The synergy of multi-agent systems and machine learning in power system applications

Multi-agent systems are widely adopted for the simulation of complex power systems and for distributed decision-making. How these agents are able to act in an intelligent and adaptive way is, however, dependent on the learning capabilities they have. It is, therefore, essential to explore the synergy between multi-agent systems and the most recent advances of machine learning. This panel addresses some of the most promising solutions that can be achieved through the combination of multi-agent systems and machine learning models. The panel will foster discussion on the advantages of combining multi-agent systems with machine learning and will identify gaps in this kind of synergy, which are expected to contribute to the opening of new research paths towards a more efficient and intelligent power and energy system. Presentations in this panel session: - Distributed Geometric Koopman Operator Learning for Sparse Networks: Application to Power Systems (23PESGM3947)