Transient disturbance detection based on hilbert phase-shifting and wavelet de-noising

The transient power quality detection and location is a quite important topic of the analysis of power quality. This detection method that based on wavelet transform is very popular. However, the results of wavelet detection are affected seriously by the system noise. In this paper, a new method based on Hilbert phase-shifting and wavelet de-noising is an advisable way. First, the signal with noise is processed by the Hilbert shifting transform, getting the signal envelope including noise. The envelope shows the change trend of disturbance signal. Then, removing the noise from the transient signal envelope by wavelet de-noising, the beginning and ending times are extracted exactly form the denoised signal. The method not only can detect and locate the disturbance signals accurately, but also unaffected by the system noise.