The Research Of Urban Rail Transit Coordinated Passenger Flow Control System Using Discrete-Event-Simulation-Based Optimization Method

Under the networked operation of urban rail transit(URT),the rapid growth of passenger flow leads to the contradiction between the original designed capacity of the station and the passenger flow demand.It is the realistic demand of the rail transit operation organization to ensure the operational safety of the rail transit network under the condition of satisfying the passenger flow demand as much as possible.Under the condition of satisfying the passenger flow demand as much as possible,ensuring the operational safety of the rail transit network is the actual demand of the URT operation organization.When the passenger flow pressure is large,the station-level passenger flow control cannot effectively alleviate the passenger flow pressure and ensure the safe operation of the rail transit.Therefore,the coordinated passenger flow control needs to be implemented at network level.The existing passenger flow control method at the line level can provide some guidance for the formulation of the control scheme in this paper.However,in the existing research,the randomness of passenger flow is considered insufficiently and the analytical model is too complex to be solved.With the scale of the network increasing,the complexity of the algorithm increases exponentially and the application range is limited.Discrete event can eliminate the shortcomings that solving explicit analytic expressions in the case of mathematical optimization needs to take a long time,which provides a new idea for model solving.Thus,this paper uses the queuing theory to describe the randomness of passenger flow,and applies the discrete-event simulation(DES)method to describe the arrival process of passenger flow and the difficulty of solving can be reduced.This paper intends to establish a hierarchical coordinated control system,and a reasonable passenger flow control scheme can be made based on optimized control model.The main work as follows:(1)The characteristics of passenger flow on the URT station and train are studied,which lay the foundation for building an optimization model.(2)Based on the operational characteristics of passenger and train,considering the evacuation capacity of the station platform,a URT queuing network model is constructed to describe the randomness of the passenger flow.(3)Combined with the queuing network model,taking Chengdu Metro Lines 1 and 2 as an example,SimEvents is used to build a DES model of URT network.The comparison between the actual and simulated passenger flow data of line section verifies the effectiveness of the DES simulation model.(4)A hierarchical coordinated control framework is established and for the coordinated passenger flow control,an optimization model with an objective of the minimum negative impact on the network which is based on DES is established.The algorithm based on Simultaneous Perturbation Stochastic Approximation(SPSA)is designed.The results show that the optimal passenger flow control scheme determined by the proposed optimization model is better than the current situation.The average density of passenger flow on the crowded platform drops to 1.54 ped/m~2(Level C)and the dwell time of passenger on the crowded platform is reduced by 35%.