城市混合交通流微观仿真建模研究

发布时间：2018-03-31 19:31

本文选题：城市混合交通流　切入点：微观仿真系统　出处：《华南理工大学》2014年博士论文

【摘要】：机非混行的混合交通流仍然是我国城市交通特别是中小城市交通的重要特征，混合交通流的微观仿真建模研究能为实际交通现象的真实再现和分析、交通系统的评价等提供有效的技术手段。本文围绕城市混合交通流微观仿真系统的建模、实现与应用，在城市混合交通微观仿真系统结构、仿真路网模型、混合交通两轮车流仿真模型、微观仿真系统设计与实现及在交通信号控制中的应用等方面展开研究与探讨，主要做了以下研究工作：（1）围绕混合交通微观仿真系统的构建，分析了仿真系统的体系结构及组成模块，建立了混合交通微观仿真系统的复合路网模型。针对由机动车、非机动车及行人所组成的城市混合交通特点，复合路网模型将机动车的连续路网模型和非机动车与行人的离散元胞自动机路网模型相结合，建立了二者的映射关系。分析了微观交通仿真系统的交通流模型，采用车道行驶轨迹和车辆动态行驶轨迹描述机动车在车道上的正常行驶和换道行驶行为。（2）针对目前城市交通特别是中小城市交通中混合交通两轮车流的特点，在传统的基于NS元胞自动机（CA）模型基础上，建立了两轮车辆混合交通流二维CA仿真模型。根据两轮车行驶特点建立了两轮车行驶的换道规则，该规则同时考虑了当前时间步和下一时间步两侧相邻车道的前后车辆的状态，使模型更符合实际。通过计算机数值仿真，分析了慢化概率、换道概率、各种两轮车的混合比例、行人穿越两轮车道等因素对两轮车流量、速度和密度等交通特征的影响，对两轮车流的一些交通特征和实际交通现象进行了合理解释，实际验证表明了模型的有效性。（3）针对非机动车行驶特性，提出了速度横向压缩效应概念，建立了基于速度横向压缩效应的非机动车多值CA模型（V-EBCA）。非机动车在行驶时随着车速的增大，出于安全性考虑，其所占用的道路横向宽度会增大，而可供其他车辆通行的道路横向剩余通行宽度会变小，呈现对道路横向剩余通行宽度的压缩效应，考虑这种速度横向压缩效应，采用多值元胞机模型对非机动车流进行建模与仿真，在分析传统多值元胞机模型（BCA和EBCA）基础上，建立了具有最大行进速度为3(cell/s)的高速EBCA模型以及基于速度横向压缩效应的V-EBCA模型，并对各种模型进行了仿真对比分析和验证。（4）围绕城市混合交通流微观仿真系统的设计与开发，对混合交通微观仿真系统进行了详细设计和功能模块设计，采用面向对象方法开发实现了一个小规模微观交通仿真系统。（5）采用所开发的微观交通仿真系统，对实际混合交通路段、定时控制交叉口以及公交优先智能控制交叉口等三种情形下的交通流进行了微观仿真分析。首先，采用基于复合路网模型的混合交通流模型，对实际混合交通路段和定时控制交叉口进行微观仿真，结果表明该系统的仿真效果与实际交通状况基本相符。然后，对于信号交叉口提出了一种公交优先智能控制方法，采用优先度值来衡量相位的通行紧急程度，将优先度值分为容量优先度和阈值优先度两种，容量优先度用于度量相位内的加权车辆数，阈值优先度用于特殊情况下的通行紧急程度，，提出了基于优先度值的公交优先相位选择方法。采用具有论域调节的两级变论域模糊控制器确定绿灯时间，模糊控制器采用粒子群优化的神经网络实现。在微观交通仿真系统中进行仿真测试，结果表明，与定时控制和传统模糊控制方法相比，该方法可以有效提高绿灯时间的利用率，减少交叉口公交停车延误，只要参数选择适当，在提高公交车辆通行效益的同时不会对社会车辆产生过大影响。
[Abstract]:Mixed traffic flow mixed traffic is still the city traffic of our country especially important features of small and medium sized city traffic, true reproduction and the analysis of the microscopic simulation model of mixed traffic flow for the actual traffic phenomena, to provide effective technical means of transportation system. This paper focuses on the modeling and evaluation of microscopic simulation system of city mixed traffic flow. Implementation and application, in the city of mixed traffic micro simulation system, the simulation model of road network, two traffic mixed traffic simulation model, discuss the microscopic simulation system design and implementation and in traffic signal control applications, mainly do the following research work:
(1) the construction of mixed traffic micro simulation system, analyzes the modules and architecture of the simulation system, establishes a complex network model of mixed traffic micro simulation system. According to the vehicle by the city, the mixed traffic characteristics of non motor vehicles and pedestrians are composed of complex network model combining discrete cellular automaton model of vehicle network the continuous network model and non motor vehicles and pedestrians, establish the mapping relationship between the two. Analysis of the traffic flow model of microscopic traffic simulation system, the trajectory of lane and vehicle dynamic driving track description from the normal traffic lane and the lane changing behavior.
