Improved Hybrid Intelligent Method for Urban Road Traffic Flow Forecasting based on Chaos-PSO Optimization

Real time traffic flow is often difficult to predict precisely because of the nonlinear and stochastic characteristics of the traffic flow data. Intelligent prediction methods such as artificial neural network (ANN), support vector machine (SVM), etc. have been proven effective to discover the nonlinear information hidden in the traffic flow data. Nevertheless, their efficiency limits in the low recognition rate. If using them independently, the prediction accuracy would be lower than integrated use. Hence, a new hybrid intelligent prediction approach base on the combination of advanced signal processing technique and intelligent data mining method is proposed for the short time traffic flow prediction in this paper. The new method is marked as the use of empirical mode decomposition (EMD) and support vector machine (SVM) to deal with the nonlinear and stochastic characteristics of the traffic flow data. Another advantage of the proposed method is that the Chaos-particle swarm optimization (PSO) is adopted for the optimization of the combination. By doing so, the local optimization of the EMD-SVM prediction model can be avoided and the forecasting rate can be enhanced significantly. The practical traffic flow data were applied to the validation of the proposed prediction model. The analysis results show that the proposed method can extract the underlying rules of the testing data and decrease prediction error by 0.53% or better when compared with single SVM approach. Thus, the new hybrid intelligent traffic flow forecasting model can provide practical application.