The accurate prediction of real-time traffic flow is the premise of realizing dynamic traffic control and guidance. In view of the nonlinear and stochastic characteristic of traffic flow, a model which combines auto regression moving average (ARMA) and Elman network based on wavelet analysis is proposed. First, original traffic flow series are decomposed by wavelet into a low frequency signal and several high frequency signals. Then the decomposed signals are reconstructed respectively. The reconstructed low frequency signal is smooth but nonlinear and nonstationary, so it is predicted with Elman network. The reconstructed high frequency signals are stationary, so they are predicted with ARMA models. Finally, the respective prediction results are synthesized to obtain the final prediction result. Using the traffic flow data of Maryland free way, the simulation experiment is conducted based on the proposed model. The simulation results demonstrate the validity and practicability of the proposed model.
A Combined Model for Traffic Flow Prediction Based on Wavelet Analysis
Second International Conference on Transportation Engineering ; 2009 ; Southwest Jiaotong University, Chengdu, China
29.07.2009
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Highways and roads , Information technology (IT) , Transportation management , Construction , Freight transportation , Traffic flow , Predictions , Water transportation , Air transportation , Rail transportation , Wavelet , Stochastic processes , Optimization , Public transportation , Traffic management
A Combined Model for Traffic Flow Prediction Based on Wavelet Analysis
British Library Conference Proceedings | 2009
|Prediction of Urban Rail Traffic Flow Based on Multiply Wavelet-ARIMA Model
Springer Verlag | 2017
|Prediction of Urban Rail Traffic Flow Based on Multiply Wavelet-ARIMA Model
British Library Conference Proceedings | 2018
|British Library Conference Proceedings | 2020
|