cs.LGAug 17, 2026

Graph Machine Learning: An Opportunity for Power Systems

Authors: Martin Sadric, Sebastian Pütz, Christian Nauck, Veit Hagenmeyer, Frank Hellmann, Dirk Witthaut, Benjamin Schäfer

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Modern power systems face growing operational complexity driven by the integration of renewable energy sources, decentralization, and the need for real-time decision-making across a wide range of timescales. Addressing these challenges traditionally relies on model-based methods that, while accurate, can be too slow for operational demands. Machine learning (ML) has therefore emerged as a faster, data-driven alternative. As grid topology plays a central role in power system operation, graph machine learning (GML) methods offer a natural framework for incorporating topological dependencies as an inductive bias. We survey nearly 800 papers at the intersection of GML and power systems, covering forecasting, state estimation, optimization, control, fault diagnosis, and cybersecurity. Power systems constitute an unusually rich benchmark setting for GML, as they combine hard physical constraints, multi-scale dynamics, safety-critical requirements, and scarce labeled data within a single, well-defined domain. Conversely, power systems can benefit from utilizing GML to complement classical solvers, as GML provide scalable, topology-aware approximations with promising generalization and computational efficiency. We identify open challenges, including limited real-world deployment and the need for interpretable models in safety-critical settings. Despite the rapidly growing number of publications, standardized benchmarks and open datasets remain scarce, leaving many results difficult to reproduce and undermining the long-term scientific credibility of the field. We further derive a structured requirements catalog for ML-ready power grid benchmarks, intended to guide future dataset development and improve reproducibility across studies. We call on the community to prioritize dedicated benchmark studies and the release of open datasets and models.

Figures & tables

Figure 1: GNN workflow for power grid tasks. The model takes as input the adjacency matrix A∈Rn×n\mathbf{A}\in\mathbb{R}^{n\times n} encoding grid topology and node feature matrix X∈Rn×nf\mathbf{X}\in\mathbb{R}^{n\times n_{f}} (e.g., voltage magnitude, angle, capacity). Through graph representation learning , the GNN employs message passing to iteratively aggregate neighborhood information, mapping the raw graph-structured input into node embeddings H∈Rn×d\mathbf{H}\in\mathbb{R}^{n\times d} . These are then used for predictions at the node level ( Y^∈Rn\hat{\mathbf{Y}}\in\mathbb{R}^{n} , e.g., risk of node failure) or at the graph level ( Y^∈R\hat{Y}\in\mathbb{R} , e.g., gird transient stability assessment).
Figure 2: Illustration of transmission, subtransmission, and distribution networks: differences in voltage levels, number of nodes, and graph topology (dense/meshed vs. radial/tree).
Figure 3: Annual GML publications in power systems (2018–2025), broken down by publication type.
Figure 4: Cumulative publication counts per application category (left) and each category’s annual share of total publications (right), revealing which areas have grown fastest since 2021.
Figure 5: Distribution of 792 surveyed publications across GML application areas in power systems. The five largest categories are reviewed as dedicated subsections in Section 4 . Data Management is discussed in Section 5 , Emerging Apps in Section 6 , and Foundational & Surveys in Section 2 .
DatasetApplicable Tasks
IEEE Standard Bus SystemsPower flow analysis, Optimal Power Flow (OPF), fault location estimation, cascading failure simulation, state estimation, transient prediction, voltage fluctuation prediction
PowerGraph [ varbellaPowerGraphPowerGrid2024a ]Power flow, OPF, cascading failure prediction, voltage stability monitoring
SafePowerGraph [ ghamiziSafePowerGraphSafetyawareEvaluation2024 ]Safety-critical power system operations, topology-aware forecasting, N-1 secure OPF, fault localization
Simbench [ meineckeSimBenchBenchmarkDataset2020 ]Comparative benchmarking of power flow and OPF solutions in realistic distribution and transmission grids
NREL PV Datasets [ draxlOverviewMeteorologicalValidation2015 ]Multi-site photovoltaic (PV) power forecasting, handling missing data, renewable integration studies
ACTIVSg / TAMU Test CasesLarge-scale power flow, OPF, and topology-aware tasks; synthetic grids (200–70,000 buses) statistically similar to real U.S. networks, with geographic coordinates and realistic load/generation profiles
Table 1: Publicly available benchmark datasets for GML in power systems. Abbreviations: NREL = National Renewable Energy Laboratory; TAMU = Texas A&M University; EPA = U.S. Environmental Protection Agency; EIA = U.S. Energy Information Administration; L2RPN = Learning to Run a Power Network.
Figure 6: Publication trends for reinforcement learning (RL), physics-informed neural networks (PINN), and explainable AI (XAI) with GML for power systems.
More in this topic

Power Systems

Power systems research studies the planning, operation, control, and protection of electrical grids and their integration with energy resources.

