The hybrid energy storage system can effectively alleviate the frequency instability caused by the strong fluctuation and randomness of wind power output. In this paper, a hybrid energy storage system composed of batteries and super capacitors is taken as the research object, and a hybrid energy storage capacity allocation method is proposed. First, adaptive wavelet transform is adopted to perform a primary distribution of the wind power output, and the grid-connected power and energy storage power satisfying the requirements are obtained. Second, HHT transform is used to decompose the energy storage power, and a series of fluctuating power components and the instantaneous frequency of each component are obtained. Third, the cutoff frequency is determined according to the instantaneous frequency, the power components with a frequency higher than the cutoff frequency are allocated to super capacitors, and the rest are allocated to batteries. Finally, the rated capacity and rated power of the energy storage system are configured according to the energy storage power of batteries and super capacitors, respectively. Simulation results show that adaptive wavelet transform and HHT transform can effectively decompose the wind power output, thus realizing the stabilization of wind power output, as well as the capacity and power allocation of hybrid energy storage system.