Applications of Machine Learning in Modern Power Systems: A Comprehensive Review

The demand for electrical power has increased due to population growth, electronic technology advancements, and environmental concerns. This has led to a significant modernization of the power system that incorporates renewable energy sources and electronic components for measurement, communication, and control. However, these changes have added new challenges to the management and control of modern power systems that cannot be overcome using traditional analytical methods. Artificial Intelligence (AI), specifically Machine Learning (ML) techniques, have recently been deployed by a wide number of researchers from different fields thanks to their adaptability and learning ability at a higher speed. These techniques are adequate for large, non-linear, and multi-variable problems such as modern power systems. Therefore, this paper aims to provide an extensive review of recent ML techniques as well as their usage in modern power systems in terms of power quality, power stability, energy and load forecasting, protection and fault diagnosis, and cybersecurity. Finally, the main challenges associated with the implementation of ML techniques for future modern power systems are pointed out.