Cross-entropy based AC series arc fault detection for more electric aircrafts

The increasing adoption of More Electric Aircraft (MEA) has introduced new challenges in ensuring the reliability and safety of onboard electrical systems. Among these challenges, AC series arc faults pose a significant risk due to their potential to degrade system integrity and cause fire hazards. Detecting such faults is particularly challenging because of their intermittent nature and similarity to normal load switching events. This article proposes a novel AC series arc fault detection technique based on time-domain current waveform analysis. The technique quantifies the asymmetry introduced by arc faults using cross-entropy and extracts the fault-imposed component to derive a fault detection index. The energy of this component is monitored over time to distinguish persistent arc faults from transient disturbances. The effectiveness of the proposed technique is evaluated through extensive MATLAB/Simulink simulations. Various case studies, including load switching events, nonlinear loads, and system parameter variations, are analyzed to assess the technique's robustness. Additionally, a sensitivity analysis is conducted to investigate the impact of key parameters on detection performance. The results confirm that the proposed technique achieves high sensitivity and reliability in detecting series arc faults while effectively discriminating against non-fault disturbances, making it a promising solution for enhancing the safety and protection of next-generation aircraft electrical systems.