Ultra-Short Term Power Prediction Of Wind Farm Based On EEMDSE-GGRU

With the advancement of energy transformation, more and more electric power departments are required to absorb new energy such as wind power in full. The output voltage will fluctuate considerably due to the non-linear and non-stationary characteristics of wind speed itself, which will cause danger to the power system. A wind power prediction method is presented based on EEMDSE-GGRU in this paper to reduce the disadvantages of wind power. The model combines Ensemble Empirical Mode Decomposition to decompose the fluctuating time series signals, then recombine them with sample entropy. Finally, the optimal parameters of the model are obtained by unsupervised learning of the GRU model using the improved whale algorithm. During prediction, each component data is predicted in turn and the final predicted value is accumulated. The proposed method is of good prediction accuracy and stability by simulation analysis.