【作者】 韦翀；

【导师】 王健；

【作者基本信息】 哈尔滨工业大学 ， 交通运输规划与管理， 2010， 硕士

【摘要】 随着城市交通需求和机动车保有量的不断增加,城市交通的供需矛盾日益突出,交通拥挤现象也越来越严重。城市交通拥挤收费作为交通需求管理的措施之一,可以有效的减少交通拥挤。本文在分析道路饱和度的基础上,建立了基于道路饱和度空间分布的交通拥挤收费模型,为最大程度的减少交通拥挤提供有效的交通拥挤收费方案。首先,分析了影响拥挤收费费率的因素,利用经济学理论阐述了交通拥挤收费的基本原理,对交通拥挤收费的分类和实施后影响进行了总结分析,并在交通配流理论的基础上,介绍了拥挤收费的双层规划模型。其次,用路网中道路饱和度的标准差来表征道路饱和度在路网中空间分布的情况。建立了基于道路饱和度空间分布的拥挤收费双层规划模型。上层模型为最小化路网用户的总出行时间,用道路饱和度的标准差的取值范围作为上层模型的约束条件;下层模型为固定需求下的用户平衡问题。在测试分析了函数迭代算法、连续平均算法和Polyak迭代平均算法这三种算法后,选取Polyak算法求解下层模型,运用二元遗传算法求解上层模型问题,因此用Polyak算法和遗传算法的组合算法来求解本文所建立的交通拥挤收费模型。最后,通过案例分析来验证所建立的交通拥挤收费模型的实用性和模型求解算法的有效性,其中采用的路网是美国南达科他州的苏福尔市的道路网,算法求解是Matlab环境下编程实现的。算例分析了能使路网用户总出行时间最小的最优收费路段和收费水平,结果发现通过增加收费路段的数目,用户总出行时间和道路饱和标准差的就会变得更小。而且道路饱和度标准差达到最小值的收费费率比用户出行时间达到最小是的要高出不少。通过计算得出了在5个道路标准差区间中的最优收费方案供交通管理者选择。更多还原

【Abstract】 With the growth of urban traffic demands and motor vehicles, the imbalance between supply and demand of urban traffic is obvious day by day. As a measure of traffic demand management, congestion pricing can decrease the traffic congestion effectively. Based on the spatial distribution analysis of road saturation, the congestion pricing model is established, and this can provide effective scheme for decreasing the congestion in greatest degree.Firstly, factors influencing the traffic congestion toll is analyzed, and the basic principle of congestion pricing is explained from the aspect of economic theory. Based on the analysis of the classification of congestion pricing and the influence after the implement, the bi-level programming model of the congestion pricing is analyzed in combining with the basic theory of traffic assignment.Secondly, the standard deviation of road saturation is used to describe the spatial distribution, and the traffic mode of private car is in consideration. The bi-level programming model based on the spatial distribution of road saturation is established. The upper level is to minimize the total travel time of users subjecting to the standard deviation of road saturation, while the lower level is the user equilibrium under fixed demand. The Genetic Algorithm combined with Polyak algorithm is selected to solve the model after the contrast of function iteration algorithm, method of successive averages algorithm, and Polyak iterate averaging algorithm.Lastly, the road network of the Sioux Falls City,South Dakota is taken as a case to validate the practicability of the model and the validity of the algorithm, and the algorithm is realized by Matlab software. Case analysis shows the optimal toll locations and toll levels, and the result also shows that the total travel time can be decreased with the increase of number of toll links. The best toll scheme is determined for the traffic managers from the 5 road deviations.更多还原