Analysis and Classification of Arcing Signals by Using MFCC

Abstract

Arc faults in electrical systems pose significant safety risks, and their early detection is crucial for preventing fires and other hazards. Traditional methods for arc fault detection in power systems often rely on conventional signal processing techniques, which may lack robustness and accuracy, especially in noisy environments. In this study, we propose a novel approach for arc fault detection using Mel-frequency cepstral coefficients (MFCCs) extracted from current signals generated by both arc and non-arc faults. MFCCs have been widely used in speech and audio processing due to their ability to capture relevant spectral features. In this paper we aim to investigate how MFCCs can differentiate between arc and non-arc faults in electrical systems. By analyzing the MFCC features extracted from current waveforms during both fault and non-fault conditions, to identify unique patterns and characteristics associated with arc faults.