Spatial And Temporal Distribution Visualization Of Transfer Passenger Flow And Site Clustering Analysis In Urban Public Transport

In recent years,with the growth of China's economy and the construction of cities,the population of cities is also increasing rapidly,and the growth of population also brings great pressure to the traffic.At present,the development of public transport in many Chinese cities is far from meeting the daily travel needs of residents,and social development and public transport construction cannot be synchronized.In this case,prioritizing the development of urban public transport has become an effective strategy to alleviate traffic pressure.In the development of public transportation,rail transit and conventional bus play an important role.Therefore,mining and analyzing the travel characteristics between urban rail transit and conventional buses,and clustering research on rail transit stations based on this,is of great significance for making reasonable urban traffic planning.The emergence of traffic big data also provides a new idea for solving urban traffic problems.With the wide use of smart card technology,data obtained from the system is also applied to all aspects of urban public transportation.In this context,this paper USES the card swiping data of intelligent card to study and analyze the temporal and spatial distribution rules of transfer passenger flow between subway and bus,conducts clustering analysis on rail transit stations by combining transfer passenger flow and the station's own properties,and proposes transfer connection strategy.Specifically,the main research of this paper is as follows:(1)This paper proposes a new transfer time,and a method for determining the threshold extraction based on transfer time change to the passenger flow between rail transit and conventional public traffic,and the time-space distribution rule of the change to the passenger flow visualization analysis,managers can use visual results more intuitive understanding of the characteristics of passenger flow distribution,as well as clustering laid the foundation for the next site.(2)A clustering model of rail transit stations based on the improved k-means clustering algorithm is designed.The initial variables were standardized by the z-score method.The dimensionality was reduced by the principal component analysis method.Finally,the improved k-means algorithm was used for the site clustering analysis.(3)This paper takes Shenzhen metro as an example study,conducts site clustering based on the analysis of card swiping data of 124 stations in 5 lines of Shenzhen metro,and uses hierarchical clustering method to verify the clustering results of k-means.Finally,the station is divided into four categories: commercial station,residential station,peripheral traffic hub station,and general station,and the bus connecting strategy is formulated based on the analysis of transfer passenger flow and transfer rate.It provides reference for the planning and construction of urban rail transit and the development of the area around the station.