A new DC arc fault detection method using DC system component modeling and analysis in low frequency range

In this paper, system component and arc modeling are presented, along with low frequency range analysis for dc arc fault detection. With an arc fault model consisting of a voltage source and variable resistor, the system during arcing can be described as an equivalent circuit model, which gives motivation for the proposed arc detection method. The presented model represents the low frequency characteristics of the main system components and arc in a dc system. Their influence on the line current can be well described and, as a result, it can be utilized for arc detection. The proposed method can distinguish normal operation from arcing and, as a result, dramatically reduces the possibility of false-positives. Experimental results verify the model validity and performance of the method.