PhD defense alert: Fabio Pavirani
Analysis and Enhancement of Monte Carlo Tree Search Techniques for Energy Applications
You are invited to join the PhD Defense of:
Name: Fabio Pavirani
Title: Analysis and Enhancement of Monte Carlo Tree Search Techniques for Energy Applications
Topic:
The energy transition is transforming power systems into increasingly decentralized, dynamic, and uncertain environments. The large-scale integration of renewable energy sources, the electrification of end uses, and the growing participation of distributed energy resources are creating a need for new control strategies capable of coordinating flexibility across the grid. Demand response has emerged as a key mechanism in this context, enabling consumers and storage assets to actively contribute to balancing supply and demand and supporting the reliable operation of the electricity system.
This dissertation investigates the potential of Monte Carlo Tree Search (MCTS) as a control and decision-making framework for demand response applications. Positioned between traditional optimization methods and data-driven artificial intelligence techniques, MCTS combines explicit planning with the flexibility to operate on complex simulation models. Through applications spanning residential energy management, grid-scale balancing, and stochastic energy optimization, the dissertation explores how this planning paradigm can support decision-making under uncertainty across multiple levels of the power system.
By advancing the methodology and demonstrating its effectiveness in diverse energy-sector settings, this work contributes to the fields of intelligent energy management, power system control, and planning-based artificial intelligence, highlighting the role that simulation-driven planning can play in enabling a more flexible and sustainable electricity grid.
Curriculum Vitae Fabio Pavirani:
Fabio Pavirani obtained both his Bachelor's and Master's degrees in Mathematics from the Università di Modena e Reggio Emilia (Italy), specializing in Data Science during his Master's studies. In 2022, he joined IDLab at Ghent University under the supervision of Prof. Chris Develder and Prof. Bert Claessens to pursue a PhD in Engineering.
His doctoral research focuses on the application of Artificial Intelligence to energy systems, with particular emphasis on sequential decision-making and planning under uncertainty. In this context, he investigated how planning techniques such as Monte Carlo Tree Search (MCTS) and reinforcement learning can be adapted and extended to address challenges in energy management, demand response, electricity markets, battery control, and electric vehicle charging. Through this work, he contributed novel methods for decision support and optimization in sustainable energy systems.
His research has resulted in nine scientific publications, including two journal articles in Energy and Buildings and Applied Energy, four conference papers, and several preprints and manuscripts in preparation, with six publications led as first author. He presented his work through international research collaborations and scientific conferences. His research interests include artificial intelligence, planning algorithms, reinforcement learning, optimization under uncertainty, and sustainable energy systems.
Date defense: 27 October 2026 | 5 PM
Place: IGent, Auditorium 1, Technologiepark 126, 9052 Zwijnaarde
Online: link to follow online
