【作者】 张璇；

【导师】 吴建平；

【作者基本信息】 北京邮电大学 ， 计算机科学与技术， 2013， 硕士

【摘要】 近年来,人均汽车保有量不断上涨,虽然部分大中型城市采取了尾号限行的措施,但是交通拥堵的状况依然愈演愈烈。尤其是早晚高峰期间及节假日的庞大交通流量使得负荷早已接近饱和的道路网络系统濒临崩溃。交通拥堵不仅造成了一定的经济损失,并且在堵塞过程中产生了大量的环境污染。出于减轻交通拥堵的状况,减少不必要的燃油空耗、尾气污染的目的,如何提高道路网络的可靠性,降低因节点拥堵造成的局部网络瘫痪可能性,已成为专家学者研究的热门课题。在学者们对于道路交通网络的长期观察研究中发现,道路网络作为一个整体,在实际情况下,拥堵的产生并非随机分布毫无规律可循的,而是主要由于某些交叉路口或路段的拥堵造成了一定范围的路网瘫痪,因此对于路网中的这些交叉路口和路段进行分析至关重要。从二十世纪70年代开始,对于网络中节点重要性的研究就已经开始。通过研究各种网络的抽象拓扑结构,借助于网络中节点的度、介数,节点生成树,最短路径等属性特征,量化地研究网络中各个节点的重要性差别。然而这些研究都侧重于网络拓扑图的结构属性,而道路网络存在自身的一些特征,例如：地理位置信息,交通流量,车流速度等。这些特征使得上述研究并不能与道路网络中关键节点的发掘完全契合。为解决上述问题,本文在对通信网络和道路网络中关键节点研究方法总结、对比和分析的基础上,提出了研究道路网络中的关键节点首先需要明确网络的功能并以此确定网络中关键节点的这一解决问题的方法,明确关键节点定义从而确定道路网络中节点重要性的评价标准,即道路网络中易受损和易引起其他节点和大面积路网瘫痪的节点为关键节点。由此提出了路网中关键节点分析的特色算法,并选取经典路网模型Sioux Falls进行分析研究,验证了所提出算法的合理性和实用性。更多还原

【Abstract】 In recent years, with the increasing of car ownership per capital, the traffic congestion is more and more intensive, even though there is tail number limited in some large and medium-sized city. Especially morning and evening rush hours and holidays, the huge traffic load makes the road network near to collapse, which is already saturated. The traffic congestion makes not only economic losses but also widespread contamination. In order to reduce traffic congestion, fuel consuming and gas pollution, it has become a hot topic to researchers that how to improve the reliability of road network and to reduce the possibility of local road network paralysis caused by node jam.In the long term study and observation of road network, researchers found that the congestion in the network does not generate randomly and irregularly, in the actual situation, but caused by the certain nodes or links. Therefore, the study of the key nodes and links is significant.Since1970s, research on the key nodes in the network has been put forward. Through studying the abstract topology structure of various network and based on the degree, betweeness, shortest path, spanning tree of network, the importance of nodes is discriminatingly quantified. However, these researches focused more on topological characteristics of road networks. The road network has some own characteristics such as geographical position information, traffic flow, traffic velocity and so on. These features make the above researches not match the study of key nodes in the road network that much.To solve above questions, based on the summary, comparison and analysis of communication network and road network, the solution is put forward that the function of road network and the definition of key nodes should be clear-cut firstly and determine the evaluation criterion of nodes’ importance. The definition of key nodes is that the key nodes should be the vulnerable ones and which make other nodes failed and local road network paralysis more easily than others. The special algorithm of key node analysis in the road network is proposed, and the classic road network, Sioux Falls, as numerical experiment is provided to demonstrate the practice and accuracy of the solution.更多还原