To obtain contamination characteristics of insulators in safe zone of leakage current

the leakage current waveforms of XP-70 insulator string under four contamination levels were extracted through a series of experiments in laboratory. The original signal was denoised by the self-adaptive threshold method to achieve three characteristic quantities of leakage current in time domain

namely

the mean value

the maximum value

and the standard deviation of the RMS value. Analysis suggests that these three characteristic quantities show exponential relation with the increase of ESDD

respectively. Moreover

a contamination level forecast model of insulators was established based on probabilistic neural network by choosing 400 groups of leakage current RMS values as training and recognizing samples. Simulation shows that the forecast accuracy of the model is 87.5%. This research may facilitate clearing decision for polluted insulators

and may provide a theoretical basis for optimal selection of characteristic quantities for pre-warning system of transmission line contamination flashover.