Short-term traffic volume prediction using GA-BP based on wavelet denoising and phase space reconstruction

Abstract

Accurate traffic volume prediction can help traffic managers to control traffic well,and can also provide convenient travel routes for passengers. In order to better describe the non-stationary, complex and spatial correlation of traffic volume, the traffic prediction model is proposed based on wavelet denoising and phase space reconstruction (WD-PSR-GA-BP). Pure traffic volume is firstly preprocessed by wavelet denoising method. Subsequently, the prediction model is built by BP neural network, which is optimized by genetic algorithm. In addition, one-dimensional traffic volume is mapped to high-dimensional space by the theory of phase space reconstruction. The inputsof the prediction model are obtained by the reconstructed traffic volume datasets. In order to verify the effectiveness of the proposed model, two groups of datasets and different models are studied in the experiment of Section 4. The experimental results show that the proposed model is superior to all other competitors in terms of MAPE, RMSE, and MAE.