Detection of Arc Fault Based on Hilbert Huang Transform Under Voltage Variation Disturbances and Harmonics

Electrical arc faults are one of the main causes of instability and equipment failure in power systems, especially when occurring simultaneously with voltage disturbances such as sag, swell, and harmonic distortion. These faults, which are classified into series and parallel arcs, are difficult to detect using conventional signal processing methods such as Fast Fourier Transform (FFT), Discrete Wavelet Transform (DWT), or Short-Time Fourier Transform (STFT), due to their limitations in analyzing non-linear and non-stationary signals. Numerous studies have demonstrated that the Hilbert-Huang Transform (HHT) is superior in analyzing complex signals and has been widely applied in fields such as mechanical condition monitoring, geophysics, and power system diagnostics. In this study, HHT is applied to detect and analyze arc faults under various Voltage Variation Disturbance (VVD) conditions such as sag, swell, and harmonics. The signal is decomposed into Intrinsic Mode Functions (IMFs) using Empirical Mode Decomposition (EMD), followed by Hilbert Spectrum analysis to extract time-frequency characteristics. Simulation results show that HHT successfully identifies arc fault signatures in the early IMFs and distinguishes voltage disturbances in the subsequent IMFs. This success demonstrates HHT's effectiveness in distinguishing between normal and faulted conditions, as well as accurately classifying types of disturbances. Moving forward, the integration of HHT with machine learning has great potential for enhancing real-time fault detection and automation in power systems.