Combined Fourier Transform and Mexican Hat Wavelet for Fault Detection in Distribution Networks

This paper presents a combined application of fast Fourier transform (FFT) and the Mexican hat wavelet for fault detection within distribution networks. The short time fast Fourier transform (STFFT) can replace FFT as well. One of the two Fourier transforms is used to determine the high frequency harmonics due to abrupt changes in line currents, those harmonics take place when various faults occur. The Mexican hat wavelet is used with a scaling factor which is found based on the frequency of the harmonics and the sampling rate of the measurement device. The results show good performance of the suggested approach, and the next step would be the classification of the fault type.