Research On Passenger Flow Prediction And Analysis For Lanzhou-yuzhong Suburban Railway Based On Spatio-Temporal Data

With the promotion of Lanzhou’s "One Centre,Two Wings" urban development strategy,the corresponding urban rail transit construction is also underway.At present,Lanzhou City Rail Transit Line 1,which serves the "One Centre",has been completed and opened for operation,while the intercity railway in the "West Wing" westward towards Lanzhou New Area is also developing steadily,while the "East Wing " The transport construction in the direction of Yuzhong has yet to take shape.In the "Outline of the 14 th Five-Year Plan and 2035 Vision for National Economic and Social Development of Gansu Province" issued by the Gansu Provincial Government in March 2021,the "planning and construction of an suburban rail transit system from Lanzhou to Yuzhong" was proposed.As a fast-speed,high-capacity,commuter rail system,the construction of suburban railways has entered a rapid development stage.Scientific planning and construction of suburban railways are all supported by passenger flow forecasting,therefore,this paper conducts research on passenger flow forecasting and related analysis of Lanzhou-Yuzhong suburban railways based on spatio-temporal data.The current road traffic characteristics of Lanzhou city are studied through Baidu Map API spatio-temporal data;a GIS-based geographic information database is constructed for the preparation of passenger corridors for the Lanzhou-Yu suburban railway;the four-stage method is improved by combining the travel characteristics of the residents in the Lanzhou-Yu region,the social economy and the current traffic situation,and the calculation of the main passenger flow indicators and sensitivity analysis are carried out for the Lanzhou-Yu suburban railway.The main work of this paper is as follows:(1)With the help of the Baidu Map open platform,a Python web crawler is used to call the Baidu Map API interface,thereby obtaining the current traffic data of the road network in Lanzhou and pre-processing it.Based on the obtained road traffic data,the basic capacity of each road in Lanzhou City and the actual traffic volume of the road sections were calculated.Based on the actual road section traffic volumes,the OD matrix is back-propagated based on the Trans CAD platform,and finally the current situation OD matrix of Lanzhou is obtained,which provides important index factors for the preparation of the passenger corridor in the following.(2)Collect data on the physical geography,social economy and population of Lanzhou,extract,analyze and store the data using GIS tools and build a regional spatial geographic database based on Arc GIS.In the regional spatial database,factors such as slope,slope direction,land use,road density,passenger flow,population density and GDP density were selected as the index factors affecting the passenger corridor of Lanzhou-Yu suburban railway.A hierarchical analysis model was established to assign weights to the index factors.Based on the selected influencing index factors,each index factor is assigned a cost by reclassification tool in Arc GIS and overlay analysis is carried out to construct a multi-cost comprehensive data set.Using tools such as cost distance analysis in GIS software,suburban railway passenger corridor from Lanzhou to Yuzhong was obtained to provide prerequisite preparation for subsequent passenger flow forecasting.(3)The applicability and shortcomings of the traditional four-stage method of suburban railway passenger flow forecasting model are analysed,and the traditional four-stage forecasting method model is improved by combining the travel characteristics of residents in Lanzhou-Yuzhong region,socio-economic attributes and the current traffic situation.In the traffic generation stage,a method based on the purpose of travel and the nature of land use in the district is used to improve the traffic generation forecasting model;in the traffic distribution stage,a method of modelling trips within and between traffic districts is used to improve the traffic distribution model.The traffic generation forecast,traffic distribution forecast,traffic modal split forecast and traffic allocation forecast are then carried out for the city of Lanzhou based on population,employment,socio-economic and travel analysis.Finally,key passenger flow indicators such as passenger flow,turnover,one-way peak cross section,passenger intensity and average distance travelled are obtained and sensitivity analysis is carried out.Passenger flow forecasting is a necessary part and an important basis for suburban railway traffic planning and design.This study uses spatio-temporal data technology to break through the limitations of the traditional data collection methods,conducts research on the current road traffic characteristics of Lanzhou City through Baidu Map API data,and prepares passenger corridors based on GIS technology,and then improves the four-stage method and conducts passenger flow forecasting by combining the regional characteristics of Lanzhou-Yu,which improves the accuracy of suburban railway passenger flow forecasting and is important for enhancing the scientific nature of urban railway planning and strengthening the It is of great significance to enhance the scientificity of suburban railway planning and strengthen the antirisk capability of passenger flow prediction.