The Application of Wavelet Neural Network in the Detection of Electrical Insulator Leakage Current

In order to master the extent of the running insulator contamination and analyze the influence of the environmental factors on the external electrical characteristics of different insulators, in this paper, by selecting leakage current peak, virtual value of leakage current, leakage current pulses and environmental temperature and humidity and something related as the parameters, introduce a kind of insulator leakage current detection algorithm which is based on Wavelet Neural Network. With wavelet transform decomposing the noise signal, use the decomposed signal as the input of the neural network, and learn and judge it by the aid of network. In these foundations, the results from the designed insulator on-line detection system simulation which is on the basis of DSP show that the application of this algorithm can enhance the forecasting performance to the extent of insulator contamination.