Artificial intelligence (AI) is becoming a key element in accelerating the energy transition in Europe. From the integration of renewable energies to the management of complex networks, AI-based systems offer significant opportunities to improve the efficiency, resilience, and sustainability of the energy system. However, the deployment of these technologies also poses critical challenges in terms of security, reliability, explainability, and regulatory compliance.
In this context, the European Union has positioned human-centered and trustworthy AI as a strategic pillar, reinforced by initiatives such as the AI Act and the creation of sectoral Testing and Experimentation Facilities (TEFs). These infrastructures enable the validation of AI solutions in real-world conditions, reducing risks and facilitating their responsible adoption by industry and the public sector.
CIEMAT: public research at the service of responsible AI in energy
The Center for Energy, Environmental and Technological Research (CIEMAT) is a Spanish public research organization with a long history of developing and validating advanced energy technologies. Through its CEDER-CIEMAT center, which specializes in distributed energy systems, renewables, and storage, CIEMAT actively contributes to the testing of innovative solutions that support the decarbonization of the European energy system.
In the field of artificial intelligence, CIEMAT provides an integrated vision that combines knowledge of the energy domain, real physical infrastructures, and a rigorous approach to the technical and operational evaluation of digital solutions. This combination is essential to ensure that AI systems are not only technically advanced, but also secure, robust, and aligned with European values.
CIEMAT's capabilities in testing and experimentation with AI applied to energy
CIEMAT's main contribution to the AI experimentation ecosystem revolves around CEDER-CIEMAT's smart microgrid, an infrastructure operated under real conditions that covers a wide range of energy technologies.
Its main capabilities include:
- Operational electrical and thermal microgrid, with diversified renewable generation (wind, photovoltaic, hydraulic).
- Electrochemical storage systems and anticipated sector coupling capabilities with hydrogen within the framework of the EnergyGuard project.
- Electric vehicle charging infrastructure and advanced demand management.
- Real-time monitoring and control systems, which generate high-quality data for training and validating AI models.
These capabilities allow AI algorithms to be evaluated in real operating scenarios, addressing critical aspects such as robustness in the face of unforeseen events, interaction with human operators, and the impact of algorithmic decisions on system stability.
Contribution to reliable AI aligned with the European framework
One of the main challenges identified by the European Commission is the gap between the development of AI models and their safe adoption in critical sectors such as energy. CIEMAT contributes to closing this gap by providing an environment where solutions can undergo:
- Reliability and resilience assessments under real conditions.
- Technical and operational risk analysis, including cybersecurity.
- Validation of explainability and human oversight, key aspects for AI acceptance.
- Preparation for regulatory compliance, in line with the emerging requirements of the AI Act.
This approach is particularly relevant for SMEs and startups, which often lack access to large-scale infrastructure or specialized resources to validate their solutions before market entry.
CIEMAT in EnergyGuard: an European TEF for AI in energy
CIEMAT is a partner in the EnergyGuard project, a large-scale Testing and Experimentation Facility funded by Horizon Europe, which aims to support the development, validation, and deployment of reliable AI solutions across the entire energy value chain. In EnergyGuard, the CEDER-CIEMAT infrastructure is integrated as one of the key physical environments, complementing other European facilities in transmission networks, buildings, hydrogen, and energy communities. This integration enables:
- Multi-vector experimentation (electricity, heat, and hydrogen).
- Access to real data and synchronized digital twins.
- Joint assessments of technical performance, energy impact, and AI risks.
- Support for regulatory sandboxes and evidence-based policy making.
More information about the project is available at: https://energy-guard.eu
Invitation to collaboration and European dialogue
At CIEMAT, we believe that the transition to reliable AI in the energy sector requires collaboration between research, industry, regulators, and citizens. Testing and Experimentation Facilities such as EnergyGuard offer a specific space for this dialogue, where technological innovation can advance responsibly and in line with European objectives.
CIEMAT invites AI developers, SMEs, public entities, and other actors in the ecosystem to explore opportunities for collaboration, experimentation, and joint learning, thus contributing to a more sustainable, resilient, and reliable European energy system.
To take a deep dive into how EnergyGuard is testing and validating AI solutions for the energy sector, please visit https://energy-guard.eu
Article produced by Marta Lafuente Cacho at Ciemat.
- Login for at skrive kommentarer