Fractal Dimension Logarithmic Differences Method for Low Voltage Series Arc Fault Detection

Series arc faults introduce singularities in the current signal and changes over time. Fractal dimension can be used to characterize the dynamic behaviour of the current signal by providing a degree of signal chaos. This measure of irregularity exhibits changes in signal behaviour that can suitably be used as a basis for series arc fault detection. In this paper, an efficient low voltage series arc fault detection method based on the logarithmic differences of the estimate of the fractal dimension of the current signal using the multiresolution length-based method is presented. The discrete wavelet transform and the hard thresholding denoising with the universal threshold are also used. Experimental results in both normal and arcing condition with different loads in compliance with the standard IEC 62606 show the effectiveness of the method that allows reliable series arc fault detection within 80ms.