A Wavelet Approach Applied to Transformer Fault Protection: Signal Processing

The use of multiresolution analysis and wavelets is explored in this paper as a means of discrimination between inrush and faults in power transformers. The discrimination is based on the analysis of the high frequency components of the differential signal after using Clarke transformation. High values of the high frequency spectrum have been found to be present at least once per fundamental cycle if the transformer presents an internal fault. Finite element analysis as well as laboratory test on a 15 kVA transformers are presented to discuss the technique