Analysis Of Rail Transit Topology Based On Passenger Flow Weighting And Its Vulnerability

The development of urbanization has brought about the continuous expansion of urban scale,quantity,and the exponential growth of urban population,which put huge pressures on the limited traffic resources of the city.The urban rail transit has the characteristics of fast,convenience,efficiency,large capacity,environmental protection and pollution-free,which have a good effect on alleviating these pressures.Urban rail system is a complex dynamic system composed of static topological structure and passenger flow loaded on it.In recent years,the researchers used a complex network theory to study the characteristics of rail transit system,but most studies focused on the network topology structure itself,and ignored the function of the network transport passenger flow properties,and the distribution of passenger flow on the rail transit network makes the network have spatio-temporal dynamic complexity.Therefore,to study the characteristics of rail transit network,we should not only study its static topology,but also explore its characteristics under the influence of dynamic passenger flow.It is of great significance for the safety operation of rail transit and the planning and optimization of rail transit network to reunderstand the network and analyze its complex characteristics from the perspective of dynamic changes.In this paper,the rail transit system loaded with dynamic passenger flow is taken as the research object,and the rail transit network weighted by passenger flow is studied by applying the theory of complex network.Firstly,this paper introduces the basic concept of complex network,explains the representation method of complex network,focuses on the modeling mechanism of complex network of rail transit.On this basis,the rail transit network model based on passenger flow weighting is proposed,also,the measurement index of characteristic parameters is extended,and the Matlab simulation method of characteristic parameters is given.Secondly,we take Chongqing light rail transit network as the research object,and build a light rail station network model and the light rail transit network model based on weighted passenger flow,Based on the characteristics of data structure,the computer acquisition method of orbit network adjacency matrix is given.Also,based on AFC card transaction data and "Transcad" modeling method,an algorithm of "OD" passenger flow acquisition is proposed to obtain flow edge weight matrix.Then,thecharacteristic parameters of the network under the two models are calculated,and found that the unweighted light rail station network has scale-free characteristics,and the weighted network with passenger flow factors taken into account presents small-world characteristics.By means of comparative analysis,the nature of the influence of passenger flow factors on the network is revealed.Then,using the efficiency of the network as the evaluation index,two different attack methods are used to attack the unweighted and weighted network to judge the vulnerability of the network,and a node importance ranking method based on the contribution matrix is proposed.The importance of the node,attacking the network,and verifying the reliability of this method.The study found that: 1)In the unweighted network,the network efficiency is reduced by 80% when the node deletion rate reaches 36% under random attack,and the track network efficiency is reduced by 80% when the node deletion rate is 15% under deliberate attacks.In the weighted network,the network efficiency drops by 80% when the node deletion rate reaches 30%,and the network efficiency drops by 80% when the node deletion rate reaches 10% under deliberate attacks.This indicates that the network is highly vulnerable to random attacks and vulnerable to deliberate attacks.Poor,but the vulnerability of deliberate attacks after considering the passenger flow factor is worse;2)The attack based on the importance ranking of the contribution matrix presents similar attack results and the network efficiency decreases the fastest,and the calculation takes the shortest time.Finally,based on the above analysis,the network and the weighted network are proposed to reduce the vulnerability of the network.