【作者】 周申培；

【导师】 严新平；

【作者基本信息】 武汉理工大学 ， 交通信息工程及控制， 2009， 博士

【摘要】 交通是现代社会的基础,是人类社会经济的命脉,人们的社会行为与交通息息相关。在我国,由于机动车保有量迅速增加、交通设施建设滞后以及管理措施不够完善等原因,交叉路口的交通堵塞现象日趋严重,从而影响到城市路网通行能力的发挥。车辆在交叉路口处反复地分流、合流及交叉,交通状况复杂,使得交叉路口已经成为制约城市道路交通功能的瓶颈。维持城市交通正常运转,需要加强交通控制与管理,积极开展对交叉路口的研究,努力提高交叉路口的通行能力。交通信号控制是运输网络的一种基本管理手段,随着智能运输系统(Intelligent Transponation System,ITS)的发展,交通信号控制的智能化研究越来越受到重视。到目前为止,已经研究了许多交通信号控制模型。但是目前这些研究建立的模型大多数都是针对交通流的,没有将其他因素考虑在内。随着城市交通和社会经济的日益发展,汽车的保有量不断创下新高,大量机动车的出现在造成严重的交通拥挤和堵塞的同时,也引起了严重的空气污染和噪声污染。据权威统计分析,主要大城市大气污染物中机动车排放分担率呈明显上升趋势。如何在发展经济的同时,建立一个人、车、路、环境和谐发展的交通体系,这个问题已经被认为是全球最严重也是最具挑战性的课题之一。因此建立一个同时兼顾交通流情况和机动车尾气排放情况的交通信号控制模型具有重要的理论意义和实用价值。论文总结了国内外城市交通信号控制的研究进展,分析了该领域研究的发展趋势。在此基础上,针对城市交通信号优化控制问题进行了智能优化的理论分析和应用方法的研究。提出了考虑排放因素的城市交通信号优化控制系统的架构、建模方法和求解算法。通过对城市信号控制路口的交通流分析,从单点两相位、单点多相位和干线协调控制的不同层面,建立了城市交通信号优化控制的双层多目标规划模型。采用遗传算法和遗传蚁群融合算法对模型进行了仿真试验,通过与传统仿真试验结果进行比较,证明了这两种优化算法的优越性,其中,遗传蚁群融合算法优化性能更好。本论文的主要研究工作总结如下:(1)设计了考虑排放因素的交通信号优化控制系统框架。通过对城市交叉口交通信号控制发展现状和趋势的分析,将双层规划模型的研究方法与考虑排放因素的交通信号配时控制联系起来进行研究,设计了交通信号优化控制系统的架构,指出信号优化模块是交通信号优化控制系统的核心,其实质是交通信号控制模型及其求解算法。(2)从单点控制和协调控制的角度,建立了城市交通信号控制的双层多目标规划模型。以双层规划模型为工具建立了交通信号控制模型,分别对上下层模型进行详细的问题描述和数学表达。其中,路网中的机动车尾气排放量作为一个优化目标,嵌入上层模型的目标函数中。模型以路网中的车辆延误和尾气排放总量为性能指标,通过改变信号控制策略,在保证交通基本畅通的前提下,将城市机动车尾气排放量限制在一定范围内。(3)提出并实现了求解函数优化问题的遗传蚁群融合算法。遗传蚁群融合算法的基本思路是首先采用遗传算法产生有关问题的初始信息素分布;然后采用蚂蚁算法,在有一定初始信息素分布的情况下,充分利用蚂蚁算法的并行性、正反馈机制以及求解效率高等特性,高速高效地得到函数的优化解。本文以Camel函数作为测试函数,同时还以单点两相位信号优化控制模型的求解为例,通过与传统优化方法的仿真试验进行比较,采用遗传蚁群融合算法可获得最优性能指标,证明该算法是一种时间效率和求解效率都比较好的求解函数优化问题的有效算法。(4)设计并实现了基于启发式遗传算法的模型求解方法。在利用惩罚策略处理双层多目标规划模型约束条件的基础上,针对具体建立的交通信号控制模型,设计了基于启发式遗传算法的模型求解方法,并运用MATLAB软件对所设计的城市单交叉口交通信号控制的双层多目标规划模型进行实时仿真,仿真表明能达到良好的控制效果。(5)完成典型城市路网交通流情况的仿真。利用VISSIM交通仿真软件对典型城市路网的交通流情况进行仿真运行,得出了在固定信号配时方案下各项模拟输出的结果。其结果一方面可以作为参数代入典型城市路网交通信号控制的双层多目标规划模型,另一方面可以作为比较值验证以上两种优化算法的优化性能。更多还原

【Abstract】 Traffic is the foundation and the economic lifelines of the modern society. In China, the quantity of vehicles increases rapidly, at the same time, traffic infrastructure and management methods have some shortages. As a result, the traffic congestions in intersections become more and more serious. It makes intersections have become bottlenecks of the urban traffic. Traffic signal is an essential element to manage the transportation network. Nevertheless, it is widely accepted that the benefits of traffic control signal systems are not being fully practiced. Along with the development of Intelligent Transportation System (ITS), the research on traffic signal control is by no means complete and traffic signal control remains one of the most heavily funded research and development items. A number of models of the traffic signal control have been developed in the past. However, those models mainly focus on the traffic flow and don’t take other factors into account.With the development of urban transportation, large numbers of vehicles cause serious air pollution and noise pollution. The authoritative statistics shows that the share rate of vehicles exhaust emission to atmosphere has an obvious increasing tendency. This problem, how to establish an harmonious traffic systems of human, vehicles, roads and environment, has been recognized as one of the most critical, yet challenging problems for the world. Therefore, it has practical significance to establish a new traffic signal control model based on the condition of the traffic flow and vehicle exhaust emissions.In this dissertation, the research of theoretical analysis and application methods are given on base of summarizing the domestic and overseas research progress and analyzing the development trends of urban traffic signal control. The principle of the traffic signal control by intelligent optimization, mainly the basic optimal theories and the methods of Genetic Algorithm as well as Ant Colony Algorithm are proposed firstly. Through the study and the analysis of traffic streams in an intersection, a bi-level multi-object optimization model of traffic signal control in an intersection is established. The simulation tests are conducted using Genetic Algorithm and The Fusion Algorithm of Genetic and Ant Colony. The simulation results show that the optimal algorithms are superior to the traditional methods, and the fusion algorithm is most suitable for determining the green split in a single intersection.The main achievements of the dissertation are as follow:(1) A bi-level multi-object model is established for optimizing the signal cycle length and green time by considering the constraint of automotive exhaust emission. The performance index function for optimization is defined to improve traffic quality and reduce emission at intersections. The research tries to limit the range of vehicle exhaust emissions on the premise of unimpeded transport by changing the traffic signal control strategy.(2) The heuristic genetic algorithm is designed to solve the problem, simplifying the model by means of penalty strategy. Subsequently MATLAB program is given to simulate the solution process. The simulation results show that very nice effects are obtained.(3) The fusion algorithm of genetic and Ant Colony applied in solving the function optimization problem is presented. By comparing with other solution methods, the fusion algorithm has the best performance.(4) The typical topological structure of urban road network is designed by combining the features of traffic signal control with the study perspective of this research.(5) Traffic simulation system (VISSIM) is used to simulate the situation of traffic flow in the typical urban road network and gains the results on the condition of fix signal timing plan. The function of the results is substituting into the model as parameters and verifying the optimization effect of above algorithms.更多还原