Research on Wind Power Fluctuation Smoothing Strategy Based on K-means Clustering and Empirical Wavelet Decomposition

With the increasing installation of wind power capacity in China, the randomness and fluctuation of the output power make it more difficult for the power grid to adjust. To address this issue, it is necessary to use energy storage systems to smooth out wind power and configure the power and capacity of the energy storage system. This has significant implications for increasing the utilization rate of power plants and reducing the storage capacity. K-means clustering algorithm is used to classify massive power plant data, and the resulting data is decomposed using empirical wavelet methods. The grid connected power and energy storage system smoothing power are obtained based on the fluctuation limits of the wind field, and the validity and applicability of the proposed smoothing method are verified through a case study of a wind farm.