Autocorrelation aided rough set based contamination level prediction of high voltage insulator at different environmental condition

Abstract

This paper presents a Rough Set Theory (RST) based approach for predicting the surface contamination level of porcelain type insulators at different environmental condition. The sample insulators have been contaminated by Solid layer Method (SLM) according to IEC60507. Leakage current of the insulators are recorded at different contamination level and at different humidity conditions. In the proposed method, autocorrelation technique is used for feature selection from leakage current because it is well suited for non-stationary leakage current signals and cancels out the effect of noise present in the signal. Rough Set Theory is used for classification of extracted features as it suits well with the correlation technique. Result shows that the proposed method is able to detect the contamination level with acceptable range of accuracy.