See all 231 papers

Explore similar work

CardsList
Oct 3, 2026

ML-OPF-Bench: Benchmarking Machine Learning for Optimal Power Flow

Machine Learning (ML) methods promise a fast solution process for Optimal Power Flow (OPF). While inconsistent test cases, implementations, and evaluation metrics across existing studies make it challenging to determine which algorithmic advances are most critical for real-world deployment. To this end, we propose ML-OPF-Bench, a unified benchmark for AC- and DC-OPF that evaluates representative ML algorithms under a consistent pipeline, stress-tests them across system sizes, distribution shifts, and resource budgets, and ranks them with a multi-objective framework. We find that prediction accuracy alone is not a reliable indicator of operational feasibility. Under heavily loaded, congested conditions, even the strongest in-distribution performers lose their advantage, while feasibility is maintained largely by post-processing that enforces the target constraints rather than by the underlying pure ML predictor. Data scaling shows that prediction accuracy and constraint violations follow different trajectories, whereas compute scaling shows that returns diminish and that larger models do not consistently perform better. These results expose critical trade-offs among ML methods' speed, accuracy, and feasibility, and offer practical guidance for future ML-OPF design. We open-source the benchmark as an extensible Python package for integrating new learning-based OPF algorithms and evaluating them under the same standard as the existing baselines.

Xinyi Liu, Xuan He, Danny H. K. Tsang +1

Dartmouth College · Hong Kong University of Science and Technology (Guangzhou) · University of Alberta

Optimal Power FlowLearning to OptimizeConstrained Optimization
AskPDF
Sep 28, 2026

PowerZooJax: A JAX-based Power System Benchmark for Reinforcement Learning

Power system operation is a safety-critical sequential decision-making problem, making it a natural testbed for reinforcement learning (RL). However, existing RL environments for power systems are often narrow in scope and computationally limited by CPU-based simulation workflows, making large-scale evaluation difficult. We introduce PowerZooJax, a JAX-based benchmark suite for RL in power system operation. It provides five constrained Markov decision process tasks spanning generation, transmission, distribution, distributed energy resources, and data center microgrid. By rewriting power flow, economic dispatch, market clearing, and device dynamics as JAX computation graphs, PowerZooJax keeps the entire training and evaluation loop on the GPU. Experiments show substantial speedups over CPU-based simulations and demonstrate standardized evaluation of policy returns, safety violations, and out-of-distribution stress conditions. Our open-source benchmark is available at: https://github.com/powerzoojax/PowerZooJax.

Zhanhua Pan, Xiao Liu, Zhilong Cao +2

Nanyang Technological University, Singapore. · Cornell University, USA. · University of Bristol, UK.

RL BenchmarksConstrained RL
AskPDF
Sep 4, 2026

Physics-Aware Random Walk Fingerprints for Scalable Power Grid Graph Classification

Recent benchmarks such as PowerGraph provide large collections of power-grid graphs for cascading-failure classification. Graph neural networks (GNNs) achieve strong predictive performance on this task, but typically require end-to-end training and model-specific tuning, while their latent representations can be difficult to relate to physically meaningful propagation patterns. Random Walk Fingerprints (RWF) offer a scalable and interpretable alternative, but existing variants primarily emphasise topology and node-level information, leaving grid-relevant operational edge states in the walk dynamics. We propose Multi-Channel Physics-Aware Random Walk Fingerprints (MC-PA-RWF) for power systems, a lightweight graph-level representation framework that introduces physical edge states into random-walk propagation. The method constructs multiple edge-weighted channels from domain-relevant attributes, extracts a channel-specific fingerprint from each weighted graph, and concatenates the resulting vectors into a compact representation. Experiments on three PowerGraph benchmark systems show substantial improvements over topology-only RWF and competitive balanced accuracy against strong GNN baselines, including Graph Convolutional Networks (GCN), Graph Attention Networks (GAT), Graph Isomorphism Networks with edge features (GINE), and Transformer-based Graph Convolutional Networks (TransformerConv). At the largest evaluated settings, the node-edge extension MC-PA-RWF+ achieves around 98.04% - 99.32% balanced accuracy and improves failure-class F1 over the strongest GNN baseline by 1.60 -- 5.84 percentage points, with statistically significant gains across all three systems.

Adnan Anwar

School of Information Technology and a member of the Deakin Cyber Research & Innovation Hub, Deakin University, Melbourne, Australia

Graph ClassificationGraph Representation LearningGraph EmbeddingPower Systems
AskPDF