The AI.grids Initiative

AI.grids is developing an AI foundation model for energy system operations. The project brings together system operators, researchers, and industry partners to create and validate AI applications in the energy domain using data from European grids and training ressources from integrated AI Factories.

As Europe's energy system becomes increasingly complex, new approaches are needed to support reliable and efficient grid operations. The integration of renewable energy sources, growing electrification, and the increasing number of distributed energy resources are creating new challenges for system operators across the continent.

To address these developments, AI.grids has been launched under the auspices of the Directorate General for Energy (DG ENER) of the European Commission (EC). The initiative constitutes one of two flagship projects of the »Strategic roadmap for digitalisation and AI in energy« (Link to EC webpage). It aims to develop the first pan-European foundation AI model for energy system operations and establish a collaborative ecosystem that brings together system operators, research organisations, and technology providers. The signing ceremony on June 3rd saw 48 partners join the consortium.

Drawing on diverse datasets from European transmission and distribution networks, AI.grids will leverage secure data-sharing frameworks and high-performance computing resources from European AI Factories to train and validate advanced AI models. The project will explore a range of operational applications, including:

  • forecasting,
  • congestion management,
  • asset monitoring,
  • planning support, and 
  • flexibility services.

Within the consortium the Fraunhofer Center Digital Energy leads WG2A, building the data foundation that enables AI.grids. Secure exchange connectors ensure all data and model sharing remains governed and aligned with the Common European Energy Data Space (CEEDS). The tasks of working group include:

  •  define the Minimum Viable Dataset for AI model training, 
  • design interoperable data architectures across European TSOs and DSOs, and 
  • develope synthetic data frameworks for cases where real operational data cannot be shared. 

Over the coming years, AI.grids will help lay the foundation for the next generation of AI-enabled grid operations, fostering collaboration across the European energy sector and accelerating the practical adoption of trustworthy AI technologies.