论文信息 - Research on the Combination Model of Short-Term Traffic Flow Forecasting

Research on the Combination Model of Short-Term Traffic Flow Forecasting

Short-term traffic flow is difficult to predict accurately and real-time, owing to the characteristics of very complexity, randomness, nonlinearity and uncertainty, etc.. In this paper, the method of combining multiple linear regression with back propagation (BP) neural network was proposed, using BP neural network to compensate the model error of multiple linear regression. The combination model and the corresponding algorithm program was made, and used to pedict the short-term traffic flow. Two different methods of selecting the input layer parameters were used and compared, while the new method has higher accuracy and stability.

Hong Wei Li | Wu Sheng Hu | Yuan Lin Liu | Su Lan Li

[1] Li Li,et al. The short-term traffic flow prediction based on neural network , 2010, 2010 2nd International Conference on Future Computer and Communication.

[2] Michael J Demetsky,et al. TRAFFIC FLOW FORECASTING: COMPARISON OF MODELING APPROACHES , 1997 .

[3] Mark Dougherty,et al. THE USE OF NEURAL NETWORKS TO RECOGNISE AND PREDICT TRAFFIC CONGESTION , 1993 .

[4] Ma Zeng-qiang,et al. Short-Term Traffic Flow Prediction Based on ANFIS , 2009, 2009 International Conference on Communication Software and Networks.

[5] Li Li,et al. Short-Term Traffic Flow Prediction Based on Genetic BP Neural Network , 2010 .

[6] Thomas Urbanik,et al. Short-Term Freeway Traffic Volume Forecasting Using Radial Basis Function Neural Network , 1998 .