Research On Traffic Flow Prediction Of Expressway Toll Station Based On Big Data

In recent years,with the rapid development of social economy,the traffic pressure of expressway is increasing,and the congestion phenomenon of expressway toll station is becoming more and more obvious.The emergence of intelligent transportation system can make traffic management effective,so as to improve the traffic capacity of the road and improve the service level.On this basis,accurate traffic flow prediction is particularly important.Therefore,this thesis takes the highway toll station traffic flow prediction as the main research object,through the deep analysis of the highway toll station traffic characteristics,establishes the genetic algorithm optimization BP neural network(GA-BP neural network)model to predict the highway toll station traffic flow in the future,and carries out capacity analysis.The main contents are as follows:Firstly,the thesis preprocessed the toll data and analyzed the traffic flow characteristics.The original data in text format was imported into MYSQL database,and the original data was preprocessed by reducing redundant data,repairing missing data and correcting noise data,etc.,so as to obtain relatively accurate traffic flow data.According to the basic data,the general characteristics and temporal and spatial characteristics of the expressway traffic flow were analyzed respectively,so as to lay a good data foundation for the subsequent expressway traffic flow prediction and traffic capacity research.Secondly,on the basis of the existing BP neural network model,by analyzing the advantages and disadvantages of the BP neural network model,the genetic algorithm was used to optimize the threshold and weight of the network,so as to avoid the problem that the BP neural network was easy to fall into the local minimum,so as to improve the accuracy of the model prediction and achieve better prediction effect.Based on the analysis of the factors affecting the traffic flow prediction of airport toll station,the GA-BP neural network prediction model was constructed by MATLAB.Finally,taking the actual gate frame charging data of Shijiazhuang Airport toll station as an example,the related charging characteristics of toll station were investigated,and the performance of BP neural network and GA-BP neural network model was analyzed and compared.The comparison results showed that the prediction accuracy of GA-BP neural network model was increased by 5.8%compared with that of BP neural network.The average relative error was reduced by0.056,which verified that the prediction performance of GA-BP neural network model was good.This model was used to predict the future traffic flow of airport toll station,and M/G queuing model was used to calculate and study the capacity of toll station lane.The actual peak-hour traffic capacity of the toll station was determined to be 2 350 pcu/h.Compared with the predicted traffic flow of 2 200 pcu/h in the future,the results showed that the traffic demand could be met.