The necessary integration of renewable energy sources, combined with the expanding scale of power networks, presents significant challenges in controlling modern power grids. Traditional control systems, which are human and optimization-based, struggle to adapt and to scale in such an evolving context, motivating the exploration of more dynamic and distributed control strategies. This work advances a graph-based distributed reinforcement learning framework for real-time, scalable grid management. The proposed architecture consists of a network of distributed low-level agents acting on individual power lines and coordinated by a high-level manager agent. A Graph Neural Network (GNN) is employed to encode the network's topological information within the single low-level agent's observation. To accelerate convergence and enhance learning stability, the framework integrates imitation learning and potential-based reward shaping. In contrast to conventional decentralized approaches that decompose only the action space while relying on global observations, this method also decomposes the observation space. Each low-level agent acts based on a structured and informative local view of the environment constructed through the GNN. Experiments on the Grid2Op simulation environment show the effectiveness of the approach, which consistently outperforms the standard baseline commonly adopted in the field. Additionally, the proposed model proves to be much more computationally efficient than the simulation-based Expert method.
Figures & tables
Figure 1: Abstract high-level representation of the model.
Figure 2: Example of substations embedding.
Figure 3: Example of lines embedding.
Figure 4: Representation of the Grid2Op “l2rpn_case14_sandbox” power grid.
Figure 5: Comparison of the performance of our RL agent against the Do-Nothing baseline over the validation set (top) and the test set (bottom).
Configuration
Validation Perf.
Test Perf.
Do-Nothing (baseline)
997.7
646.6
No DQfD
1985
1878
No GNN
1206.4
785.2
No Reward Shaping
5667.1
5324.3
Complete System
5452.2
6114.4
Table 1: Ablation study results: average survival time on validation and test sets. The best result per column is in bold. The first row (Do-Nothing) reports the performance of the baseline.
Figure 6: Heatmap of the weights in the first GNN layer.
Model
Inference Time (avg ± std) [s]
Proposed Model
0.187 ± 0.145
Expert Agent
2.56 ± 0.223
Table 2: Inference Time Comparison: Proposed approach against Expert agent.
As the integration of volatile renewable energy sources increases the strain on modern power grids, the use of Reinforcement Learning (RL) for autonomous topological reconfiguration has emerged as a promising research field to keep strained grids stable and operational. Compared to traditional redispatching measures, topological actions offer a cheaper and more cost-effective way to manage grid congestion. However, their implementation is hindered by a vast combinatorial action space and strict operational constraints. This paper investigates the effectiveness of model-based AlphaZero-inspired approaches that utilize Monte Carlo Tree Search (MCTS) for proactive grid management. We systematically evaluate how reward functions, observation density, and search guidance influence an agent's survivability. Our results demonstrate that the optimized AlphaZero approach achieves a peak survivability of 98.43%, significantly outperforming the proximal policy optimization (PPO) variant. We find that conducting the MCTS without guidance from a prior learned policy or value function can enhance training efficiency, and that a straightforward binary survival reward provides more effective search guidance than complex, multi-objective functions. Our findings demonstrate that while AlphaZero is a powerful framework for topological control, pure reinforcement learning is not sufficient; rather, an effective and reliable system requires a 'minimalist' integration of domain-specific heuristics, binary rewards, and a restricted observation space of line loads.
Lukas Zetto, Benjamin Schäfer, Qiong Huang
Karlsruhe Institute of Technology, Karlsruhe, Germany
Self-healing smart grids can quickly adjust their network configuration during outages to minimize power disruptions. During an outage, several actions can be taken, such as network reconfiguration through switching operations and emergency load shedding. However, traditional machine learning methods for outage mitigation are not well suited for smart grids due to their slow response time and high computational cost. To address these challenges, recent studies have explored reinforcement learning to automatically perform network reconfiguration. In these approaches, the control policy is typically modeled using a graph neural network (GNN). However, conventional GNNs operate in the spatial domain and may fail to capture important relationships in the frequency domain. Frequency-domain information is particularly useful for modeling global structural patterns and system-wide interactions in power networks. In this paper, we propose a spectral graph reinforcement learning framework for outage management in distribution networks to enhance system resilience. Our model learns the optimal power restoration policy using a spectral graph neural network. We evaluate the proposed method on three modified IEEE test systems: the 13-bus, 34-bus, and 123-bus networks. Experimental results show that our approach achieves near-optimal performance in real time and generalizes well across a wide range of outage scenarios.
Reinforcement learning has shown promise for automating power-grid operation tasks such as topology control and congestion management. However, its deployment in real-world power systems remains limited by strict safety requirements, brittleness under rare disturbances, and poor generalization to unseen grid topologies. In safety-critical infrastructure, catastrophic failures cannot be tolerated, and learning-based controllers must operate within hard physical constraints. This paper proposes a safety-constrained hierarchical control framework for power-grid operation that explicitly decouples long-horizon decision-making from real-time feasibility enforcement. A high-level reinforcement learning policy proposes abstract control actions, while a deterministic runtime safety shield filters unsafe actions using fast forward simulation. Safety is enforced as a runtime invariant, independent of policy quality or training distribution. The proposed framework is evaluated on the Grid2Op benchmark suite under nominal conditions, forced line-outage stress tests, and zero-shot deployment on the ICAPS 2021 large-scale transmission grid without retraining. Results show that flat reinforcement learning policies are brittle under stress, while safety-only methods are overly conservative. In contrast, the proposed hierarchical and safety-aware approach achieves longer episode survival, lower peak line loading, and robust zero-shot generalization to unseen grids. These results indicate that safety and generalization in power-grid control are best achieved through architectural design rather than increasingly complex reward engineering, providing a practical path toward deployable learning-based controllers for real-world energy systems.
Gitesh Malik
Department of Electrical Engineering, Delhi Technological University (DTU), India