Font Size: a A A

Research On Passenger Flow Parameters Estimation Of Congestion Points At Rail Transit Stations

Posted on:2019-10-30Degree:MasterType:Thesis
Country:ChinaCandidate:S H DongFull Text:PDF
GTID:2392330596956520Subject:Transportation engineering
Abstract/Summary:PDF Full Text Request
In recent years,with the rapid development of urban rail transit,a mode of network operation has been realized gradually.The passenger flow "blowout" brought by the operation network has applied a dramatic stress on the road network.Urban rail transit station,as a node in the network,frequent traffic congestion phenomenon often occurs,the imbalance between the limited space resources and the growing passenger demand has become increasingly prominent.Take Shanghai as an example,the existing running mileage of the rail transit has reached a number of 617 km(December 2017),which is expected to reach 800 km in 2020 and 1000 km in the future.Coupled with the unfriendly transportation environment above the ground,the Urban has been taking a more significant role in the traffic,and the daily traffic that is over 10 million has become a "new normal" of Shanghai Urban railway.Rail transit stations,as the concentration nodes of all kinds of facilities,equipment and passengers,will become t the most affected part of the whole system.During rush hour or a sudden huge passenger flow,congestion and detention of passengers will often occur in the stations,which has brought great challenge to the management of the station.Under this circumstance,how to realize the accurate monitoring and identification of the passenger flow status,in a station that the space is cramped narrow and enclosed,has become an urgent problem to be solved,which is a great significance to improve the management of the passenger flow and to ensure the safety of the operation of the rail transit.Firstly,this thesis analyzes the causes of large passenger flow,and classified the large passenger flow according to whether the occurrence of it can be predicted.Secondly,take the collection and distribution process of passengers in stations as the research subject,analysis the factors that affect the process and the streamline characteristics,based on the analysis,this thesis summarized the causes of the large passenger flow,classified the congestion points,including the static capacity-based static congestion point and sensitive coefficient-based dynamic congestion point and listed the identification method of the two kinds of congestion points.Meanwhile,based on the area of a station that a congestion point occurs,congestion points are divided into three categories,including station hall congestion points,Pedestrians walkway congestion points and platform congestion points.Based on the analysis,discussed the characters of the facilities and the congestion points.At the end of the chapter,established an indexes system that used to describe the passenger flow statues of the congestion points.Finally,on the basis of all the analysis above,Combined with the existing measures in the urban railway,some measures of station congestion point management under the condition of a large passenger flow are summarized.Finally,this thesis introduced the BP neural network algorithm to the prediction of the station congestion points passenger flow statues and the distribution law under the condition of large passenger flow,as well as improved algorithm with Genetic algorithm(GA).To test the prediction model,this thesis take a station of Shanghai Urban railway as simulation station and collected a data set of 15 days.The result shows that the model can predict the passenger flow statues of all the kinds of congestion points,both the accuracy and the stability are relatively high,which can be used to monitoring the passenger flow statues of the whole statin.
Keywords/Search Tags:Rail transit, large passenger flow, Congestion points, BP neural network
PDF Full Text Request
Related items