Abstract: Based on the negative impact of stochasticity and volatility of wind power on the stability of power grid operation, an improved function filter of wavelet threshold with limiting links is proposed, and the improved sparrow search algorithm is used to optimize the improved wavelet threshold function based on the regulatory factors of different engineering models, and the grid-connected power curve is obtained after the fluctuation is smoothed to meet different demands. After the wind power is smoothed, the grid-connected power curves under different scenarios are evaluated using accumulation of energy storage output and Euclidean distance of index fluctuation flattening. The numerical examples show that the proposed improved wavelet threshold function limit-filtering can well smooth wind power fluctuations, and limit grid-connected power below the MPFR limit of 1 min and 10 min scale in China, and the weight of the objective function can be adjusted according to different industrial needs to obtain the corresponding grid-connected power. At the same time, the superiority of NTSSA algorithm in the optimization performance is verified.