Mathematical Fundamentals of Machine Learning in Power Systems

Dear Colleagues,

Machine learning is increasingly employed in power systems to address complex challenges related to optimization, control, forecasting, and decision-making. However, the reliable deployment of data-driven methods in electrical power systems requires a solid understanding of their underlying mathematical principles, particularly in relation to system constraints, physical consistency, and operational robustness.

This Topic focuses on the mathematical fundamentals of machine learning in power systems, emphasizing models and algorithms grounded in optimization theory, statistical learning, and numerical analysis. Contributions are encouraged that investigate the mathematical structure, stability, convergence, and interpretability of machine learning techniques applied to power generation, transmission, and distribution systems.

Relevant applications include smart grids, renewable energy integration, electric vehicle coordination, voltage and loss optimization, demand response, and energy management systems. Particular attention is given to hybrid approaches that combine physical modeling with data-driven learning, as well as to explainable and uncertainty-aware machine learning methods. By bringing together researchers from applied mathematics, machine learning, and power engineering, this Topic aims to advance trustworthy, transparent, and mathematically sound machine learning solutions for modern power systems.

Dr. Bogdan Neagu
Dr. Andrea Mazza
Topic Editors