拥挤网络交叉口多目标配时优化研究
发布时间：2019-04-11 16:11
【摘要】：交叉口信号配时优化作为城市交通信号控制系统的基础环节,也是交通控制管理中的重要内容,对于缓解拥堵、规范秩序、减少污染、提高运行效率等交通问题的解决有着重大意义。本文以我国大量混合、拥挤的交通环境为背景进行了单点交叉口和交通网络信号配时优化设计,主要做了以下几方面工作：1.总结了交通信号控制参数、交叉口性能评价指标以及交通信号控制方案,介绍了多目标规划原理、双层规划原理以及用户平衡网络分配模型。2.以混行拥挤单点交叉口为研究对象,综合考虑机动车和行人两种交通主体的通行效益,以机动车平均延误、行人平均延误以及通行能力为指标,以饱和度为约束,建立了单点交叉口多目标配时优化模型。在机动车平均延误指标的计算中,将其分为机动车驶过交叉口的固有延误和行人违章引起的机动车延误两部分处理。设计了遗传算法求解多目标配时优化模型,以重庆市典型交叉口为例,实地调查数据并带入模型进行优化,验证了改进模型的有效性。3.针对拥挤交通网络,建立了流量分配与信号配时相结合的双层规划模型。模型下层为UE均衡网络模型,上层为以路网总运行时间最小、停车次数最少为优化目标的多目标模型。设计了启发式算法与遗传算法求解模型。以“田”字型网络为算例,对网络中各交叉口进行综合配时优化,优化结果显示,路网总延误与停车次数均得到有效减少。
[Abstract]:As the basic link of urban traffic signal control system, intersection signal timing optimization is also an important content of traffic control management. It can alleviate congestion, regulate order and reduce pollution. It is of great significance to solve traffic problems such as improving operation efficiency. In this paper, based on a large number of mixed and crowded traffic environment in our country, the optimal design of single-point intersection and traffic network signal timing is carried out, and the main work is as follows: 1. The traffic signal control parameters, intersection performance evaluation index and traffic signal control scheme are summarized. The principle of multi-objective programming, double-level programming and user balanced network assignment model are introduced. 2. Taking the mixed crowded single-point intersection as the research object, considering the traffic benefits of the vehicle and pedestrians, taking the vehicle average delay, the pedestrian average delay and the traffic capacity as the index, and taking the saturation as the constraint, the traffic efficiency of the vehicle and pedestrians is considered. The multi-objective timing optimization model of single-point intersection is established. In the calculation of motor vehicle average delay index, it is divided into two parts: the inherent delay of motor vehicle passing the intersection and the vehicle delay caused by pedestrian violation. The genetic algorithm is designed to solve the multi-objective timing optimization model. Taking the typical intersection in Chongqing as an example, the field survey data and the model are brought into the optimization model to verify the effectiveness of the improved model. 3. A bilevel programming model combining traffic assignment and signal timing is established for traffic congestion networks. The lower layer of the model is the UE equilibrium network model, and the upper layer is a multi-objective model with the minimum total running time of the road network and the minimum number of parking times as the optimization objectives. Heuristic algorithm and genetic algorithm are designed to solve the model. Taking the "Tian" shape network as an example, the integrated time allocation optimization is carried out at each intersection in the network. The optimization results show that the total delay and parking times of the network are effectively reduced.
【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U491.54
本文编号：2456564
[Abstract]:As the basic link of urban traffic signal control system, intersection signal timing optimization is also an important content of traffic control management. It can alleviate congestion, regulate order and reduce pollution. It is of great significance to solve traffic problems such as improving operation efficiency. In this paper, based on a large number of mixed and crowded traffic environment in our country, the optimal design of single-point intersection and traffic network signal timing is carried out, and the main work is as follows: 1. The traffic signal control parameters, intersection performance evaluation index and traffic signal control scheme are summarized. The principle of multi-objective programming, double-level programming and user balanced network assignment model are introduced. 2. Taking the mixed crowded single-point intersection as the research object, considering the traffic benefits of the vehicle and pedestrians, taking the vehicle average delay, the pedestrian average delay and the traffic capacity as the index, and taking the saturation as the constraint, the traffic efficiency of the vehicle and pedestrians is considered. The multi-objective timing optimization model of single-point intersection is established. In the calculation of motor vehicle average delay index, it is divided into two parts: the inherent delay of motor vehicle passing the intersection and the vehicle delay caused by pedestrian violation. The genetic algorithm is designed to solve the multi-objective timing optimization model. Taking the typical intersection in Chongqing as an example, the field survey data and the model are brought into the optimization model to verify the effectiveness of the improved model. 3. A bilevel programming model combining traffic assignment and signal timing is established for traffic congestion networks. The lower layer of the model is the UE equilibrium network model, and the upper layer is a multi-objective model with the minimum total running time of the road network and the minimum number of parking times as the optimization objectives. Heuristic algorithm and genetic algorithm are designed to solve the model. Taking the "Tian" shape network as an example, the integrated time allocation optimization is carried out at each intersection in the network. The optimization results show that the total delay and parking times of the network are effectively reduced.
【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U491.54
【参考文献】
相关期刊论文 前4条
1 尹宏宾,徐建闽,周其节;基于知识的信号交叉口配时设计系统[J];华南理工大学学报(自然科学版);1999年08期
2 孙超;徐建闽;;城市单点交叉口的信号配时优化研究[J];交通与计算机;2008年06期
3 张凌煊;祝进城;帅斌;;考虑行人效益的拥挤交叉口多目标配时优化[J];计算机工程与应用;2015年16期
4 四兵锋;孙壮志;赵小梅;;基于随机用户平衡的混合交通网络流量分离模型[J];中国公路学报;2006年01期
，本文编号：2456564
本文链接：https://www.wllwen.com/kejilunwen/daoluqiaoliang/2456564.html
