There is a problem of mode mixing in empirical mode decomposition (EMD) denoising. Thus a denoising algorithm combining variational mode decomposition and moving mean filtering is proposed. First, the noise signal is decomposed into several inherent modes by finding the optimal solution of the variational model. Then, the optimal decomposition layer K and its corresponding correlated modes are determined by the correlation coefficient criterion, and the uncorrelated modes are processed by the moving mean filter to get the useful components. Finally, the denoised signal is constructed based on the useful components extracted from the correlated and uncorrelated modes. Simulation results show that compared with EMD denoising and wavelet denoising, the proposed algorithm can effectively remove the noise in the transient disturbance while retaining the characteristic information in the transient disturbance. This work is supported by the Science and Technology Project of the Headquarters of State Grid Corporation of China (No. 52130421000S).