Dynamic Estimation Model And System Of Passenger Flow In Urban Rail Transit Network

Under the background of network operation,urban rail transit shows a number of characteristics,such as huge road network size,sharp increase of passenger flow,greater impact of emergencies,more complex passenger flow characteristics and evolution rules,etc,which made the transportation’s organizational decision-making be faced with great challenges.Therefore,technological means and methods are urgently required to timely and accurately estimate or predict the status of passenger flow in road network as well as its dynamic changes for the basis of quantitative decision.In face of this goal,the main content of this paper is as follows:(1)It analyzed the requirements for the status analysis of passenger flow during the real-time operation process,and made clear that the study should be carried out from the following three perspectives:conduct the estimation and prediction of road network passenger flow OD in passenger flow OD layer,study the deduction and simulation system of passenger flow distribution and the development system in the road network passenger distribution layer,carry out the quantitative analysis of the impact of emergencies in the network operation effect layer.(2)This paper analyzed the known state of passenger flow OD information under different operating periods,proposed a multi-level and progressive dynamic estimation framework of passenger flow OD,and constructed the passenger flow OD prediction model based on clustering and the time-sharing passenger flow OD timing sequence prediction model based on LSTM.The both were organically integrated,which not only excavated the rules contained in the historical data,but also made full use of the actual data in the operation of the day to improve the prediction accuracy and better accord to the actual situation.(3)Based on the existing research,the passenger flow distribution deduction and simulation model and system which mix real-time and time-out,have variable simulation region and adapt to normal and emergencies were constructed,and the simulation system was designed and developed.Meanwhile,in order to analyze the impact of emergencies,the design and calculation of quantitative indexes were studied.(4)Finally,taking Beijing subway for example,the constructed estimation and simulation deduction of passenger flow OD was for simulation experiment.Later,the passenger flow status was analyzed from different aspects,and then the impact of emergencies was analyzed.