Analysis And Research On Passenger Flow Of New Urban Rail Transit Lines Opened

In recent years,in order to further solve urban traffic problems,many cities have vigorously developed urban rail transit,and the proportion of urban rail transit in urban public transportation is also increasing.Due to the high cost and long period of urban rail transit construction,it is difficult to build a relatively complete rail transit road network at one time.The development of most urban rail transit in my country has experienced the process of gradually expanding the road network.Except for a few first-tier cities with relatively complete road networks,there are still many unopened lines in the rail transit road network planning of other cities,and each new line opening will change the road network structure and affect the passenger flow of the existing road network.Therefore,it is necessary to analyze and study the passenger flow when the new line is opened.The main research contents of the paper are:(1)Analyze the distribution of passenger flow characteristics of rail transit.Based on the real passenger flow data in Zhengzhou,the characteristics of passenger flow are analyzed from two aspects: time and space.In terms of time characteristics,the change of passenger flow is presented on the one hand in a weekly cycle,and on the other hand,the passenger flow in the morning and evening peaks is larger at different time periods on the same date,and the passenger flow at other times is smaller;in the spatial characteristics,the more developed areas around the station The passenger flow is large,and vice versa,the passenger flow is small.(2)Based on the passenger flow characteristics,the stations are clustered,and the passenger flow of the new line stations is predicted based on the distribution of the working and living population.Based on the real passenger flow data of Zhengzhou rail transit,the hierarchical clustering algorithm is used to cluster the stations,and the rail transit stations are divided into five categories,and the passenger flow characteristics of each category are summarized.This paper introduces the forecasting method of passenger flow at the new line station.First,through the statistics of the distribution of the working and living people around the station,the correlation between the distribution of the working and living population and the characteristics of the passenger flow is verified,and the results of the station classification according to the two are basically consistent;According to the distribution of working and living population,the new line stations are matched with the existing line station categories;then,according to the characteristics of the station population,station category and other characteristics,a BP neural network model is established to predict the passenger flow of the new line stations and compare with the real passenger flow.,the error is less than 20%.(3)Predict natural passenger flow based on historical data,and analyze the pros and cons of different driving intervals during morning rush hour.Based on the historical passenger flow data,a passenger flow prediction model based on XGBoost was established to predict the passenger flow of some stations at a granularity of 5 minutes,and compared with the actual value,the change trend of the passenger flow of the forecast results is consistent with the real value,and the accuracy reaches 86%.In order to maximize the operational efficiency of urban rail transit,taking the minimum number of people waiting and the highest operating economic benefit as the objective function,the advantages and disadvantages of different driving intervals during the morning peak period are analyzed.(4)Analysis of the impact of the opening of the new line on the passenger flow of the existing line stations.Compare and analyze the passenger flow data of the existing line network,line and station before and after the opening of the new line;according to the different passenger flow changes,the passenger flow changes of the existing line stations are divided into diverted passenger flow and induced passenger flow for analysis;for diverted passenger flow,the station The location and distance from the new line are analyzed,and it is found that the stations with more diverted passenger flow are basically located in the section parallel to the new line,or near the interchange point between the new line and the existing line station;for the induced passenger flow,real passenger flow data are used.Based on the variables of ride time,transfer times,destination attraction,property factor,and scale factor,the utility function is calculated by using the multinomial Logit model,and the model parameters are estimated by the maximum likelihood function,and the utility-based reachability is established.Sexuality Metrics Model.By calculating the station accessibility before and after the opening of the new line,it is found that the stations with a larger increase in accessibility are basically consistent with the stations with a larger increase in passenger flow after the opening of the new line.