Presentation in PriTEM workshop, March 22-23, 2023, by Mehdi Foroughi from IFI, UiO
Abstract: The emergence of networked microgrids has transformed energy systems into multi-agent systems, enabling microgrids to proactively determine policies and make intelligent decisions based on their objectives. However, the centralized energy market presents challenges for direct peer-to-peer trading between microgrids, potentially restricting their ability to act proactively. To address this, we propose a decentralized architecture that employs distributed deep reinforcement learning to model peer-to-peer energy trading among microgrids as a partially observable Markov game. The approach integrates peer-to-peer trading with the physical network through grid sensitivity analysis, allowing microgrids to navigate the complex and competitive environment of peer-to-peer trading. Results from simulations using the CIGRE distribution network demonstrate the effectiveness of our proposed approach in facilitating market clearing for peer-to-peer energy trading.
Bio of Mehdi Foroughi: Mehdi is a passionate researcher who has a strong interest in energy and environmental issues. He holds a Bachelor's degree in Aerospace and a Master's degree in Energy Systems. Mehdi gained valuable experience as a consultant for start-ups in the oil sector, and is now pursuing a PhD in energy informatics at the University of Oslo. His research focuses on developing autonomous and distributed management solutions for energy market participants, with the aim of addressing challenges through innovative approaches.