Error Correction and Wavelet Neural Network Based Short-term Traffic Flow Prediction

摘要

Real-time and accurate short-term traffic flow prediction is an important part of intelligent transportation system research. Wavelet neural network is a preferable method for predicting traffic flow. However, its performance is not satisfactory since it's easy to fall into local optimum. This paper proposed an Error Correction Wavelet Neural Network prediction method (EC-WNN) to predict short-term traffic flow. First, we use Wavelet Neural Network to predict the traffic flow, and build the error prediction model of Auto-Regressive Integrated Moving Average (ARIMA) based on the error series. Then we use the prediction errors to update the prediction results. Finally, the real detected traffic data are used to evaluate the precision of the model, the results show that EC-WNN is superior to traditional WNN in accuracy of prediction.