RL2Grid: Benchmarking Reinforcement Learning in Power Grid Operations
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908414388469760 |
|---|---|
| author | Marchesini, Enrico Donnot, Benjamin Crozier, Constance Dytham, Ian Merz, Christian Schewe, Lars Westerbeck, Nico Wu, Cathy Marot, Antoine Donti, Priya L. |
| author_facet | Marchesini, Enrico Donnot, Benjamin Crozier, Constance Dytham, Ian Merz, Christian Schewe, Lars Westerbeck, Nico Wu, Cathy Marot, Antoine Donti, Priya L. |
| contents | Reinforcement learning (RL) can provide adaptive and scalable controllers essential for power grid decarbonization. However, RL methods struggle with power grids' complex dynamics, long-horizon goals, and hard physical constraints. For these reasons, we present RL2Grid, a benchmark designed in collaboration with power system operators to accelerate progress in grid control and foster RL maturity. Built on RTE France's power simulation framework, RL2Grid standardizes tasks, state and action spaces, and reward structures for a systematic evaluation and comparison of RL algorithms. Moreover, we integrate operational heuristics and design safety constraints based on human expertise to ensure alignment with physical requirements. By establishing reference performance metrics for classic RL baselines on RL2Grid's tasks, we highlight the need for novel methods capable of handling real systems and discuss future directions for RL-based grid control. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_23101 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | RL2Grid: Benchmarking Reinforcement Learning in Power Grid Operations Marchesini, Enrico Donnot, Benjamin Crozier, Constance Dytham, Ian Merz, Christian Schewe, Lars Westerbeck, Nico Wu, Cathy Marot, Antoine Donti, Priya L. Machine Learning Artificial Intelligence Reinforcement learning (RL) can provide adaptive and scalable controllers essential for power grid decarbonization. However, RL methods struggle with power grids' complex dynamics, long-horizon goals, and hard physical constraints. For these reasons, we present RL2Grid, a benchmark designed in collaboration with power system operators to accelerate progress in grid control and foster RL maturity. Built on RTE France's power simulation framework, RL2Grid standardizes tasks, state and action spaces, and reward structures for a systematic evaluation and comparison of RL algorithms. Moreover, we integrate operational heuristics and design safety constraints based on human expertise to ensure alignment with physical requirements. By establishing reference performance metrics for classic RL baselines on RL2Grid's tasks, we highlight the need for novel methods capable of handling real systems and discuss future directions for RL-based grid control. |
| title | RL2Grid: Benchmarking Reinforcement Learning in Power Grid Operations |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2503.23101 |