Study On Short Term Traffic Flow Forecasting Method In Nanchang

Traffic flow forecasting is an important part of intelligent transportation system.The important premise of traffic signal control,traffic assignment and route guidance is fast and accurate short-term traffic flow forecasting.However,the traffic system has high complexity,nonlinearity and uncertainty characteristics,it is a complex system composed of human,vehicle,road and other objects,real-time,accurate prediction of traffic flow is one of the hot and difficult research in the field of intelligent transportation.However,due to the short-term traffic flow information,uncertainty of noise signal interference,and because of the complex topology of city road network,that how to realize the city road traffic flow forecasting has hindered the long-term development of intelligent transportation.In order to solve the above problems,many prediction methods have been proposed,but because of not considering the influence of uncertain complexity caused by interference signal or city road network of short-term traffic flow,leading to the accuracy of the prediction results and the real-time performance is not ideal.This paper adopts Mallat algorithm,wavelet decomposition and reconstruction of the short-term traffic flow signal,strong noise interference signal to filter out the short-term traffic flow,this method can improve the short-term traffic flow information preprocessing speed and accuracy.Traffic flow data are highly complex and nonlinear,and the neural network has strong nonlinear processing ability,self-organization,self adaptation and self-learning ability.In summary,the city road traffic flow variability and complexity and nonlinear characteristics,in order to improve the short-term traffic flow prediction accuracy,this paper proposes a wavelet denoising and short-time traffic optimization BP neural network adaptive genetic algorithm combined forecasting model based on flow.Wavelet transform can be decomposed into a smooth traffic flow with different frequency sub sequences,and then the sub sequences were predicted,which can effectively solve the traffic flow prediction is time-varying,complicated and nonlinear problems,and adaptive genetic algorithm with global searching ability,can solve the defects of the neural network is very good-easy to fall into local minimum.The prediction results are compared with the results of wavelet neural network and BP neural network based on genetic algorithm.The results show that the proposed model average absolute error,mean absolute percentage error and root mean square error is smaller,and the fitting degree of EC was larger,which verified the model in short-term traffic flow prediction on the validity and accuracy.