Research on Traffic Flow Forecasting Method Based on CATformer Model

Traffic flow forecasting is of great significance to today’ s high-speed management. The accuracy of forecasting has an important relationship with avoiding congestion. Traditional time series forecasting methods have certain limitations when dealing with nonlinear and non-stationary time series data. This paper proposes A CATformer model based on the seq2seq architecture, which introduces both CNN and TCN modules to enhance the model's sequence decomposition ability. The proposed model is compared with the other four models. The prediction results show that compared with the other four prediction models, the CATformer model have better predictive ability..