A SHORT-TERM TRAFFIC FLOW FORECASTING METHOD BASED ON STATE IDENTIFICATION

Abstract

For most of the short-term traffic flow forecasting model is under the single state of traffic flow prediction, unable to meet the actual traffic induced demand. Therefore, this paper puts forward the a applicable in a mixed state of discrimination of state based on short term traffic flow prediction model. The short-time traffic flow forecasting method, the wavelet analysis method for decomposition and reconstruction of history traffic data (traffic flow, speed and density) and obtain the freedom flow state and flow state and blocking flow state state of traffic flow parameters threshold congestion. According to the traffic condition criteria, the current traffic flow parameters are judged, and the corresponding forecast model is selected. The experimental results show that the short-term traffic flow forecasting model based on state estimation has a higher prediction accuracy than the single forecasting model, and it is more suitable for the traffic flow prediction.