Application of wavelet analysis to feature extraction of noisy corona discharge signals

As transmission line voltages increased in time, corona occurring on conductors had more importance in terms of both corona losses, their impact on power transmission as well as leading to electromagnetic interference. One of the major challenges of corona discharge measurements is separation of corona signals from different type of noises resulting from measurement circuit or surrounding environment. In this paper, wavelet analysis based de-noising was used to separate corona discharges from noisy data obtained during the measurements. For that purpose, necessary procedure for accurate de-noising of corona discharge signals was defined. Obtained results by following the wavelet based de-noising procedure indicate that de-noising and future extraction of corona signals acquired from high-voltage devices by using the proposed method is successful to remove noise from the measured data as effectively as possible while preserving the signal features.