Detection method and hardware deployment for series arc faults in residential power circuits

Congxin Han; Zhiyong Wang; Fengyi Guo; Min Wang

Detection method and hardware deployment for series arc faults in residential power circuits

Vol. 25, No. 12, pp. 2338-2350, Dec. 2025

10.1007/s43236-025-01054-w

Series arc fault

Improved empirical wavelet transform

Fault feature

Hiking optimization algorithm

Support Vector Machine

Abstract

The diversity of series arc fault characteristics caused by different load types makes the detection of residential series arc faults complex and challenging. To address this issue, this paper proposes a series arc fault detection method and a hardware deployment approach suitable for residential power circuits and various load conditions. Initially, the energy operator is introduced to optimize frequency spectrum partitioning, which enhances the adaptability and accuracy of the improved empirical wavelet transform (IEWT) in decomposing current signals. Subsequently, empirical mode functions (EMFs) rich in fault features are selected using the Spearman correlation coefficient and energy entropy (EE). The EE, slope entropy (SE), and dispersion entropy (DE) are used as fault features. Finally, an arc fault recognition model is established using a hiking optimization algorithm optimized support vector machine (HOA-SVM). Compared to other methods, the proposed method demonstrates a faster computation speed and a higher detection precision. Additionally, the practicality of the proposed method is assessed through implementation on a Raspberry Pi 5, confirming its feasibility for practical application. Detection results indicate that the recognition accuracy is over 97%, and the recognition time for a single sample is lower than 25.01 ms. These contributions highlight the potential of the model as a reliable and efficient solution for arc fault detection.

Statistics

Show / Hide Statistics

Cumulative Counts from September 30th, 2019
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.

Cite this article

[IEEE Style]

C. Han, Z. Wang, F. Guo, M. Wang, "Detection method and hardware deployment for series arc faults in residential power circuits," Journal of Power Electronics, vol. 25, no. 12, pp. 2338-2350, 2025. DOI: 10.1007/s43236-025-01054-w.

[ACM Style]

Congxin Han, Zhiyong Wang, Fengyi Guo, and Min Wang. 2025. Detection method and hardware deployment for series arc faults in residential power circuits. Journal of Power Electronics, 25, 12, (2025), 2338-2350. DOI: 10.1007/s43236-025-01054-w.