Partial Discharge Denoising Method for Switchgear Based on CEEMD-Wavelet Threshold

Abstract

In the process of PD(Partial Discharge) signal denoising, wavelet threshold denoising and CEEMD (Complete Ensemble Empirical Mode Decomposition) denoising have problems of effective signal loss leading to waveform distortion and insufficient adaptability. A PD signal denoising method based on CEEMD-wavelet threshold is proposed. The denoising principle of the method is introduced in detail. The advantage of this method is that not only the noise in the PD pulse signal is effectively removed, but also the integrity of the effective PD signal is well preserved. According to the denoising effect of the simulated PD signal and the experimental PD signal, and the obvious advantages of the proposed method in terms of signal-to-noise ratio and amplitude error, the validity and reliability of the proposed method are verified.