A new approach based on a nonlinearity measure to study the leakage current signature is proposed. Using this approach profiles for clean insulators, contaminated insulators with various contamination levels in dry and wet conditions are established. The nonlinearity measure captures a range of magnitudes for leakage currents exhibiting different signatures. The insulator nonlinearity measure decreases when the insulator is either wetted, contaminated or both wetted and contaminated. The change in this measure is significant in identifying the 'health state' of contaminated insulators and in flashover prediction, especially since contamination is one of the main causes of insulator failure, which thereafter may cause power system outages and threaten life. Further, this approach has the added advantage of application in identifying other types of disturbance waveforms which model-based and spectrum analysis methods lack.
↧