A Review on Arc Fault Diagnosis Methods for DC Distribution Networks

Photovoltaic power generation are becoming an important trend in the composition of power grid systems, and are playing an increasing role in DC distribution networks due to their environmentally friendly, easy-to-install, and easy-to-maintain characteristics. However, the difficulties in the diagnosis of DC arc faults are beginning to emerge. Compared with the reciprocating zero crossing point of the AC transmission grid, the DC circuit current has no natural zero crossing point, and when a DC arc fault occurs, it cannot be detected using traditional AC arc detection methods. For this reason, a great deal of work has been done over the years by workers in the electrical field, focusing on analyzing the characteristics of arc faults in DC power systems and their effects on the system. This technology is essential to ensure the safe operation of DC power systems, prevent equipment damage, and reduce unplanned outages. However, although there are many studies around the detection of DC arcs, the current research on real-time diagnosis and early warning systems for the development of DC arc dynamics is still insufficient. This paper introduces the DC arc fault potential hazards in DC circuits through the policy initiatives and applications of new energy technologies in various countries, and in the main part of the article, it introduces the diagnostic methods based on electromagnetic radiation, arc sound, time-domain features, and frequency-domain features, as well as the arc diagnostic techniques through artificial intelligence techniques that have emerged in recent years. At the same time, the article also describes the advantages and shortcomings of the various technical routes and predicts the possibility of combining DC arc diagnostic technology with industrial Internet of Things and other technologies in the future.