A Denoising Algorithm for Partial Discharge Measurement Based on the Combination of Wavelet Threshold and Total Variation Theory

In electrical engineering, partial discharge (PD) measurement is frequently employed to detect insulation defects and judge insulation conditions of high-voltage electrical apparatus. However, it is easily corrupted by white noises in the field. In this article, a joint algorithm is proposed, in which the wavelet threshold and total variation (TV) denoising methods are combined by the convex optimization theory to denoise ultrahigh frequency (UHF) PD signals corrupted by white noises. Since the two respective methods are incorporated into the joint algorithm, it is with high potential to reduce oscillation error introduced by the wavelet threshold method and eliminate stair error introduced by the TV denoising method. In order to validate the effect of the proposed algorithm, a numerical simulation is carried out to compare it with several existing methods. Indicators of their performance are computed, and the results verify that the proposed algorithm outperforms all the other methods.