, the signal timing is refined and rationalized. In urban traffic, because the adjacent intersections are often interrelated and influence each other, it is not possible to solve the traffic problems of the urban main roads by only paying attention to the traffic control of a certain intersection. The reasonable allocation of signal cycle, green signal ratio and phase difference are the methods to improve the efficiency of traffic signal control system. Due to the limitation of traffic flow composition, road driving conditions, traffic signal timing scheme and so on, it is not easy to get good results from many traffic flows. The effect of coordinated control is composed of traffic flow, cycle, green signal ratio, travel time and phase sequence. Most of the previous studies were based on the premise that the phase sequence of the intersection was invariant, because the introduction of the phase sequence optimization into the coordination control will increase the difficulty of optimization and solution. The invariant phase sequence limits the solution space of the coordinated control scheme and limits the maximum function of the traffic signal control system. In order to improve the efficiency of the trunk line coordinated control system, this paper designs a coordinated trunk line control strategy based on the signal phase phase sequence adjustment, and adjusts the phase sequence of the trunk line coordinated control intersection by adjusting the phase sequence of the trunk line coordinated control intersection. In order to maximize the benefit of the coordinated control model, the solution range of the basic parameters of coordinated control is expanded. Firstly, the structure of the coordinated control system is designed, and the control principle of the system is put forward. Secondly, the optimization method of signal cycle, green signal ratio and phase difference of coordinated control is analyzed and studied. On the basis of this, the restriction relation of phase sequence to coordinated control is analyzed. Thirdly, the conflict strategy of coordinated intersection is proposed in this paper, and the phase sequence is coded in binary form. A phase sequence model based on the minimum average vehicle delay is established, and the optimal phase sequence combination scheme is solved by using multi-population genetic algorithm. Based on the phase phase sequence, the basic parameters of each intersection are optimized and adjusted. Finally, a case study on the intersection of the North third Road, a coordinated control trunk line in Shihezi City, is carried out. The simulation results show that the multi-population genetic algorithm can improve the efficiency of phase sequence optimization, and use VISSIM software to establish road network model to simulate, compare the current control schemes of trunk intersection, the simulation results show that multi-population genetic algorithm can improve the efficiency of phase sequence optimization. The feasibility of the research method is illustrated by comparing the average delay time of vehicles.
【学位授予单位】：上海应用技术大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U491.54

【参考文献】