Towards generalization of graph neural networks for AC optimal power flow

Journal Article (2026)
Author(s)

Olayiwola Arowolo (TU Delft - Electrical Engineering, Mathematics and Computer Science)

Jochen L. Cremer (TU Delft - Electrical Engineering, Mathematics and Computer Science, Austrian Institute of Technology)

Research Group
Intelligent Electrical Power Grids
DOI related publication
https://doi.org/10.1016/j.egyai.2026.100842 Final published version
More Info
expand_more
Publication Year
2026
Language
English
Research Group
Intelligent Electrical Power Grids
Journal title
Energy and AI
Volume number
25
Article number
100842
Page Views
162
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

AC Optimal Power Flow (ACOPF) is computationally intensive for large-scale grids, often requiring prohibitive solution times with conventional solvers. Machine learning offers significant speedups, but existing models struggle with scalability and topology flexibility. To address these challenges, we propose a Hybrid Heterogeneous Message Passing Neural Network (HH-MPNN) that integrates a heterogeneous graph neural network (GNN) with a scalable transformer and physics-informed positional encodings. Our approach differs from previous GNN-based ACOPF approaches, which rely on local message passing alone. Our architecture explicitly models distinct power system components to capture local features while using global attention for long-range dependencies. Evaluated on diverse benchmarks including PGLearn and GridFM-DataKit datasets, HH-MPNN achieves <1% optimality gap on default topologies across grid sizes from 14 to 2000 buses on average. For N-1 contingencies, our approach demonstrates zero-shot N-1 generalization with <3% optimality gap on several test cases despite training only on default topologies. We further develop an approach that ensures robust N-1 generalization to high-impact contingencies through targeted augmentation of the training data, showing that exhaustive simulation is unnecessary for topologically flexible models. Finally, size generalization experiments demonstrate that pre-training on small grids significantly improves performance on large-scale systems. Achieving computational speedups of up to 5000 times compared to interior point solvers, these results advance practical, generalizable machine learning for real-time power system operations.