Identifying DC Series and Parallel Arcs Based on Deep Learning Algorithms

Abstract

Arc phenomena are usually related to the undesired disengagement of two electrical connections. The emission power discharge from the failure arc may damage wiring and can present a fire hazard. Numerous studies have been proposed to detect arc events and quickly isolate them from an electrical system. DC arc faults are often sorted into two types: series and parallel arcs. A series arc may be the outcome of discharging links in electrical wiring. By contrast, the parallel arc occurs between two electric wires, or between a link and a ground, owing to contamination or poor isolation. The currents in a system with an arc fault are considerably greater when the arc parallel in nature than when the arc is series in nature. In this paper, the electric activities of a network are investigated for the duration of series and parallel arc failures in both the time and frequency domains. The arcing behavior investigated is selected to allow for the identification of series and parallel arcs. The sorting of electrical arcs in an accurate and reliable manner is useful for electrical protection schemes. The identification process used here is based on data related to different domains, such as load current and voltage. In this study, eight learning techniques are investigated with the aim of detecting series and parallel arc faults. The arc behaviors were studied in the various domains. We used the load current and voltage characteristics as an statistic for categorizing a given arc failure. This study could be beneficial to enhance the stability and reliability of arc-fault detectors.

Various Feature-Based Series Direct Current Arc Fault Detection Methods Using Intelligence Learning Models and Diverse Domain Exclusion Techniques

“…In this case, the decision tree type used is the classification and regression tree (CART), suitable for both classification and regression tasks. The depth of the tree, which determines the maximum number of splits, is set to 4 with 14 leaf nodes [26]. Ensemble learning, a powerful technique in machine learning, combines predictions from multiple models to enhance accuracy and reliability.…”

“…Machine learning algorithms have shown promise in DC arc fault detection. Nevertheless, current methodologies frequently concentrate exclusively on time or frequency domain currents, overlooking the necessity for inclusive preprocessing of signals [18][19][20][21][22][23][24][25][26][27][28], although an approach with simple indexes has been tried for DC arc detection [19]. This research presents a novel methodology to detect arc fault recognition by extracting and utilizing various key features for DC arc detection.…”

Empirical Filtering-Based Artificial Intelligence Learning Diagnosis of Series DC Arc Faults in Time Domains

“…However, it is worth noting that FFT analysis has limitations, including the requirement for more time to transform the time-domain signals into frequency-domain signals and an inability to analyze the time-domain characteristics effectively. Modern research endeavors have increasingly turned to artificial-intelligence-driven methods for fault detection and diagnosis, a trend that has proven highly beneficial in the context of arc fault diagnosis [17,18].…”

DC Series Arc Failure Diagnosis Using Artificial Machine Learning With Switching Frequency Component Elimination Technique

“…Machine learning-driven approaches have been increasingly harnessed in contemporary research endeavors to ascertain and diagnose faults. These methodologies have exhibited their utility in arc fault diagnosis [15], [16], [17], [18]. The scholarly community has adeptly employed these sophisticated techniques within the DC arc fault diagnosis context, leading to notable and affirmative outcomes.…”