The AI for Energy Grids Lab centers its research on its core theme: extending electricity grid capabilities.
To drive these advancements, the lab leverages key technical AI components, focusing primarily on:
With a commitment to explainable, and responsible AI systems.
The AI for Energy Grids Lab is dedicated to bridging the gap between technical AI research and electricity grid operations to extend grid capabilities.
The lab’s mission and vision are to improve the currently limited transportation capability of today’s electricity grid, facilitating the seamless integration of renewable energy and the connection of distributed energy resources, such as photovoltaics and electric vehicles, directly through end-users to accelerate the transition to low-carbon renewable energy.
The impact of the lab lies in driving a sustainable energy transition by increasing the capacity of the power grid, reducing CO2 emissions, and making energy transportation more cost-effective. This is achieved by validating AI breakthroughs directly within the grid operation teams of Alliander, translating research code into practical open-source software with applied sciences partners, and embedding explainable and responsible AI frameworks to ensure safe, fair, and transparent decision-making across critical public infrastructure.
Graph neural networks & reinforcement learning – Focuses on using Graph Neural Networks combined with Reinforcement Learning to drastically speed up complex grid planning, switching, and simulation calculations.
State-estimation with AI – Focuses on using AI and Bayesian statistics to improve the accuracy, computational speed, and robustness of electricity grid state estimators for future distribution grids.
Risk-based investments and operation – Focuses on quantifying risks and ranking short-term flexibility market options against traditional network reinforcements when grid capacity is limited.
Decentralized control of energy flows – Focuses on shifting from centralized systems to decentralized controllers using edge hardware to process data and make decisions locally at station levels.
Responsible & trustworthy AI for distribution system operation – Focuses on deploying AI safely, fairly, and ethically within critical grid infrastructure to ensure transparent and autonomous decision-making.
The power of assumptions: A literature review on how algorithmic design influences energy justice in electrical distribution grids Journal Article
In: vol. 133, pp. 104605, 2026, ISSN: 2214-6296.
Uncertainty Quantification Using GNN Deep Ensembles for Contingency Decisions in the MV-Grid Journal Article
In: vol. 20, no. 1, pp. e70262, 2026.
Exploring the impact of adaptive rewiring in Graph Neural Networks Journal Article
In: pp. arXiv–2602, 2026.
Home Energy Management using End-to-End Learning Proceedings Article
In: 2025 IEEE PES Innovative Smart Grid Technologies Conference Europe (ISGT Europe), pp. 1–5, 2025.
Towards a system-theoretic approach to algorithmic (un) fairness Proceedings Article
In: European Workshop on Algorithmic Fairness, pp. 303–309, PMLR, 2025.
GNN deep ensembles for N-1 contingency decisions Proceedings Article
In: IET Conference Proceedings CP922, pp. 2435–2439, The Institution of Engineering and Technology, 2025.
Scalable quantile predictions of peak loads for non-residential customer segments Proceedings Article
In: 2025 IEEE PES Innovative Smart Grid Technologies Conference Europe (ISGT Europe), pp. 1–5, IEEE, 2025.
Adapting to limited grid capacity: Perceptions of injustice emerging from grid congestion in the Netherlands Journal Article
In: vol. 122, pp. 103962, 2025.
Relying on Artificial Intelligence in Electrical Distribution Grids: Opportunities, Risks, and Regulation Journal Article
In: no. 145, 2025.
Message passing-based inference in an autoregressive active inference agent: 6th International Workshop on Active Inference, IWAI 2025 Journal Article
In: International Workshop on Active Inference, 2025.
Relation Between Electrical Grid Congestion and Bus Characteristics Proceedings Article
In: 2024 IEEE PES Innovative Smart Grid Technologies Europe (ISGT EUROPE), pp. 1–6, 2024.
Alliander is a network company that emerged from Nuon, now Vattenfall. Alliander consists of the components Liander , Qirion . Liander, a regional grid operator , provides energy distribution in a third of the Netherlands
Delft University of Technology (TU Delft) is a technical university in Delft. Top education and research are at the heart of the oldest and largest technical university in the Netherlands.
HAN University of Applied Sciences is one of the largest universities of applied sciences in the Netherlands.
Radboud University (RU) is a general university in The Netherlands, active in almost all scientific fields, and one of the leading universities worldwide.
University of Twente is a public technical university located in Enschede, Netherlands.
