Abstract: With the development of high-frequency power electronic devices and the massive deployment of distributed new energy sources, supraharmonics have caused a series of problematic. However, it is difficult to accurately monitor supraharmonics due to its non-stationary and wide frequency domain. Therefore, a new monitoring method for supraharmonics based on dynamic sampling and compressed sensing is proposed. In sampling node, a dynamic compression sampling algorithm is designed based on a flexible time window, and the window width and sliding position can be adjusted by parameters. Moreover, the sparsity of the superharmonics signal within the time window is proved to break through the limitation of Nyquist and realize the low-speed dynamic compression sampling. In construction node, the variable step sparsity estimation subspace pursuit-dynamic basis pursuit (VSSESP-DBP) algorithm is designed for recovering signal continuously. First, the initial solution is reconstructed by the variable-step sparsity self-estimating subspace pursuit algorithm. Secondly, a dynamic basis pursuit algorithm is proposed, which takes the previous solution as priori information, and quickly solves the current value. Tests based on the wind power grid-connected model show that, compared with existing methods, the method in this paper could realize supraharmonics dynamic monitoring and accurate reconstruction with lower sampling frequency.