A Spatiotemporal Correlative Study On Travel Flow Distribution With Expressway Toll Data

The expressway serves as a transport carrier connecting the cities,the construction of expressway and the development of city complement each other.The long-term spatiotemporal effect of expressway network has a significant impact on the travel flow between cities.The expressway toll data records OD information of vehicles in and out expressway toll stations all day,which is easy to obtain,complete in content,strong in timeliness and high in accuracy.Therefore,the massive toll data obtained from toll collection systems can provide us spatiotemporal traffic flow information of the whole expressway network.To describe the relationship between urbanization and expressway travel flow qualitatively and quantitatively from the perspective of time and space can not only better predict the OD travel flow of expressway,but also better understand the internal effect of urbanization on the spatiotemporal distribution of expressway travel flow,which is meaningful and valuable for urban planning and regional traffic development.This paper mainly completes the following work:First of all,in the data preprocessing,a method combining the characteristics of toll data is further proposed to eliminate the abnormal data from the expressway's own properties according to the main reasons for the abnormal data of expressway toll data.And the feasibility of the method if verified by the actual toll data.Secondly,the expressway traffic is divided into passenger flows and cargo flows according to the types of vehicles,which represent the mobility of residents and cargo respectively.The travel flow on expressway network is related in time and space,and the distribution and evolution of traffic flow is a spatiotemporal phenomenon,which is difficult to describe travel accurately from the point of view of single time or space.Therefore,this paper excavates the toll data from different angles of time and space,and the spatiotemporal correlation is determined by analyzing the structure of expressway network and the characteristics among the time series of actual toll data.Thirdly,this paper proposes a spatiotemporal correlative regression model based on travel flows between cities considering the spatiotemporal effects from the micro and macro perspectives,and the marginal effects are calculated to measure the spatiotemporal effects of expressway travel flow and the influence of urbanization on expressway travel flow quantitatively.In this paper,the description effect and applicability of the model are improved by combining the characteristics of the data itself.The results show that the spatiotemporal correlation of the origin dependence,destination dependence and OD dependence can significantly affect the travel flow,and if we ignore the spatiotemporal correlation will lead to an overestimation to the impact on expressway travel flow of GDP and the proportion of urban population,and an underestimation of accessibility in the urbanization variables.The quantitative measurement of the interaction and effect between urbanization variables and expressway travel flow reveals the urbanization's impact on the spatiotemporal distribution of expressway travel flow,which is helpful to the coordinated development of expressway planning,operation,management and distribution of urban resources.