Hybrid Model Based on EEMD, ARMA and Elman for Photovoltaic Power Prediction

To increase the accuracy of photovoltaic (PV) power prediction results, a hybrid prediction model based on ensemble empirical mode decomposition (EEMD), autoregressive moving average (ARMA) and Elman neural networks is proposed in this paper. To reduce the complexity of the PV power signal, EEMD is first used to decompose the PV power signals. Then the all components are divided into low frequency and high frequency components according to the zero crossing rate. The ARMA and Elman are used to predict the low frequency and the high frequency components, respectively. Finally, the results of the component predictions are superimposed. The experimental results demonstrate that the proposed method has higher prediction accuracy of PV power.