Short-time prediction method based on fractal theory for traffic flow

It´s very difficult to predict the nonlinear traffic flow message, especially short-time traffic flow. To solve the issue, the nonlinear fractal phenomenon of urban traffic is analyzed. Based on the fractal prediction technology applied in other fields, a dedicated improved fractal model is raised to predict short-time traffic flow parameter. A minimum fractal dimension is given according to the distinguished concrete traffic circumstance in the model. The weekly traffic similarity is also inducted to improve the accuracy of the model. Finally, the improved fractal model is employed to predict the traffic flow in Hangzhou city. The experiment result shows the improved fractal method proposed here possesses a high prediction precision.