The power fluctuation is serious when wind power generation is connected to network, which will bring negative impact on the economy, security and reliability of the power system. Through the control of energy storage system, the output power of wind farm can be suppressed. Based on the historical power output data of a wind farm, the target signal is obtained by wavelet decomposition. The charging and discharging model of the energy storage system is established, and the control strategy is also formulated. Energy storage capacity configuration is carried out with the optimal objective of best economy of energy storage system. To consider the economy and stabilization effect of the energy storage system, the probability of wind power output fluctuation meeting the grid connection requirements is introduced into the target model as a flexible constraint by chance constrained programming. The method of modifying the stabilization target in the future time period according to the current grid connected power is also studied. Finally, the genetic algorithm is used to configure the optimal energy storage capacity through simulation experiments. The results show that the proposed energy storage capacity allocation method can greatly reduce the economic cost of energy storage system, and the wind power fluctuations can meet the grid requirements at a high level of confidence.