Abstract:In view of the nonlinear and uncertainty characteristics of short-term traffic flow data, a short-term traffic flow prediction model based on ensemble empirical mode decomposition (EEMD) and random forest (RF) was proposed. Firstly, the space mean speed sequences of the original traffic flow data was decomposed into several intrinsic mode functions (IMF) and a residual component (RES) by using the EEMD algorithm, which extracted the information of traffic flow data at different time frequencies.Next, The first component is decomposed by EEMD to refine the random information of traffic flow data.Secondly,each component after decomposition was predicted by using RF,which can obtain several submodels. Finally, the prediction values of all submodels were summed linearly to obtain the final prediction results. The actual traffic flow data of a certain section road in Alar City was used for experiments. The results show that the prediction performance of the EEMD and RF combined model is better than that of the single RF model, and the secondary decomposition of high-frequency sequence can further improve the accuracy of the combined model.