Scalable Heterogeneous Graph Foundation Models for Data-Driven Optimal Power Flow in Smart Grids

• URL: http://arxiv.org/abs/2605.23194v1
• Date: Fri, 22 May 2026 03:25:25 GMT
• Title: Scalable Heterogeneous Graph Foundation Models for Data-Driven Optimal Power Flow in Smart Grids
• Abstract summary: This paper presents a scalable heterogeneous graph neural network (GNN) workflow for data-driven OPF surrogate modeling and OPF-GFM development.<n>The workflow preserves the distinct node and edge types of power grids and supports distributed preprocessing, training, and downstream fine-tuning on leadership-class supercomputers.
• Score: 7.737263980113792
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Fast and reliable optimal power flow (OPF) approximation is essential for reliable smart-grid operation, yet many learning-based surrogates either flatten the native heterogeneous structure of power networks, target a limited set of grid topologies, or lack scalable infrastructure for graph foundation model (GFM) training. This paper presents a scalable heterogeneous graph neural network (GNN) workflow, built on HydraGNN, for data-driven OPF surrogate modeling and OPF-GFM development. The workflow preserves the distinct node and edge types of power grids -- buses, generators, loads, shunts, AC lines, transformers, and device-to-bus couplings -- and supports distributed preprocessing, training, hyperparameter optimization (HPO), and downstream fine-tuning on leadership-class supercomputers. Using three million heterogeneous graph instances spanning ten PGLib-OPF cases, from 14 to 13,659 buses, we conduct DeepHyper-driven HPO on the ORNL Frontier supercomputer. The campaign identifies compact models ($\sim$1.6--1.7M parameters) with the lowest validation losses. Downstream experiments on feasibility classification and N-1 contingency regression show that fine-tuning pretrained OPF GFM improves low-data accuracy, stabilizes training, accelerates convergence, and reduces adaptation cost when partial or head-only fine-tuning is used.

Related papers

• GENCO - A Unified Neural Solver Embedded in a Development Framework for Steady-State Grid Analysis [6.9646837562713]
  We present GENCO, a unified solver for steady-state transmission grid analysis.<n>It handles power flow (PF), optimal power flow (OPF), runtime state estimation (SE) within a single architecture.<n>We release large-scale datasets with millions PF and OPF scenarios across diverse grid topologies to support reproducible benchmarking.
  arXiv  Detail & Related papers  (2026-08-10T17:57:49Z)
• MxGPS: Multiplex Graph Transformers for a Power Grid Foundation Model [0.7951418336539415]
  Single-task fine-tuning of graph neural networks (GNNs) for power grid problems exhibits a systematic failure mode.<n>We term this topology overfitting: the tendency of task-specific gradient signals to encode relational structure particular to the training topologies rather than the underlying physics.<n>We introduce MxGPS (Multiplex GPS), a multiplex graph transformer that runs K task-specialised GPS branches over a shared node encoder.
  arXiv  Detail & Related papers  (2026-07-15T12:22:36Z)
• F\textsuperscript{2}LP-AP: Fast \& Flexible Label Propagation with Adaptive Propagation Kernel [6.834330245672611]
  textbfF$2$LP-AP is a training-free, computationally efficient framework that adapts to local graph topology.<n>textbfF$2$LP-AP achieves competitive or superior accuracy compared to trained GNNs.
  arXiv  Detail & Related papers  (2026-04-22T16:23:17Z)
• EUGens: Efficient, Unified, and General Dense Layers [56.498769704575544]
  We propose a new class of dense layers that generalize standard fully-connected feedforward layers, textbfEfficient, textbfUnimat and textbfGeneral dense layers (EUGens)<n>EUGens leverage random features to approximate standard FFLs and go beyond them by incorporating a direct dependence on the input norms in their computations.
  arXiv  Detail & Related papers  (2026-01-30T05:01:03Z)
• Towards Generalization of Graph Neural Networks for AC Optimal Power Flow [0.0]
  Machine learning approaches offer computational speedups but struggle with scalability and topology adaptability without retraining.<n>We propose a Hybrid Heterogeneous Message Passing Neural Network (HH-MPNN)<n> HH-MPNN models buses, generators, loads, shunts, transmission lines and transformers as distinct node or edge types.<n> Applied zero-shot to thousands of unseen topologies, HH-MPNN achieves less than 3% optimality gap despite training only on default topologies.
  arXiv  Detail & Related papers  (2025-10-08T10:28:46Z)
• TGLF-SINN: Deep Learning Surrogate Model for Accelerating Turbulent Transport Modeling in Fusion [18.028061388104963]
  We propose textbfTGLF-SINN (Spectra-Informed Neural Network) with three key innovations.<n>Our approach achieves superior performance with significantly less training data.<n>In downstream flux matching applications, our NN surrogate provides 45x speedup over TGLF while maintaining comparable accuracy.
  arXiv  Detail & Related papers  (2025-09-07T09:36:51Z)
• Constraints and Variables Reduction for Optimal Power Flow Using Hierarchical Graph Neural Networks with Virtual Node-Splitting [0.24554686192257422]
  Power system networks are often modeled as homogeneous graphs, which limits the ability of graph neural network (GNN) to capture individual generator features at the same nodes.<n>By introducing the proposed virtual node-splitting strategy, generator-level attributes like costs, limits, and ramp rates can be fully captured by GNN models.<n>Two-stage adaptive hierarchical GNN is developed to (i) predict critical lines that would be congested, and then (ii) predict base generators that would operate at the maximum capacity.
  arXiv  Detail & Related papers  (2024-11-09T19:46:28Z)
• FusionLLM: A Decentralized LLM Training System on Geo-distributed GPUs with Adaptive Compression [55.992528247880685]
  Decentralized training faces significant challenges regarding system design and efficiency. We present FusionLLM, a decentralized training system designed and implemented for training large deep neural networks (DNNs) We show that our system and method can achieve 1.45 - 9.39x speedup compared to baseline methods while ensuring convergence.
  arXiv  Detail & Related papers  (2024-10-16T16:13:19Z)
• GP CC-OPF: Gaussian Process based optimization tool for Chance-Constrained Optimal Power Flow [54.94701604030199]
  The Gaussian Process (GP) based Chance-Constrained Optimal Flow (CC-OPF) is an open-source Python code for economic dispatch (ED) problem in power grids. The developed tool presents a novel data-driven approach based on the CC-OP model for solving the large regression problem with a trade-off between complexity and accuracy.
  arXiv  Detail & Related papers  (2023-02-16T17:59:06Z)
• Learning to Solve the AC-OPF using Sensitivity-Informed Deep Neural Networks [52.32646357164739]
  We propose a deep neural network (DNN) to solve the solutions of the optimal power flow (ACOPF) The proposed SIDNN is compatible with a broad range of OPF schemes. It can be seamlessly integrated in other learning-to-OPF schemes.
  arXiv  Detail & Related papers  (2021-03-27T00:45:23Z)
• A Meta-Learning Approach to the Optimal Power Flow Problem Under Topology Reconfigurations [69.73803123972297]
  We propose a DNN-based OPF predictor that is trained using a meta-learning (MTL) approach. The developed OPF-predictor is validated through simulations using benchmark IEEE bus systems.
  arXiv  Detail & Related papers  (2020-12-21T17:39:51Z)
• High-Fidelity Machine Learning Approximations of Large-Scale Optimal Power Flow [49.2540510330407]
  AC-OPF is a key building block in many power system applications. Motivated by increased penetration of renewable sources, this paper explores deep learning to deliver efficient approximations to the AC-OPF.
  arXiv  Detail & Related papers  (2020-06-29T20:22:16Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.