基于MPC的快速路入口匝道协调控制策略研究

发布时间：2018-07-06 09:49

本文选题：快速路系统 + 匝道控制　；参考：《浙江大学》2014年硕士论文

【摘要】：随着城市化进程的加速,城市交通拥堵问题已成为全球大城市面临的共同问题。发生在城市快速路上的交通拥堵如果不及时消除将可能导致路网大面积瘫痪,因此,对快速路拥堵的研究和治理变得尤为重要。本文以城市快速路为研究对象,从宏观交通流模型入手,以入口匝道控制为手段,将模型预测控制(MPC)理论应用到快速路匝道控制系统中,探究解决城市快速路交通拥挤的匝道优化控制策略。全文的主要研究成果总结如下。 (1)基于交通流模型,以匝道调节率为控制变量,提出了匝道MPC调节的离散时间非线性动态控制命题。MPC策略以METANET模型作为过程模型,采用离散系统的极小值原理进行求解。仿真结果表明,匝道MPC调节能够显著缓解交通拥堵,改善路网总体运行效率；同时与最优控制相对比,匝道MPC策略在模型和过程失配情况下依然有很强的控制效果。 (2)针对匝道控制节点过多带来的负面效应,在MPC框架基础上扩展了匝道控制节点选择模块。以无控制时各控制节点的梯度信息作为特征值,将控制节点按照对路网性能潜在影响的大小进行K-均值聚类,提出一种快速有效的动态选择控制节点的方法。仿真结果表明,该方法在显著改善路网通行效率的同时能够减少算法的计算时间,提高算法的求解效率。 (3)针对城市机动车尾气污染严重的问题,提出一种兼顾通行效率和环境效益的快速路入口匝道多目标协调控制策略。以METANET模型为基础,将宏观交通变量转化为微观变量,然后采用微观排放模型计算尾气排放量。在MPC框架下,综合交通效率和环境效益指标,给出非线性动态优化控制命题,并基于极小值原理进行求解。在多匝道快速路网的仿真表明,该策略在兼顾交通通行效率的同时能有效减少机动车尾气排放,显著改善燃油消耗。 (4)基于上述工作,设计了入口匝道控制通用软件,以便于比较不同路网结构、不同交通需求等场景的仿真效果。该软件系统由参数设置和仿真分析组成。其中参数设置部分为软件的输入,包括路网参数、模型参数和仿真参数的设置;仿真分析部分为软件的输出,包括交通需求量、匝道调节率、路网优化前后的性能指标与交通流状态。
[Abstract]:With the acceleration of urbanization, urban traffic congestion has become a common problem faced by big cities in the world. If the traffic congestion on the urban expressway is not eliminated in time, it may lead to the paralysis of the road network. Therefore, it is very important to study and control the traffic congestion on the expressway. In this paper, taking the urban expressway as the research object, starting with the macroscopic traffic flow model and taking the on-ramp control as the means, the model predictive control (MPC) theory is applied to the ramp control system of the expressway. To explore the optimal ramp control strategy to solve urban expressway traffic congestion. The main research results are summarized as follows: (1) based on the traffic flow model and taking the ramp regulation rate as the control variable, the discrete time nonlinear dynamic control proposition of ramp MPC regulation is proposed. The METANET model is used as the process model. The minimum principle of discrete system is used to solve the problem. The simulation results show that the ramp MPC regulation can significantly alleviate traffic congestion, improve the overall operation efficiency of the road network, at the same time, compared with the optimal control, Ramp MPC strategy still has a strong control effect in the case of model and process mismatch. (2) in view of the negative effect caused by the excessive number of ramp control nodes, the ramp control node selection module is extended on the basis of MPC framework. Taking the gradient information of each control node as the eigenvalue, the control nodes are clustered according to the potential impact on the road network performance, and a fast and effective method for dynamic selection of control nodes is proposed. The simulation results show that the proposed method can significantly improve the efficiency of the road network and reduce the computational time of the algorithm and improve the efficiency of the algorithm. (3) aiming at the serious pollution problem of urban motor vehicle exhaust gas, the proposed method can reduce the computational time and improve the efficiency of the algorithm. This paper presents a multi-objective coordinated control strategy for freeway on-ramp with both traffic efficiency and environmental benefit. Based on the METANET model, the macro traffic variable is transformed into the micro variable, and the exhaust emission is calculated by the micro emission model. Under the framework of MPC, the nonlinear dynamic optimal control proposition is given by synthesizing the traffic efficiency and environmental benefit index, and the solution is based on the minimum principle. The simulation of multi-ramp expressway network shows that the strategy can effectively reduce vehicle exhaust emissions and improve fuel consumption while taking into account traffic efficiency. (4) based on the above work, a general software for on-ramp control is designed. In order to compare the simulation effect of different road network structure and different traffic demand. The software system consists of parameter setting and simulation analysis. The parameter setting part is the input of the software, including road network parameter, model parameter and simulation parameter setting, the simulation analysis part is the software output, including traffic demand, ramp adjustment rate, Performance index and traffic flow state before and after road network optimization.
【学位授予单位】：浙江大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U495;U491.54

【参考文献】