Research On The Urban Rail Transit Passenger Flow Early-warning System

With the rapid development of the urban rail transit system,the rail transportation has been becoming the first traffic trip choice of the most passengers.However,a large number of passengers entering the rail transit system causes enormous effects to the urban rail transit system.In order to ensure a normal operation of this system,mastering the real-time variation of passenger flow,publishing early-warning of passenger flow in time and carrying out the traffic contingency plans has great significant to the normal operation of the rail transit system.The research mainly conduct from the following aspects(1)The design for early-warning system of the rail transit passenger flow.The urban rail transit passenger flow early-warning system is elaborated from four aspects,such as the information needs,system configuration,logical structure and system function realizing.According to the design principle,the four subsystem of the rail transit passenger flow early-warning system including the passenger flow data collection and management,the passenger flow data analysis and forecast,passenger flow early-warning level and passenger flow early-warning output is designed,and the logical structure and hierarchy diagram of the rail transit passenger flow early-warning system is drew.(2)From the relevant concepts of passenger flow,the characteristics of the rail transit passenger flow is analyzed from three perspectives including time,space and time-space.This article describes the concept and classification of passenger flow,the relative content of the investigation of passenger flow and the passenger flow influence.The rail transit passenger flow characteristics is analyzed from the time,space and time-space.The deep reasons in different time dimension and spatial dimension of the passenger flow features is analyzed.(3)The rail transit short-term passenger flow forecasting model based on the support vector machine is constructed,and the grid search,particle swarm optimization and genetic algorithm is used for the parameter optimization.Firstly,the pre-processing of the passenger flow data including identification and repairment of these data is introduced.After accessing the recovery data,the article chose the SVM model that has good generalization ability to forecast the short-term passenger flow of the rail transit,according to the nonlinear and complex characteristics of short-term passenger flow and considered the parameter optimization method and the normalized range of independent variable and dependent variable.According to Chongqing Rail Transit line 1 Shapingba station’s historical data,the prediction and evaluation analysis on two types of passenger flow(5min and 15min)is made.(4)The early-warning evaluation model for urban rail transit passenger flow based on improved matter-element extension model with variable weight is built.Six indicators including train load rate,pedestrian density of station platform,density of pedestrian channel,pedestrian speed,proportion of inbound pedestrian peak flow and proportion of outbound pedestrian peak flow is chose for constructing the early-warning evaluation model for urban rail transit passenger flow.Besides,taking into account of the specificity of rail passenger warning,this article built the evaluation model of the urban rail transit passenger flow early-warning,which is based on the improved matter-element extension model with variable weight,and made the Chongqing Rail Transit line 1 Shapingba station as an example to empirical analysis the passenger flow warning evaluation model.Talking the urban rail transit passenger flow early-warning system as the research objects,the essay forecasted the short-term passenger flow from the parameter optimization method and the normalized range of independent variable and dependent variable,constructed the urban rail transit warning level evaluation model,and stated the relationship between the rail transit security and early-warning level.The research results could be used for evaluating the security of rail transit passenger flow,which made a preparation for responsing to large passenger flow events and guaranteed the normal operation of urban rail transit system.