Analysis Of Temporal And Spatial Characteristics Of Urban Traffic And Identification Modeling On Key Nodes And Sections Of Road Network Based On Taxi GPS Data

In recent years,with the rapid growth of car ownership,urban road traffic congestion has become an increasingly serious problem.Traditional methods such as expanding the number of urban roads are unable to meet the increasing traffic demand anymore.With the popularization of GPS equipment,taxi GPS data provides a new data source for the study of urban traffic problems.It is an important way to ease the traffic congestion by using the floating car data to analyze the temporal and spatial characteristics of urban traffic and identify the key nodes and sections of urban road network.This paper uses the GPS data of more than 40,000 taxis from January 12 to 18,2015 in Beijing,firstly systematically analyzes the temporal and spatial characteristics of Beijing taxis from the aspects of overall travel characteristics and hotspot mode.Then,based on the directed weighted complex network,the identification of key nodes and sections of the urban road network is carried out from the aspects of road network topology and road network traffic attributes.The main research contents include the following three parts:(1)On the basis of data processing,this paper analyzes the overall travel characteristics,travel hot spot mode,and travel hot trend of taxis with statistical analysis and space-time cube model.First of all,the overall travel characteristics of taxis in Beijing are studied in this paper by using the methods of statistical analysis and nuclear density analysis.Then,based on the space-time cube model,the hot spot pattern recognition and hot spot trend analysis are carried out from two aspects: the overall travel trajectory and the travel trajectory of passengers.It is found that taxi travel trajectory indicates obvious daily cycle effects,and the working day and weekends demonstrate two kinds of spacetime characteristics.The paper also founds that taxi travel hot spots can be divided into five types.In addition,the hot spot mode and hot spot trend of the taxi travel trajectory both show a big difference over time.(2)Based on the directed weighted complex network model,the identification model of key nodes and sections of the urban road network is constructed from the perspective of road network topology and traffic attributes.Firstly,on the basis of previous research,the definition of key nodes and sections is proposed.Then,the weighted complex network is used to construct the urban original road network model and dual road network model based on the identification of key nodes and sections.Furthermore,the evaluation index system of key nodes and sections is constructed from the aspects of traffic attributes and road network topology,and each index is weighted according to the improved analytic hierarchy process.Finally,a set of key nodes and sections identification algorithm is proposed.(3)Taking Xicheng district of Beijing as the research object,the key nodes and sections of Xicheng District are identified.First of all,the basic situation of Xicheng District is described,and the traffic congestion situations of each grade of road in Xicheng District on weekdays and weekends are compared and analyzed.Then,based on the GPS data and the proposed model,the identification cases of key nodes and sections of urban road network in different periods are illustrated,and the identification results are compared with the traditional methods that only rely on the static characteristics of complex network and the methods that only consider traffic attributes.The results show that the identification model proposed in this paper realizes the expression of both topology structure and traffic attributes of road network,therefore is more practicable and effective in the large sense.Finally,according to the identification results,the paper put forward the management suggestions for key nodes and sections of urban road network.