Abstract: In order to improve the wind farm prediction power accuracy,for wind farm and hybrid energy storage,a hybrid energy storage compensation scheme based on prediction information is proposed. Firstly,the probabilistic statistical analysis is carried out based on the prediction error,and the stratified compensation strategy of wind power prediction error is proposed,and the allowable error domain and confidence interval compensation domain are formulated for compensating the prediction error. Then,the corresponding energy storage compensation scheme is given for the prediction errors of different error layers by using energy storage,decomposing the prediction errors outside the area to be compensated based on wavelet packet decomposition,determining the critical point as the target allocation value of power-based energy storage and energy-based energy storage by the unbalanced power DFT frequency division method,and making secondary correction for the load state overrun problem by adaptive fuzzy control. Finally,a wind farm in Xinjiang is used as an example for simulation analysis,and the results show that the proposed strategy can effectively reduce the wind power prediction error.