Research On The Prediction Method Of Flashover Based On Leakage Current Of Polluted Insulators

Compared with porcelain insulator or glass insulator, composite insulators possess of many excellences, including its light weight, good seismic performances, excellent anti-contamination performances, expedient maintenance, which is widely used in power grid. High voltage composite insulators contribute for the substation and the transmission line, providing electric isolation, supporting some power equipments, and carrying the high-voltage transmission line. But the composite insulators consist of a polymer material, which bond is weaker than the ceramic material, and thus more easily aging of composite insulators than ceramic insulators. Subjected to effects of stress and chemical during operation leads composite insulators to accelerating aging, the loss of hydrophobicity, anti-fouling flashover performances deteriorates, decreasing the insulation reliability, resulting in the occurrence of flashover accident. Insulator leakage current can comprehensively reflect the insulator voltage level and contamination state, which takes effect throughout the entire development process of the insulator flashover. Based on the study of leakage current characteristics can predict the surface contamination of the insulator and provide an effective pre-warning insulator pollute flashover.This paper is based on artificial pollution tests and studies on silicon rubber composite insulators, extracted leakage current signal at different flashover development stages of mild, moderate, severe contamination. Study on the development process of the partial discharge arc and leakage current waveform characteristics and established a correspondence between the partial discharges and leakage current waveform. Based on leakage current waveform can preliminary estimate the surface discharged state of insulators.Five time-frequency characteristic values were extracted:leakage current RMS, maximum, the third harmonic of the fundamental amplitude ratio, the sum of 3,5 times harmonic of the fundamental harmonic amplitude ratio, waveform deviation factor, according to the time domain and frequency domain analysis of leakage current. The time characteristic values can be used to pre-warning contamination under low humidity, while the frequency characteristic values can be used to pre-warning contamination at higher humidity. That there exists some change rules between leakage current waveform deviation factor and insulator flashover development process can provide a reference for the pollute flashover pre-warning.Fractal dimension characteristic values of composite insulators leakage current is derived by using fractal theory, the variation features of waveform and fractal dimension of leakage current in different sections of the whole process of flashover development is analyzed, and leakage current waveform fractal characteristics in three different contamination levels is compared. The results show that in the process of flashover, fractal dimension of leakage current waveform descends on the whole; and there is a slight difference in the leakage current waveform fractal characteristics depending on different contamination level. Experimental analysis shows that the leakage current waveform fractal dimension can directly reflect its waveform characteristics and provide effective early warning for insulator flashover.