Scalable reinforcement learning for large-scale coordination of electric vehicles using graph neural networks

Abstract As the adoption of electric vehicles (EVs) accelerates, addressing the challenges of large-scale, city-wide optimization becomes critical in ensuring efficient use of charging infrastructure and maintaining electrical grid stability. This study introduces EV-GNN, a novel graph-based solutio...

Full description

Saved in:
Bibliographic Details
Main Authors: Stavros Orfanoudakis, Valentin Robu, E. Mauricio Salazar, Peter Palensky, Pedro P. Vergara
Format: Article
Language:English
Published: Nature Portfolio 2025-07-01
Series:Communications Engineering
Online Access:https://doi.org/10.1038/s44172-025-00457-8
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849768642545188864
author Stavros Orfanoudakis
Valentin Robu
E. Mauricio Salazar
Peter Palensky
Pedro P. Vergara
author_facet Stavros Orfanoudakis
Valentin Robu
E. Mauricio Salazar
Peter Palensky
Pedro P. Vergara
author_sort Stavros Orfanoudakis
collection DOAJ
description Abstract As the adoption of electric vehicles (EVs) accelerates, addressing the challenges of large-scale, city-wide optimization becomes critical in ensuring efficient use of charging infrastructure and maintaining electrical grid stability. This study introduces EV-GNN, a novel graph-based solution that addresses scalability challenges and captures uncertainties in EV behavior from a Charging Point Operator’s (CPO) perspective. We prove that EV-GNN enhances classic Reinforcement Learning (RL) algorithms’ scalability and sample efficiency by combining an end-to-end Graph Neural Network (GNN) architecture with RL and employing a branch pruning technique. We further demonstrate that the proposed architecture’s flexibility allows it to be combined with most state-of-the-art deep RL algorithms to solve a wide range of problems, including those with continuous, multi-discrete, and discrete action spaces. Extensive experimental evaluations show that EV-GNN significantly outperforms state-of-the-art RL algorithms in scalability and generalization across diverse EV charging scenarios, delivering notable improvements in both small- and large-scale problems.
format Article
id doaj-art-33c69aa306424b8abd6eae4edb4e5ea8
institution DOAJ
issn 2731-3395
language English
publishDate 2025-07-01
publisher Nature Portfolio
record_format Article
series Communications Engineering
spelling doaj-art-33c69aa306424b8abd6eae4edb4e5ea82025-08-20T03:03:42ZengNature PortfolioCommunications Engineering2731-33952025-07-014111210.1038/s44172-025-00457-8Scalable reinforcement learning for large-scale coordination of electric vehicles using graph neural networksStavros Orfanoudakis0Valentin Robu1E. Mauricio Salazar2Peter Palensky3Pedro P. Vergara4Intelligent Electrical Power Grids, Delft University of TechnologyIntelligent & Autonomous Systems Group, Centrum Wiskunde & Informatica (CWI)Electrical Energy Systems, Eindhoven University of Technology (TU/e)Intelligent Electrical Power Grids, Delft University of TechnologyIntelligent Electrical Power Grids, Delft University of TechnologyAbstract As the adoption of electric vehicles (EVs) accelerates, addressing the challenges of large-scale, city-wide optimization becomes critical in ensuring efficient use of charging infrastructure and maintaining electrical grid stability. This study introduces EV-GNN, a novel graph-based solution that addresses scalability challenges and captures uncertainties in EV behavior from a Charging Point Operator’s (CPO) perspective. We prove that EV-GNN enhances classic Reinforcement Learning (RL) algorithms’ scalability and sample efficiency by combining an end-to-end Graph Neural Network (GNN) architecture with RL and employing a branch pruning technique. We further demonstrate that the proposed architecture’s flexibility allows it to be combined with most state-of-the-art deep RL algorithms to solve a wide range of problems, including those with continuous, multi-discrete, and discrete action spaces. Extensive experimental evaluations show that EV-GNN significantly outperforms state-of-the-art RL algorithms in scalability and generalization across diverse EV charging scenarios, delivering notable improvements in both small- and large-scale problems.https://doi.org/10.1038/s44172-025-00457-8
spellingShingle Stavros Orfanoudakis
Valentin Robu
E. Mauricio Salazar
Peter Palensky
Pedro P. Vergara
Scalable reinforcement learning for large-scale coordination of electric vehicles using graph neural networks
Communications Engineering
title Scalable reinforcement learning for large-scale coordination of electric vehicles using graph neural networks
title_full Scalable reinforcement learning for large-scale coordination of electric vehicles using graph neural networks
title_fullStr Scalable reinforcement learning for large-scale coordination of electric vehicles using graph neural networks
title_full_unstemmed Scalable reinforcement learning for large-scale coordination of electric vehicles using graph neural networks
title_short Scalable reinforcement learning for large-scale coordination of electric vehicles using graph neural networks
title_sort scalable reinforcement learning for large scale coordination of electric vehicles using graph neural networks
url https://doi.org/10.1038/s44172-025-00457-8
work_keys_str_mv AT stavrosorfanoudakis scalablereinforcementlearningforlargescalecoordinationofelectricvehiclesusinggraphneuralnetworks
AT valentinrobu scalablereinforcementlearningforlargescalecoordinationofelectricvehiclesusinggraphneuralnetworks
AT emauriciosalazar scalablereinforcementlearningforlargescalecoordinationofelectricvehiclesusinggraphneuralnetworks
AT peterpalensky scalablereinforcementlearningforlargescalecoordinationofelectricvehiclesusinggraphneuralnetworks
AT pedropvergara scalablereinforcementlearningforlargescalecoordinationofelectricvehiclesusinggraphneuralnetworks