A Method of Second Wavelet Transform Automated Threshold for Partial Discharge Signal Extraction

The existence of partial discharge in high voltage insulation can lead to insulated degradation, and even breakdown. Wavelet transform has the characteristics of multi-resolution analysis, and it has time-frequency locaLization characteristics, which is suitable for smooth, the non-stationary signal analysis. In paper, Auto threshold algorithm of second wavelet transform is used in extraction of partial discharge. Exponential damping and exponential damping oscillating partial discharges are used in numeric simulation, and effectiveness of extraction of partial discharge by first wavelet transform and second wavelet transform is studied. Numerical simulation results show that wavelet transform can effectively extract partial discharge from random noise because it has multi-resolution, and second wavelet transform auto threshold reserves more features of original partial discharge than conventional de-noise method of first wavelet transform. The method of auto threshold second wavelet transforms process swift and it is adaptive to dispose on Line and off Line.