Research On Improved Betweenness For Assessing The Anti-attack Capability Of Transportation Network Systems

With the rapid development of the city scale and the rapid increase of population,the structure of the urban traffic network system is becoming more and more complex,when it is hit by any small-scale security risk events and loses its service function,due to the cascade effect of the network system,the influence will spread through the network topology connection,and even cause the paralysis of the transportation function of the whole urban traffic network.Therefore,based on the traditional concept of intermediate number,this thesis proposes an improved concept of intermediate number,which is suitable for traffic network system,to identify the key components of traffic network,so as to improve the anti-attack ability of traffic network.It includes the following research contents.First,by analyzing three kinds of situations in the research of anti-attack ability of network system,pointed out the limitation of the betweenness definition,thereby puts forward an improved betweenness definition of network node and link,as follows: the number of times that all OD pairs passed through a node or link to meet the given constraint condition.Second,this thesis,based on the method of improved betweenness method,further considered passengers travel route choice preference and different station and section the transport flow characteristics.And in this thesis,a weighted improved betweenness was proposed to further distinguish suitable paths satisfying given constraints among all the origin-destination(OD)pairs in rail transit network.When calculating the times of all appropriate paths of all OD pairs passing through a node or link,the shorter appropriate paths were given higher weight of probability traffic,which overcomed the defect of treating these appropriate paths equally by traditional methods.Thirdly,in order to efficiently calculate the improved betweenness and weighted improved betweenness,a natural heuristic algorithm ripple diffusion algorithm is improve,and a many to many ripple-spreading algorithm for solving the first k shortest paths is proposed.The algorithm is very efficient in large-scale networks.Fourthly,a case study is carried out based on the Beijing rail transit network model,and the traditional betweenness value,improved betweenness value and flow weighted improved betweenness value of each station and section of Beijing rail transit network are calculated.Verified by comparative analysis of calculation results,the new network properties improved betweenness and flow weighted improved betweenness in improving transportation network with resistance to strike capability.