南京市区公交、地铁加权复合网络建模与实证分析

发布时间：2018-04-14 19:29

本文选题：公共交通网络 + 复杂网络　；参考：《南京邮电大学》2017年硕士论文

【摘要】：城市公共交通网络作为城市交通运输系统的核心组成部分,在居民的日常生活中扮演着重要的角色。伴随着城市的发展和人口的不断增加,城市居民不断增长的出行需求与拥堵的公共交通之间的矛盾日益突出。为了提升公共交通的服务质量,针对基于公交和地铁为主的城市公共交通网络的研究受到了广泛的关注。本文以复杂网络理论为基础,在深入分析了城市公共交通网络的拓扑结构和功能特性之后,构建了一种公交地铁加权复合网络模型,并对此模型进行了进一步的分析研究。本文的主要工作和成果如下:1.在公交地铁双层网络的基础上,通过添加地铁站点与其周围一定范围d内的公交站点间的连边,提出了一种公交地铁复合网络模型。考虑公交地铁载客量、行驶速度、路段运营线路数、路段距离和拥堵等因素的影响,定义了连边容量、时间和容量效率三种权值,对复合网络进行加权处理。研究表明,模型参数d的增大,使复合网络中节点和线路间的连接更加紧密。同时,共享单车行业的兴起使公交地铁连边的出行更加便捷高效,提高了各种出行方式的衔接配合效率,这将引导人们更多地在公交和地铁间换乘,满足人们的出行需求。2.由于常规和出行高峰时段网络存在很大的差异,本文定义了时变条件下网络连边容量、时间和容量效率权值。高峰时段为了增大运力,各线路的发车时间间隔会缩短,同时,随着道路上车流的增多,道路发生拥堵的风险增大。研究表明,出行高峰时段,连边容量会有显著的增大,但容量效率变化较小。增加公交站点和地铁站点周围共享单车的投放量,可以有效的提升整个网络的出行效率。3.本文通过局部评价指标和全局评指标对网络进行节点重要性分析。同时,根据本文定义的连边容量、时间和容量效率权值对网络进行连边重要性分析。研究表明:节点攻击策略下,基于容量效率的点权攻击方式对网络最大连通子图率和全局效率的影响较大;连边攻击策略下,基于时间成本和容量效率权值的攻击方式分别对网络全局效率和网络最大连通子图率的影响比较大。
[Abstract]:As the core part of urban transportation system, urban public transport network plays an important role in the daily life of residents.With the development of the city and the increase of population, the contradiction between the increasing travel demand of urban residents and the congested public transport is becoming increasingly prominent.In order to improve the service quality of public transport, the research on urban public transport network based on public transport and subway has been paid more and more attention.Based on the theory of complex network, after analyzing the topological structure and functional characteristics of urban public transportation network, this paper constructs a weighted composite network model of public transit subway, and makes further analysis and research on this model.The main work and results of this paper are as follows: 1.On the basis of the double-layer network of public transportation and subway, a model of the compound network of public transportation subway is proposed by adding the connecting edge between the subway station and the bus station within a certain range of the stations around it.Considering the influence of the factors such as the carrying capacity, the driving speed, the number of operating lines, the distance and the congestion of the subway, the paper defines three kinds of weight values, such as the connected capacity, the time and the capacity efficiency, and carries on the weighted processing to the compound network.It is shown that the increase of model parameter d makes the connection between nodes and lines closer in the composite network.At the same time, the rise of bicycle sharing industry makes the transit subway trip more convenient and efficient, and improves the efficiency of various trip modes, which will lead people to transfer more between public transport and subway, to meet the travel needs of people.Due to the great differences between conventional and travel peak time networks, this paper defines the capacity, time and capacity efficiency weights of the network under time-varying conditions.In order to increase the capacity in rush hour, the departure time interval of each line will be shortened, at the same time, with the increase of traffic flow on the road, the risk of road congestion will increase.The results show that during the travel peak period, the connected capacity increases significantly, but the capacity efficiency changes little.Increasing the amount of bicycles shared around bus stations and subway stations can effectively improve the travel efficiency of the whole network.This paper analyzes the node importance of the network by local evaluation index and global evaluation index.At the same time, according to the capacity, time and capacity efficiency weights defined in this paper, the importance of continuous edge is analyzed.The results show that the point-weight attack based on capacity efficiency has a great influence on the maximum connected subgraph rate and global efficiency under the node attack strategy, while in the continuous edge attack strategy, the maximum connected subgraph rate and global efficiency of the network are greatly affected.The attack methods based on the weight of time cost and capacity efficiency have great influence on the global efficiency of the network and the maximum connected subgraph rate of the network respectively.
【学位授予单位】：南京邮电大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：O157.5;U12

【参考文献】