ACDSA

Prof. Dr. GIANCARLO STORTI GAJANI

Politecnico di Milano, Italy

He received the Dr. Ing degree in electronic engineering and the Ph. D. degree in electronic systems from Politecnico di Milano in 1986, and 1991 respectively. Since 1992 he is Assistant Professor and since 2002 Associate Professor of Circuit Theory in the same University. His research interests have initially focused on the development of architectures for signal processing (in particular, for audio and music applications) and neural networks. More recently, his research interests include non linear circuits and machine learning methods applied to power engineering. He is a senior member of IEEE and the author of more than 90 papers in international Journals and Conferences.

Title: Artificial Intelligence in Power Systems and Electric Mobility

Abstract: The changing nature of power systems alongside the increasing presence of electric vehicles, which presents both challenges and opportunities, makes the availability of tools capable of prediction of the utmost importance.
AI and ML applications improve power systems' ability to predict outcomes, allowing for better electricity demand, price, and generation forecasts, which is needed to balance supply with demand and avoid blackouts. While vital for lowering emissions, the incorporation of renewable energy creates stability issues that ML can solve through better forecasting and optimized dispatch and storage. The expansion of EV charging infrastructure impacts grid stability, potentially causing voltage fluctuations and overloading, which calls for intelligent management systems.
The aggregator model presented, which uses RL to optimize charging and reduce grid overloads, is one such system. When Optimal Power Flow (OPF) is addressed with AI and RL, it allows for the lowest operational costs while maintaining system security, which is critical for efficient grid operation. Microgrid control in Renewable Energy Communities (RECs) benefits from Model Predictive Control (MPC) models that are enhanced with LSTM-based predictors. This optimizes resource management and promotes renewable energy use.
Finally, AI's adaptability enables dynamic stability control and adjustments to changing grid conditions, assuring better energy management and dispatch.