R&D Units
Artificial Intelligence
Trustworthy and explainable AI to support multi-scale energy decision-making under uncertainty
What we do?
The Unit investigates the application of trustworthy and explainable AI to support decision-making in multi-scale energy systems.
We apply the AI techniques best suited to each challenge—including forecasting, optimisation, simulation and LLM-based interfaces—to deliver actionable, accessible and interpretable outputs. Our research addresses challenges across multiple scales, from buildings and local energy communities to energy markets and regulatory policy.
We integrate real-world data, computational models and physical principles to develop AI systems that improve energy efficiency, demand flexibility, hybridisation and the integration of renewable energy into energy systems.
Research Areas
- Explainable AI for trustworthy energy models, ensuring that AI model outputs are understandable, traceable and useful for both technical and non-technical users.
- Application of machine learning and optimisation methods to demand response, charging and discharging scheduling, ancillary service price forecasting, market decision-making, dispatch and operational planning under risk and uncertainty.
- Complementing the expertise of energy systems specialists with the most appropriate AI methods and knowledge representation techniques, with a particular focus on reproducibility, validation and the transition from research prototypes to solutions geared towards industrial Minimum Viable Products (MVPs).
- Use of what-if scenarios and causal models to assess alternative decisions, policies and interventions, including energy storage deployment, tariff changes, flexibility activation, demand response and regulatory measures.
Strategic Objectives
- AI-based decision support for complex energy systems under uncertainty.
- Integrated approaches combining forecasting, scenario simulation and optimisation to support action-oriented objectives that go beyond descriptive data analysis.
- Prototyping with real-world data to advance scalable research, create opportunities for public-private collaboration and facilitate technology transfer.
Projects
Milestones
- More than 15 years of experience in AI across academia and industry, applying artificial intelligence to a wide range of challenges in sustainability, biotechnology and energy.
- Development of more than 10 AI/ML-based products for biotechnology companies, including recommender systems, predictive models and decision-support tools.
- Development of three patented AI-based technologies designed to transform scientific data into scalable applications.
