Prediction Of Short-term Passenger Flow Distribution After The Opening Of New Urban Rail Transit Line

During the "13th Five-Year Plan" period,the construction of urban rail transit in our country enters a peak period,and urban rail transit in various cities ushers in a large-scale network operation phase.In order to give full play to the metro advantage of urban rail transit and improve the service level,operators and participants have higher expectations for short-term passenger flow forecast after the opening of the new line.Accurate passenger flow forecasting can provide an important basis for the station size,transport organization and operation management after the new route is opened up.At the same time,passenger flow forecast is also an important part of the preliminary work of urban rail transit and an important guarantee for the healthy development of urban rail transit.However,there are many shortcomings and difficulties in the traditional passenger flow forecasting method for urban rail transit.For example,after the new line is switched on,the change of the network topology leads to the failure of the traditional growth factor method and the Follet method.Passenger flow distribution needs to be determined through manual surveys,which consumes a great deal of energy and financial resources.Most cities in our country are at a stage of rapid development.The traffic regulations of urban master plan and urban complex are adjusted frequently.The differences in the topological structure of each stage lead to the change of the basic conditions of forecasting,which leads to the generally low prediction accuracy.Passenger flow distribution is an important part of passenger flow forecasting.In order to solve the above practical problems,this paper focuses on the distribution of passenger flow based on historical data.Accurate passenger flow forecasting will provide the basis for optimizing which arrange the station facilities,equipments,refining transportation organization and route selection.The main contents of this paper are:(1)Based on the introduction of the traditional urban rail transit passenger flow forecasting method,this paper summarizes the research of road network topology,path solving and data mining,and points out the research goal and content of the paper.Based on the analysis of the passenger flow distribution model of the traditional urban rail transit,a new method of passenger flow distribution model for urban rail transit based on the ticket information is proposed.(2)Using the programming language,database and data integration tools to analyze the status quo data of Chengdu Metro and get the key data.(3)By analyzing the data classification model adopted by Chengdu Rail Transit Group,the indicators such as congestion degree and transfer penalty time are unified.According to the current passenger flow data,the full path selection algorithm and model written by C # are used to calculate the current passenger flow related indicators of Chengdu Metro.The passenger traffic indexes of the two are compared,and then the model parameters are calibrated.(4)The standard gravity model,the single constrained gravity model and the double constrained gravity model are respectively introduced.The calibration is further discussed.Based on the existing achievements of predecessors,they are respectively calibrated by least squares method and trial calculation method.(5)Using the method of cluster analysis,linear programming and gravity model,taking Chengdu Metro Line 7 and 10 as an example,it predicts the passenger flow of full-time after the opening of new line.And calculate the passenger flow density,transfer volume,average distance and other related passenger flow indicators.(6)The predictions of the passenger flow distribution of the urban rail transit based on historical data are contrasted and analyzed.