Testing and validating AI solutions in the energy sector
The EnergyGuard project is developing a European-scale testing and experimentation facility for AI in the energy sector.

AI is rapidly transforming how Europe produces, distributes and consumes energy. To maintain technological leadership and ensure ethical, secure AI deployment, the European Commission launched the AI continent action plan – a comprehensive roadmap to position the EU as a global hub for trustworthy AI.
At the centre of this plan are testing and experimentation facilities (TEFs) – large-scale, open-access environments that allow innovators to validate AI applications under real operational conditions. While such facilities already exist in healthcare, manufacturing and smart cities, EnergyGuard is pioneering their application in the energy domain.
Funded under Horizon Europe and co-financed by the European Union, EnergyGuard is establishing a network of Energy AI TEFs to accelerate the testing, validation and deployment of AI-driven energy solutions across Europe. Its overarching aim is to support the EU Green Deal and Digital Decade objectives by integrating trustworthy AI into the heart of the energy transition.
What Is EnergyGuard?
EnergyGuard is built on a European structure of five interconnected TEFs that cover different key areas of the energy domain including smart grid and microgrid, hydrogen, flexible communities and energy efficient buildings.
Each TEF focuses on a specific energy vector, but they operate as one pan-European infrastructure. The facilities offer both virtual and physical testing environments that replicate real-world operating conditions for AI validation.
Its mission is clear: to accelerate the integration of trustworthy, efficient and secure AI across the entire energy value chain, allowing innovators, start-ups and SMEs to bring cutting-edge AI solutions from the lab to the market faster.
At its core, EnergyGuard integrates high fidelity digital twins representing complex energy infrastructures such as:
• Electricity networks with high shares of renewables;
• Microgrids and energy communities;
• Hydrogen production and storage systems;
• Buildings and mobility applications.
These digital twins serve as the foundation for AI experimentation. Developers can deploy algorithms, train models and benchmark performance against real operational data, all within a secure, interoperable framework.
Powered by MeluXina, one of the most energy efficient supercomputers in Europe, EnergyGuard combines high performance and low carbon computing to enable large-scale AI testing without compromising on sustainability. This conscious use of high performance computing ensures energy innovators can reliably train, simulate and validate complex AI models securely and responsibly.
Each TEF provides access not only to datasets and simulation models but also to AI inference for APIs, software services, and validation tools – all accessible through a unified AI development and testing environment.
Accelerating AI innovation
Trustworthiness is central to EnergyGuard’s mission. The project has developed a framework that can help innovators to design and verify AI-based systems that are consistent with European principles of safety, transparency and compliance.
Through the EnergyGuard AI sandbox, developers can evaluate and refine their solutions by evaluating them under real world energy conditions, using shared datasets, digital twins and benchmarking tools before advancing to the acceptance environment, where performance and conformity with the EU AI Act are formally assessed to ensure readiness for real-world deployment.
This approach allows innovators to show both compliance and quality, building trust among energy stakeholders, regulators and investors.
This pathway enables innovators to demonstrate both compliance and quality, fostering confidence among energy stakeholders, regulators, and investors.
Beyond testing, EnergyGuard acts as a bridge across Europe’s digital and AI ecosystems. It connects seamlessly with:
• AI-on-demand platforms;
• European data spaces;
• Digital innovation hubs;
• Other sectoral AI TEFs.
This interoperability ensures that participants can integrate EnergyGuard’s assets with other EU initiatives, enabling a collaborative, data-driven innovation environment that strengthens Europe’s position as a global AI hub.
Why EnergyGuard matters
By offering open, scalable and secure access to world-class facilities, EnergyGuard allows European energy innovators to move from prototypes to market-ready, AI-driven solutions faster than ever before.
Its impact reaches beyond technology and directly supports the EU’s green and digital transition, helping to:
•Accelerate renewable integration;
•Enhance grid flexibility and efficiency;
•Strengthen energy resilience and cybersecurity;
•Foster cross-sector innovation between electricity, hydrogen, heat and mobility systems.
Through collaboration, transparency, and innovation, EnergyGuard is paving the way toward a smarter, more sustainable European energy landscape powered by trustworthy AI.
Discover how EnergyGuard is transforming AI innovation in Europe’s energy sector at its website.
About the author
Sotiris Pelekis is a Senior Researcher and AI Engineer in the Decision Support Systems Laboratory at the National Technical University of Athens. He holds a PhD and an MSc from NTUA and specialises in trustworthy AI and advanced data analytics for the energy and buildings sectors including energy forecasting, demand response optimisation and cybersecurity.











