Study on Joint Noise Reduction Method of Local Discharge Signal Based on SVD and Parameter Optimized VMD

Difficulty of subsequent identification and localization of transformer partial discharge signals in multiple noise backgrounds, In this paper, a joint noise reduction method based on SVD and parameter-optimized VMD for local discharge signals is proposed. Firstly, Singular Value Decomposition (SVD) is performed on the original local-amplified signal, and the effective singular values are selected and reconstructed to obtain the preliminary noise reduction signal after suppressing the periodic narrowband interference; Secondly, the preliminary noise reduction signal is processed by Variational Mode Decomposition (VMD) optimized by Sparrow Search Algorithm (SSA) to obtain a series of Intrinsic Mode Function (IMF), which removes the small variance contribution of the modes; IMF is used to remove the small variance contribution of the modes. IMF), and the modes with small variance contributions are removed; Finally, the improved wavelet thresholding method is applied for the residual modes for denoising and the denoised signal is obtained. The analysis results of the simulated signals show that the method can significantly suppress the periodic narrow-band interference and white noise, which provides effective support for the subsequent localized pattern recognition and localization work.