路段容量和路票约束下的交通网络均衡模型

发布时间：2018-04-25 16:46

本文选题：路票交易约束 + 路段容量约束　；参考：《东南大学》2015年硕士论文

【摘要】：可交易电子路票系统是交通需求管理的一种新方式,为缓解城市交通拥堵问题提供了新思路。本文回顾了路票交易体系、路段容量约束和双层规划模型这三个方面的国内外研究概况,总结了现有的研究成果,在此基础上确定了本文的研究目标和研究内容。首先,本文概述了路票交易约束下的交通网络流问题,路票交易体系是一种路权转让的交通需求管理制度,通过通行权的自由交易实现；实施过程中主要确定路票发行总量、发行时间间隔、初始分配制度、收费方案和交易市场,从城市交通拥堵角度分析了路票交易体系的经济学原理；基于此提出了路票交易约束下的用户均衡模型及等价的数学规划模型,并证明了交通网络均衡状态下路段流量的唯一性和路票交易市场中路票价格的唯一性。最后,在公平性、可实施性方面,将路票交易体系与拥堵收费进行比较。对路票交易体系进行了两个重要的算法设计：路票收费方案的算法设计和路票价格的算法设计。根据用户均衡与系统最优这两个原则之间的差别——边际成本,设计每个路段需要收取的路票数,由路票守恒条件确定路票发行总量。然后,在给定的路票收费方案条件下,用外惩罚函数法将路票约束转化成无约束的交通网络流问题,结合牛顿算法求得路票价格。最后,通过编程,用C#实现了牛顿算法、改进的牛顿算法、外惩罚函数法的求解程序,对算例路网设计其路票收费方案并求解路票价格,验证算法的正确性。介绍拥堵交通网络流问题,用等价的数学规划模型表示路段容量约束,并用增强拉格朗日乘子法对路段容量约束下的拥堵网络进行配流。将路段容量约束添加到路票交易体系中,构造路段容量和路票交易两个约束下的双层规划模型,对部分拥堵路段收取路票时,通过Kuhn-Tucker条件将双层规划模型转化成单层非线性规划问题,利用松弛算法进行交通分配,得到局部最优的路票收费方案及路段流量。最后,将该算法应用于算例路网中,说明一个合理的路票收费方案能有效地缓解交通拥堵。
[Abstract]:Tradable electronic ticket system is a new way of traffic demand management, which provides a new way to alleviate urban traffic congestion. This paper reviews the domestic and international research situation of the three aspects of road ticket trading system, section capacity constraint and bilevel programming model, summarizes the existing research results, and then determines the research objectives and contents of this paper. First of all, this paper summarizes the traffic network flow problem under the constraint of the road ticket transaction. The road ticket transaction system is a traffic demand management system for the transfer of road rights, which is realized through the free transaction of the right of passage; in the process of implementation, the total amount of road ticket issuance is mainly determined. From the perspective of urban traffic congestion, this paper analyzes the economic principles of the ticket trading system from the point of view of the issue time interval, the initial distribution system, the charging scheme and the trading market. Based on this, the user equilibrium model and equivalent mathematical programming model under the constraint of road ticket transaction are proposed, and it is proved that the uniqueness of road flow under the condition of traffic network equilibrium and the uniqueness of ticket price in the road ticket trading market. Finally, in the aspect of fairness and implementability, we compare the traffic ticket trading system with congestion charge. There are two important algorithms for ticket trading system: the algorithm design of ticket pricing scheme and the algorithm design of ticket price. According to the difference between the two principles of user equilibrium and system optimization-marginal cost, the number of road tickets to be collected in each section is designed, and the total number of road tickets issued is determined by the conservation conditions of road tickets. Then, under the condition of given road ticket pricing scheme, the ticket constraint is transformed into an unconstrained traffic network flow problem by using the external penalty function method, and the price of the ticket is obtained by using Newton algorithm. Finally, the program of Newton algorithm, improved Newton algorithm and external penalty function method is implemented by C #, and the pricing scheme of road ticket is designed for example road network, and the price of ticket is solved to verify the correctness of the algorithm. This paper introduces the congestion traffic network flow problem. The equivalent mathematical programming model is used to express the section capacity constraints, and the augmented Lagrange multiplier method is used to allocate the congestion network under the section capacity constraints. The capacity constraint of road section is added to the traffic ticket trading system, and a two-layer programming model is constructed under the two constraints of section capacity and ticket transaction, and when collecting tickets for some congested sections, The bilevel programming model is transformed into a single-layer nonlinear programming problem by Kuhn-Tucker condition, and the relaxation algorithm is used to allocate the traffic, and the local optimal tolling scheme and section flow are obtained. Finally, the algorithm is applied to an example road network, which shows that a reasonable road ticket charging scheme can effectively alleviate traffic congestion.
【学位授予单位】：东南大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U491

【参考文献】