| With the growth of freeway mileage,the traffic management problems of the freeway are becoming more and more serious.The analysis and prediction of freeway traffic flow have become a hot topic in recent years.The freeway traffic flow usually shows the characteristics such as cyclicality,uncertainty,temporal correlation,non-stabilization and high nonlinearity.The main factors which affect freeway traffic flow are road construction,traffic accidents,weather conditions,and major holidays and so on.Because of the lack of specifically detailed data of road construction and traffic accidents,this project only researches on using the freeway traffic flow factor of weather conditions to forecast the freeway traffic flow during the non-holidays.This paper designs the freeway traffic flow analysis and prediction system,and realizes the traffic flow information management,weather information management and traffic flow prediction function.The freeway traffic flow information is obtained through the Java Socket UDP technology from the freeway traffic flow collection system.Weather information is obtained through the standard HTTP API interface from the K780 data site.The traffic flow information,weather information statistics and analysis function are realized through the Echarts technology.According to the characteristics of freeway traffic flow and the degree of the weather information influencing on the freeway traffic flow,we analyze that time,road surface influence rank which is caused by weather conditions and air visibility influence level are the main influence factors.The pretreatment process of traffic flow and the weather information are explained in detail.This paper presents a prediction model of freeway traffic flow based on BP neural network,and takes G2 freeway qingxian main line station for example to do the traffic flow forecasting experiment.The training and prediction process of the experimental model are described in detail and the results are analyzed.The experiment shows that the forecasting model has high prediction accuracy.The achievements of this project can facilitate the traffic management department to grasp the traffic flow situation,provide the basis for real-time traffic flow guidance,diversion and information release,greatly reduce traffic congestion and ensure the smooth flow of the freeway.Our research can also provide decision basis for the national freeway planning and construction. |
