Traffic Flow Forecasting Based On Artificial Bee Colony Algorithm And Neural Network

With the continuous development of China's economy,the improvement of the level of urbanization and the frequent traffic congestion,it has greatly affected the people's work and life.The intelligent transportation system improves the efficiency of transportation through real-time feedback and efficient control of information,and can reduce the traffic congestion.As an important part of ITS,traffic flow forecasting has deep theoretical and practical significance.In this thesis,the artificial bee colony optimization algorithm is used to optimize the BP neural network and to establish the model of traffic flow prediction,which is devoted to improving the accuracy of the prediction.The specific work is carried out from the following points:(1)This thesis first describes the status quo of domestic and international traffic flow forecasting and various models and related theories,and summarizes the characteristics of various methods.(2)For the raw data that is not processed,the Lagrange interpolation method is used to repair the abnormal data,and then the original data is denoised using the wavelet analysis method.(3)Then,this thesis analyzes the advantages and disadvantages of BP neural network,and selects the artificial bee colony algorithm according to the flaw of BP neural network to optimize its weight and threshold.(4)Aiming at the traffic flow data after pretreatment,this thesis adopts BP neural network,Elman neural network and BP neural network based on artificial bee colony optimization algorithm respectively to build prediction model and predict training.According to the model evaluation formula,the prediction performance of the three models is quantitatively calculated.Through comparative analysis,it is concluded that the BP neural network based on artificial bee colony optimization algorithm has the best prediction result and the best fitting degree,which can be used for traffic flow prediction.