Research On Vulnerability Assessment And Application Of Urban Rail Transit Network Based On Passenger Flow Impact

With the development of urban rail transit,the passenger flow of urban rail transit is increasing day by day,and the safety of rail transit is becoming more and more important.However,frequent rail transit safety accidents in recent years have also exposed its vulnerability in the operation process.Therefore,in order to improve the operational safety level of the rail transit network,it is particularly necessary to grasp the changing law of network vulnerability after emergencies and give corresponding application measures to improve network security.This paper takes rail transit network vulnerability as the research object,constructs a data-driven rail transit network vulnerability assessment framework based on rail transit operation passenger flow,and applies the vulnerability assessment results in the emergency rescue station siting model,which makes the research of this paper more practical.Firstly,based on cascade failure theory and coupled image lattice model,a rail network vulnerability simulation model was established with rail traffic data,and the model was validated from two aspects:different time steps and different stations,and it was found that the established model reflected the characteristics of passenger interchange and the risk resistance of stations,and the simulation model was close to the actual scenario.Secondly,the TOPSIS model was used to quantitatively identify the key stations,and the network vulnerability assessment indexes were established from the perspectives of network structure and service level,and the network vulnerability values were quantified by weighting the indexes using hierarchical analysis.the sensitivity of large transit stations to coupling coefficients is higher than that of other stations.Finally,based on the vulnerability of the urban rail network,the traditional emergency rescue station location model(set coverage model)is improved by adding a new objective function weighted by the shortest rescue time,in which the vulnerability weight is incorporated into the objective function.A simulated annealing algorithm is used to compare the solution of the improved siting model before and after the improvement.The results showed that the improved location model is more close to the actual situation.