Suppressing white-noise in partial discharge measurements part 2: the optimal de-noising scheme

摘要

Suppressing white noise in partial discharge (PD) measurements is a very challenging task. Complex Daubechies wavelet (CDW) transform is proved to be an effective method of noise rejection because of its shift-invariance and available of phase information. The major challenge of noise rejection based on CDW transform is the determination of the optimal mother CDW, complex threshold and combined information. In this paper, a new method of determination of the optimal de-noising scheme is proposed to solve this problem. The optimal mother CDW is determined by comparing the shift-invariance among CDW2 to CDW12. Combined information and complex threshold are determined by calculating the waveform similarity between original signals and de-noised signals as well as peak-signal-to-noise ratio (PSNR) of de-noised signals. On-site applications of extracting PD signals from noisy background by the optimal de-noising scheme based on CDW are illustrated. The on-site experimental results show that the optimal de-noising scheme is an effective way to suppress white noise in PD measurement. It preserves more energy of original signals and enables the waveform distortion small after de-noising.