Wind speed forecasting based on EEMD and ARIMA

This paper proposes a prediction model based on Emsemble Empirical Mode Decomposition (EEMD) and Autoregression Integrated Moving Average (ARIMA) model for the characteristics of the wind speed as the nonlinear and non-stationary sequence. Firstly, the wind speed time series is decomposed into a number of Intrinsic Mode Functions (IMFS) and one residual series which are smoother than the original sequence using EEMD. Then the ARIMA model is applied to forecast the IMF and residue series. Finally, the prediction result of the wind speed is obtained by summing the predicted results of each IMF and residue component. The results gained in this paper show that the prediction accuracy of EEMD-ARIMA model is higher than that of EMD - ARIMA model and ARMA model.