Abstract: Based on the adaptive ensemble empirical mode decomposition (EEMD),the new method was proposed to reduce the wind power fluctuation.Firstly,the order with the largest difference of the energy entropy of the intrinsic mode function was calculated by the information entropy theory.And then,the order with the lowest self-similarity was calculated by the sample entropy theory.At last,the power distribution of the hybrid energy storage system was modified by the theory of fuzzy control optimization.The results show that the proposed strategy can adaptively realize the reasonable power distribution and grid-connected power smoothing.And the hybrid energy storage system works in the reasonable range of the state of charge,which effectively improves the system stability and service life.

Key words: ensemble empirical mode decomposition (EEMD), energy entropy, sample entropy, fuzzy control, state of charge