Application of wavelet-based de-noising to online measurement of partial discharges

For online detection of partial discharge (PD) activity within power cables and cable accessories, one of the non-conventional PD detection techniques, VHF capacitive coupling has been proved to be suitable. However, the existence of excessive interference can significantly influence the measurement sensitivity and reliability. This paper investigates the application of wavelet transforms for de-noising signals obtained using capacitive couplers. The experimental work has been carried out on a 132 kV cable loop with a known defect within the cable joint. Obtained results indicate that the wavelet analysis technique can effectively discriminate internal PD pulses among corona discharges and pulse-like interference. Further, removal of interference has been achieved by applying level dependent threshold values.