Artificial Intelligence in Electrical Power Systems: A Comprehensive Review of Machine Learning Techniques and Applications

Abstract

The electrical power system is undergoing a significant transformation through the use of Artificial Intelligence (AI) and Machine Learning (ML) for advanced forecasting, monitoring, optimization, fault detection and intelligent control of electrical power systems. It reviews the broadly used ML methods such as regression, decision trees, support vector machines, neural networks, deep learning, ensemble methods, and reinforcement learning, and their implementations in the field of load forecasting, renewable energy integration, stability assessment, predictive maintenance, and smart grids. The review also addresses issues of data quality, computational complexity, cybersecurity, interpretability and model generalization. The future directions of intelligent power systems are highlighted to be promising with emerging approaches such as explainable AI, physics-informed learning, federated learning, digital twins, edge AI and autonomous operation.