Artificial intelligence methodology for separation and classification of partial discharge signals

Results of investigations performed in order to improve the current diagnostic techniques used for the evaluation of insulation systems of HV apparatus are presented in this paper. Improvements come from the development of a new measuring system which allows the digital acquisition of Partial Discharge (PD) signals and a new separation method, based on a Fuzzy Classifier, for the analysis of the PD-pulse shape signals. The identification of the classes, relevant to different PD phenomena, is then performed by means of PD-pulse height and phase analysis. The proposed approach is supported by the analysis of PD data obtained from insulation systems of stator bars with artificially-reproduced defects.