(2) for the current city traffic characteristics of mixed traffic flow especially two small city traffic in NS, based on the traditional cellular automaton (CA) model based on a two-dimensional CA simulation model of two wheeled vehicle mixed traffic flow. According to the characteristics of Two Wheeler lane changing rules established two wheel driving, the the rules also considered vehicles before and after the current time step and the next time step on both sides of the adjacent lane state, which makes the model more realistic. Through numerical simulation, analysis of the slow probability, lane changing probability, all kinds of bicycle mixed ratio, pedestrian crossing two lane on the bicycle traffic, affect the speed and the density of traffic characteristics, carry out a reasonable explanation for some of the traffic characteristics of two traffic and actual traffic phenomena, the actual results show that the validity of the model.
(3) according to the characteristics of non motor vehicle, the transverse compression effect of speed is established based on the non motor vehicle lateral velocity compression effect of multi valued CA model (V-EBCA). The non motor vehicle when driving as speed increases, for security reasons, the road width occupied will increase, and for other vehicles on the road transverse residual passage width becomes smaller, showing the compression effect of the transverse residual width of the road traffic, considering the effect of the transverse compression speed, cellular automata model for modeling and Simulation of the non motor vehicle flow with multi value, based on the analysis of traditional multi value cellular automata model (BCA and EBCA) on the basis of established with the maximum speed for the 3 (cell/s) high speed EBCA model and based on the velocity transverse compression effect of V-EBCA model, and the models are simulated and verified.
(4) around the design and development of the microscopic simulation system of urban mixed traffic flow, the detailed design and functional module design of the mixed traffic microscopic simulation system is carried out, and a small-scale microscopic traffic simulation system is developed by object-oriented method.
(5) the microscopic traffic simulation system is developed, the actual mixed traffic intersection, timing control and bus priority intelligent control of the intersection of three cases of microscopic traffic flow simulation analysis. Firstly, the mixed traffic flow model based on complex network model, micro simulation of actual mixed traffic the timing and control intersection, basically consistent with the simulation results show that the effect of this system and the actual traffic condition. Then, the signal intersection presents a bus priority intelligent control method, the priority value of traffic emergency measure phase, the priority value of capacity is divided into priority and threshold priority two. The capacity of priority is used to measure the weighted number of vehicles in the phase, the priority for traffic threshold level of emergency under special circumstances, the priority value of phase selection based on bus priority Choose the method of two grade. Using a variable universe fuzzy controller to determine the green time domain adjustment, fuzzy controller based on neural network and particle swarm optimization. The simulation test, the microscopic traffic simulation system. The results show that compared with the traditional fuzzy control and timing control method, this method can effectively improve the utilization rate of the green time to reduce the delay of intersection, bus stop, as long as the parameters are selected properly, will not increase the traffic efficiency of bus and the impact of social vehicles.

【学位授予单位】：华南理工大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：U491.112

【参考文献】