| Road traffic flow prediction is the core of intelligent transportation systems and traffic information services, traffic control and induced an important foundation. Intelligent traffic system control and induced the need for accurate road network, traffic flow, rapid short-term prediction and discrimination. Therefore, the study of effective traffic flow analysis and prediction of the theory and method, from the information obtained quickly and accurately predict traffic flow and determine the status of the current urgent need for the development of intelligent transportation systems, is to study the important and difficult problem. In this paper, real-time data for forecasting short road network theory and method of the title in the traffic flow data based on the analysis, a path based on fractal theory model of short-term traffic flow forecasting. The basic idea is:the distribution of traffic has a certain self-similarity, so called self-similarity refers to the part and the whole in form, function, information, time and space has a statistically significant similarity, and fractal theory has just to study a powerful tool for self-similarity.This paper first described the concept of short-term traffic flow forecasting, basic processes, characteristics and requirements, review the domestic and foreign short-term traffic flow forecasting research. Then, we introduce the concepts of fractal, the various definitions of fractal dimension and the traffic data of the Hurst exponent and correlation dimension calculation shows that traffic flow with fractal geometry. Finally, Traffic Software Integrated System(TSIS) simulation platform for the forecast model. With examples of the application of the forecasting model of traffic flow forecasting to describe the specific methods and steps, given prediction, the results were analyzed. Simulation results show that the method of fractal forecasting short-term prediction of traffic flow can be achieved relatively good results. |
PDF Full Text Request