# Physics Informed Reinforcement Learning with Gibbs Priors for Topology Control in Power Grids

arXiv: https://arxiv.org/abs/2604.01830 · HTML: https://mlsanity.com/paper/2604.01830/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in RL for energy, networks and operations, one of 15 directions in Deep reinforcement learning methods.
- Builds on 14 of its 20 references that are in the map.
- First posted on arXiv 2 Apr 2026.

## Placed in

- Deep reinforcement learning methods / [RL for energy, networks and operations](https://mlsanity.com/deeprl/rl-for-energy-networks-and-operations/)

## Builds on (most cited first, 14 of 14)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Learning to run a Power Network Challenge: a Retrospective Analysis](https://mlsanity.com/paper/2103.03104/) (2021)
- [Action Set Based Policy Optimization for Safe Power Grid Management](https://mlsanity.com/paper/2106.15200/) (2021)
- [Reinforcement Learning for Electricity Network Operation](https://mlsanity.com/paper/2003.07339/) (2020)
- [Power Grid Congestion Management via Topology Optimization with AlphaZero](https://mlsanity.com/paper/2211.05612/) (2022)
- [PowRL: A Reinforcement Learning Framework for Robust Management of Power Networks](https://mlsanity.com/paper/2212.02397/) (2022)
- [Managing power grids through topology actions: A comparative study between advanced rule-based and reinforcement learning agents](https://mlsanity.com/paper/2304.00765/) (2023)
- [Exploring grid topology reconfiguration using a simple deep reinforcement learning approach](https://mlsanity.com/paper/2011.13465/) (2020)
- [Hierarchical Reinforcement Learning for Power Network Topology Control](https://mlsanity.com/paper/2311.02129/) (2023)
- [HUGO -- Highlighting Unseen Grid Options: Combining Deep Reinforcement Learning with a Heuristic Target Topology Approach](https://mlsanity.com/paper/2405.00629/) (2024)
- [Curriculum Based Reinforcement Learning of Grid Topology Controllers to Prevent Thermal Cascading](https://mlsanity.com/paper/2112.09996/) (2021)
- [Optimizing Power Grid Topologies with Reinforcement Learning: A Survey of Methods and Challenges](https://mlsanity.com/paper/2504.08210/) (2025)
- [KCLNet: Physics-Informed Power Flow Prediction via Constraints Projections](https://mlsanity.com/paper/2506.12902/) (2025)
- [Learning Topology Actions for Power Grid Control: A Graph-Based Soft-Label Imitation Learning Approach](https://mlsanity.com/paper/2503.15190/) (2025)

## Built on by (most cited first, 1 of 1)

- [Reinforcement Learning versus Optimization for Optimal Transmission Switching: A Comparative Study](https://mlsanity.com/paper/2607.10948/) (2026)

## Abstract (arXiv)

Topology control for power grid operation is a challenging sequential decision making problem because the action space grows combinatorially with the size of the grid and action evaluation through simulation is computationally expensive. We propose a physics-informed Reinforcement Learning framework that combines semi-Markov control with a Gibbs prior, that encodes the system's physics, over the action space. The decision is only taken when the grid enters a hazardous regime, while a graph neural network surrogate predicts the post action overload risk of feasible topology actions. These predictions are used to construct a physics-informed Gibbs prior that both selects a small state-dependent candidate set and reweights policy logits before action selection. In this way, our method reduces exploration difficulty and online simulation cost while preserving the flexibility of a learned policy. We evaluate the approach in three realistic benchmark environments of increasing difficulty. Across all settings, the proposed method achieves a strong balance between control quality and computational efficiency: it matches oracle-level performance while being approximately 6× faster on the first benchmark, reaches 94.6% of oracle reward with roughly 200× lower decision time on the second one, and on the most challenging benchmark improves over a PPO baseline by up to 255% in reward and 284% in survived steps while remaining about 2.5× faster than a strong specialized engineering baseline. These results show that our method provides an effective mechanism for topology control in power grids.