Smart detection technology of serial arc fault on low-voltage indoor power lines

Abstract

Abstract This paper employs the discrete wavelet transform (DWT) and an artificial neural network to identify the occurrence of serial arc faults on indoor low voltage power lines. Electric arc faults on power lines must be detected in order to turn off the electric power sources before fire events occur. However, since the characteristics of line current waveforms during serial arc faults are complicated, smart detection technology is required to have high accurate recognition. The DWT is utilized to obtain the time-domain characteristics of line current waveforms, and the signal energy of some sub-bands is useful information to reflect the serial arc fault patterns. And then, a radial basis function neural network (RBFNN) is trained by using the data of signal energy obtained from DWT. After the training process, the RBFNN has excellent ability to identify the serial arc-fault conditions. At last, the accumulative RBFNN outputs of 30 power cycle line current data are used to certify the occurring of a serial arc fault on the line. This study also compares the results of detecting serial arc faults with a commercial arc-fault circuit interrupter (AFCI) to reveal the goodness of the purposed method.