Intelligence Detection of DC Parallel Arc Failure With Featuring From Different Domains

Abstract

DC microgrids are increasingly becoming the backbone of renewable energy integration. Their ability to efficiently manage intermittent sources like solar and wind power is transforming the energy landscape. However, a critical challenge remains in the form of DC arc faults, which can significantly compromise the reliability and safety of these systems. Parallel arc faults represent a particularly challenging scenario due to their unique electrical behavior. Unlike series arc faults, which cause a decrease in system current, parallel arcs can lead to a significant increase in current due to the low resistance path they create. This research delves into the electrical behavior of DC systems during parallel arc faults. By analyzing the source current signals in different domains, the authors aim to identify specific characteristic features of the source current that can serve as reliable indicators combined with artificial learning models for arc fault diagnosis. The findings of this research can have significant implications for the improvement of advanced arc failure recognition systems. This research represents a valuable step towards safe and reliable DC systems by addressing the challenge of parallel arc fault detection.

Keywords

• DC parallel arc fault
• feature characteristics
• artificial learning models