Fault detection is crucial for ensuring the reliability and safety of electrical systems. This study explores the capabilities of artificial intelligence (AI) in fault detection and compares them with conventional techniques. AI-based approaches leverage advanced machine learning algorithms and deep learning models to accurately identify faults and anomalies in electrical systems. Our analysis reveals that AI-based fault detection approaches demonstrate superior performance compared to conventional techniques across various evaluation metrics. However, challenges such as dataset bias, interpretability, and scalability pose significant hurdles. Recommendations are proposed to address these challenges, including efforts to mitigate dataset bias, enhance model interpretability, optimize scalability, promote collaboration, and encourage continued research and development. By implementing these recommendations, stakeholders can develop more robust and reliable fault detection solutions that contribute to the safety, reliability, and sustainability of electrical infrastructure.
Jun-Jun A. Amparado (Mon,) studied this question.