Simulation Of Urban Rail Transit Spatio-Temporal Route Deduction Based On Afc Data

The development of urban rail transit system tends to be cyberized,the network is becoming more and more complex,and the number of reachable paths between stations is proliferating.Accurately grasping the dynamic distribution of passenger flow in the network meets the need of fine operation development of urban rail transit.To address this issue,the main contents of this paper are as follows.1.Introduced data mining techniques,clarify the sources and basic formats of passenger data and train data in this paper,and perform data cleaning.At the macro level,by using the data classification method,extract and analyze the basic passenger flow characteristics in three different views: network,station and passenger.At the micro level,based on the travel logic and travel process decomposition of passengers in the urban rail transit system,the distribution of uncertain components of travel time are estimated,and established the estimation model of station detention rate to describe the boarding probability of passengers waiting for trains.2.Analyzed the basic theory and method of the effective path search problem in urban rail transit system,and supported by the theory of aggregation,this paper designed an efficient path search algorithm based on the characteristics of the interchange network.Used hybrid Gaussian model to describe the OD travel time distribution due to that the travel time of paths obeys normal distribution.An improved EM algorithm based on the method of density peak clustering is used to solving the model.By disasseming the passenger travel decisions,this paper established the probabilistic matching model of spatio-temporal routes based on Bayes Theorem,and got the results of the analytical method.3.This paper analyzed the applicability of Anylogic system simulation software in the field of urban rail transit and the lackness in the application in network level.The construction method of passenger spatio-temporal route simulation and deduction model based on multibody modeling is detaily proposed.Designed a visualized passenger-flow simulation system of urban rail transit,which based on the reality of Chengdu Rain Transit.Displaye and anylyzed the output data,including spatio-temporal route and interval-section passenger flow etc.