城市快速路入口匝道流量控制研究

发布时间：2018-05-04 23:08

本文选题：城市快速路 + 短时交通流预测　；参考：《西南交通大学》2016年硕士论文

【摘要】：在城市经济与城市建设的高速发展需求下,道路供给资源不足与不合理的交通管理造成了城市快速路的拥堵和利用率低下。由于城市用地的不足,解决快速路交通拥堵不能只靠盲目的通过扩大道路设施规模来增加通行能力,更重要是充分利用通行能力和相关服务设施。匝道控制无论是在理论与仿真层面还是在实际各大城市的实例应用中,都已被证实是解决快速路效率问题最见成效的管理方式之一。合理的匝道控制对于充分利用快速路通行能力,改善交通条件,显著预防其交通堵塞,协调快速路与关联道路的交通组织,提高道路的服务水平起着关键的作用。通过分析了城市快速路的基本特性以及快速路匝道的组合形式,为交通仿真以及交通流分析提供依据。接着对匝道基础控制原理进行分析,提出了匝道控制的各项评价指标,为控制效果的提供依据。同时分析了不同入口匝道的控制策略与方法并进行分类,为建立自适应匝道控制提供理论支持。对于城市快速路入口匝道的准确的短时流量预测能够对提前预判交通环境,为交通主动控制、改善交通环境提供支持。由于短时交通流具有非线性、不稳定性等特征,本文建立了基于遗传算法优化的小波神经网络,与普通的神经网络以及以上一个周期流量代替本周期流量的预测方法相比,具有较强的学习能力和避免陷入局部最优的特点,并通过实例验证了其预测的准确性。在对快速路入口匝道不同的控制策略与方法的分析基础上,针对现有控制方法的存在三个主要问题：入口匝道信号控制与主线可插入车头时距不足造成匝道排队长度回溢、排队快速清理对快速路主线造成的不良影响以及控制滞后的问题,结合短时交通流预测,提出了综合考虑快速路利用效率、车辆可插入间隙、匝道排队长度的快速路入口匝道分级控制方法。最后通过VISSIM与COM接口进行匝道控制仿真评价,结果表明控制效果良好和具有实用性。
[Abstract]:Under the rapid development demand of urban economy and urban construction, the shortage of road supply resources and unreasonable traffic management have resulted in the urban expressway congestion and low utilization rate. Because of the shortage of urban land, to solve the traffic congestion of expressway can not only increase the capacity by expanding the scale of road facilities, but also make full use of traffic capacity and related service facilities. Ramp control has been proved to be one of the most effective management methods to solve the problem of expressway efficiency, both in theory and simulation level and in practical examples of major cities. Reasonable ramp control plays a key role in making full use of expressway capacity, improving traffic conditions, preventing traffic jam significantly, coordinating traffic organization between expressway and related roads, and improving the service level of roads. By analyzing the basic characteristics of urban expressway and the combined form of ramp, this paper provides the basis for traffic simulation and traffic flow analysis. Then, the principle of ramp basic control is analyzed, and the evaluation indexes of ramp control are put forward, which provide the basis for the control effect. At the same time, the control strategies and methods of different on-ramp are analyzed and classified to provide theoretical support for the establishment of adaptive ramp control. Accurate short time flow prediction for urban expressway ramp can provide support for early prediction of traffic environment, active control of traffic and improvement of traffic environment. Because of the nonlinear and unstable characteristics of short-time traffic flow, a wavelet neural network based on genetic algorithm optimization is established in this paper, which is compared with the conventional neural network and the forecasting method of replacing the current periodic flow with the previous periodic flow. It has the characteristics of strong learning ability and avoiding falling into local optimum, and its prediction accuracy is verified by an example. Based on the analysis of the different control strategies and methods of the on-ramp on the expressway, there are three main problems in the existing control methods: the signal control of the on-ramp and the lack of the main line can be inserted into the front of the ramp, which results in the queue length spilling back. The bad effect of queue quick cleaning on the main line of expressway and the problem of control lag, combined with short-term traffic flow prediction, the paper puts forward a comprehensive consideration of the expressway utilization efficiency and the interchangeable clearance of the vehicle. The method of ramp grading control for the length of ramp queue. Finally, the simulation evaluation of ramp control is carried out through the interface of VISSIM and COM, and the results show that the control effect is good and practical.
【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：U491

【相似文献】