Abstract: Aiming at the problem that wind power fluctuation reduces the consumption level of power grid, a power distribution strategy of hybrid energy storage system is designed, which adopts the GWO(Grey wolf optimizer) to optimize and improve ICEEMDAN(Improved complete ensemble empirical mode decomposition with adaptive noise). Firstly, using the cross-correlation coefficient between the wind power grid-connected power and the original power, and the sample entropy of the intrinsic modal function(IMF) obtained by ICEEMDAN decomposition as the fitness function, GWO is used to calculate the parameter signal-to-noise ratio in the ICEEMDAN algorithm. The sum of μ and the boundary point k of high-frequency and low-frequency power components are optimized. Secondly, the wind power is decomposed by ICEEMDAN, the low-frequency IMF signal is used as the wind power grid-connected power, the high-frequency IMF signal is used as the power of the hybrid energy storage system, and the information entropy of each adjacent high-frequency IMF signal is used as the basis to realize the power of the hybrid energy storage system. According to the state of charge of the super capacitor, the power of the battery and the super capacitor is corrected by fuzzy control to realize the secondary distribution of the power of the hybrid energy storage system. Finally, the above-mentioned wind power fluctuation control strategy is compared with other wind power mitigation strategies to verify the effectiveness and superiority of this method.