To mitigate wind power fluctuations effectively and smoothly, a two-level fuzzy optimal control strategy for hybrid energy storage system (HESS) is proposed in this paper. Firstly, the wavelet packet decomposition method is used to decompose the outputs of the wind power, and the power distribution is carried out based on the characteristics of the HESS. Then the first fuzzy controller coordinates the charge and discharge power and the state of charge (SOC) of the energy storage system. On this foundation, the real SOC of the HESS and its reference as the inputs, a second fuzzy controller optimizes the charge and discharge power of the supercapacitor for a second time, and realize the smooth outputs of the active power of the wind farm simultaneously. Finally, based on the actual wind farm data, the mathematical simulation model is built with Matlab/Simulink, and the control strategy is proved availably by simulation analysis.