Reinforcement learning for power systems control

Adaptive topology control in power systems using multi-agent reinforcement learning and spectral clustering.

Collaborators: Erica van der Sar, Sandjai Bhulai, Jan Viebahn (TenneT)

See below for the list of related publications and preprints.

Reinforcement Learning Operations Research Applied Probability Optimization Graph Theory Power systems reliability
Alessandro Zocca
Alessandro Zocca
Tenured Assistant Professor

Related

Publications

Multi-Agent Reinforcement Learning for Power Grid Topology Optimization (2023)

E. van der Sar, A. Zocca, S. Bhulai Submitted to PSCC 2024
How can we use multi-agent reinforcement learning to dynamically adjust the topology of power grids?
arXiv preprint PDF BibTeX

Mixed-integer linear programming approaches for tree partitioning of power networks (2023)

L. Lan, A. Zocca Submitted to Transactions on Power Systems
How to optimally tree-partition power systems as alternative to intentional controlled islanding?
arXiv preprint PDF BibTeX

A Spectral Representation of Power Systems with Applications to Adaptive Grid Partitioning and Cascading Failure Localization (2021)

A. Zocca, L. Chen, L. Guo, S.H. Low, A. Wierman Submitted
How to use spectral graph theory to improve power networks robustness against cascading failures?
arXiv preprint PDF bib