Anzeige
Erschienen in:
2022 | OriginalPaper | Buchkapitel
Traffic Flow Prediction Model Based on Deep Learning
verfasst von : Bowen Wang, Jingsheng Wang, Zeyou Zhang, Danting Zhao
Erschienen in: Man-Machine-Environment System Engineering: Proceedings of the 21st International Conference on MMESE
Verlag: Springer Singapore
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Abstract
ARMA_LSTM model was constructed for short-term traffic flow prediction of urban road sections. Firstly, the grid search method was used to find the best parameter combination of Auto-Regressive and Moving Average Model (ARMA), so as to fit the linear characteristics of traffic flow. Then Long Short-Term Memory model (LSTM) was used to fit the nonlinear features in the reconstructed residual sequence. Experimental results show that ARMA_LSTM model has higher prediction accuracy and lower Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) values than some traditional models and artificial intelligence models at different sampling intervals. The model can be used to forecast traffic flow at different time intervals.