Abstract: With the increase of wind power installed capacity in China,the randomness and fluctuation of wind power make it more difficult for the frequency adjustment of power grid. To solve such problem,a hybrid energy storage system composed of energy-type（lithium-ion batteries）and power-type（flywheel）energy storage arrays can be used to smooth the wind power output. Therefore,it is urgent to study the capacity configuration method of the hybrid energy storage system to increase its utilization rate and reduce its requir ment capacity. Firstly,the 250 kW/50 kWh flywheel and 250 kW/150 k Wh lithium-ion battery energy storage unit models are established,and corresponding control strategies are introduced. Secondly,according to the characteristics of the energy storage system and the amplitude-frequency characteristics of the output power of the wind farm,an improved wavelet packet decomposition method is proposed to decompose the output power data of the wind farm and obtain the overall charging and discharging power commands of the hybrid energy storage system. Thirdly,the distribution method of charging and discharging power commands for each type of energy storage array are obtained through the response time of different types of energy storage arrays and the energy distribution of each node with the improved wavelet packet decomposition. Finally,the fitting function is adopted to fit the probability distribution characteristics of the charge and discharge power commands above mentioned. Based on the limitation of confidence interval under a certain confidence level and the introduction of measurement factor for correction,the capacity configuration method for the hybrid energy storage system is proposed to determine the number of various energy storage units,which can increase the utilization rate of hybrid energy storage system and reduce its requirement capacity. Based the actual output power data of a wind farm with the rated installed capacity of 15 MW,the developed capacity configuration method of the hybrid energy storage system is verified through simulation.