Smart Detection Of AC Series Arc Fault On Home Voltage Line Based On Fast Fourier Transform And Artificial Neural Network

Abstract

A fast fourier transform (FFT) and artificial neural network (ANN) were used and explained on this paper, for detect series arc fault on home voltage line. Detection of series arc fault is very needed to avoid fires caused by the series arc fault. But, the characteristic of current waveform and current harmonic spectrum during series arc fault are unique and complicated, so the smart detection was required to have the best recognition. To obtain the current spectrum characteristic of line current waveform, the FFT was very required. Then, a feed forward back propagation neural network(FFBPNN) is learned using some part of the data from current harmonics amplitude from FFT. After the learning process,the FFBPNN has a very intellegent ability to detect the serial arc fault condition. At last, the output value of ANN method will decide the occurence series arc fault or not with 12 variation load. Also in this study, compares the results of detection series arc fault using FFT with detection series arc fault using FFT and ANN.