Research On Robust Optimization Of Dynamic And Coordinated Passenger Inflow Control For Urban Rail Transit Stations

The over-crowded situation of local rail transit network is becoming increasingly serious due to the ever-burgeoning travel demand caused by the continuous expansion of urban rail transit.In order to reduce the transportation pressure and ensure the operation safety,passenger inflow control measures have been taken by more and more urban rail transit stations.However,the passenger inflow control strategies of most stations are mainly formulated by its managers independently according to their experience and historical passenger volume data,which lack the support of scientific and effective theories and methods in selecting control periods and determining control strength.Besides,the cooperation among multiple stations is not considered.Moreover,affected by various factors,the passenger demand is often uncertain.Therefore,it has important research significance and practical application value to study the multi-station dynamic and coordinated passenger inflow control for urban rail transit lines under uncertain passenger demand.This paper mainly focuses on how to alleviate the congestion of each station on an urban rail transit line by taking dynamic and coordinated passenger inflow control measures among multiple stations,and to reduce the sensitivity of passenger inflow control strategy to the disturbance of uncertain passenger demand.The main research work of this paper is as follows:(1)The concept of equivalent control time interval is introduced for the problem of multi-station dynamic and coordinated passenger inflow control to establish the corresponding relationship between trains and passengers,so as to calculate the remaining capacity of the carriage after the trains arrive at the station,which lays the foundation for the construction of dynamic and coordinated passenger inflow control models of urban rail transit stations.(2)The dynamic and coordinated passenger inflow control for urban rail transit stations based on certain passenger demand is studied.In order to generate a passenger inflow control strategy satisfying time-varying passenger demand and system optimum,a mixed integer programming model with the aim of minimizing the accumulated number of passengers stranded outside the stations on the whole line is constructed,which is mainly constrained by the passenger demand,train capacity,the maximum number of people allowed to gather at the platform and transfer capability.The case study of Beijing Metro Batong line shows that the strategy generated by the model can reduce the accumulated number of stranded passengers by 90.3% compared with the coordinated but non-dynamic passenger inflow control strategy,and 20.6% compared with the dynamic but non-coordinated strategy.(3)When the passenger demand is uncertain and cannot be predicted in advance easily,in order to ensure the effectiveness of the passenger inflow control strategy in the case of massive passenger demand,a robust optimization model is constructed by using continuous numbers in a section to describe the uncertain demand,and then the strong dual theory is used to transform it into a robust counterpart model which is easy to solve.The case study of Beijing Metro Batong line shows that when the uncertain passenger demand fluctuates greatly,the robust strategy generated by the model can still maintain a good performance by selecting appropriate robust control parameters.Compared with the deterministic model,the accumulated number of passengers stranded outside the stations on the whole line can be reduced by 9.73% in the worst case of passenger flow demand,which avoids the risk caused by uncertain passenger demand.(4)When the passenger demand is uncertain but can be predicted in advance,the problem of how to generate a passenger inflow control strategy that is less sensitive to the disturbance of uncertain passenger demand and better for all possible passenger demand scenarios is studied.A robust optimization model for dynamic and coordinated passenger inflow control is constructed by using the scenarios with known probability to describe the uncertain passenger demand,and the optimal number of passengers allowed to enter each station during each equivalent control period is solved.The case study of Beijing Metro Batong line shows that the relative error of each passenger demand scenario between the accumulated number of passengers stranded outside the stations on the whole line under the robust strategy and its optimal value calculated by the deterministic model is less than 2%,which proves that the robust passenger inflow control strategy has a better performance for the scenario set and can effectively reduce the sensitivity to the disturbance of uncertain passenger demand.