Description

This project studies and demonstrates the stakeholder needs for explainability and the required technologies to achieve explainable AI systems that support a sustainable energy system. We currently focus on demand-side flexibility or increased flexibility in how we consume energy to better match the characteristics of the energy grid.

Problem Context

Due to the trend in the way we use utility buildings (~480,000 in NL), these buildings have a significant potential to provide flexibility and relieve the grid which is currently congested in large parts of the Netherlands. Therefore, we are scoping this project around these utility buildings, with the ambition to expand to residential buildings. Through this project we are creating the capability for TNO to be positioned in the supporting role towards actors such as DSOs, aggregators and building owners on the implementation of explainable AI systems to unlock the potential of demand side flexibility.

The research question: How should AI systems be explained towards users to guarantee trustworthiness, as support tools that provide the necessary insights to make utility buildings use and generate their energy in a more flexible manner?

Solution

With the development of the FUN demo we have a tool to raise awareness on the value and necessity of XAI at energy system actors and AI developers. The tool can be utilized as a resources for existing AI to add explainability and for future AI development to include explainability by design. To this end we’ve build two demonstrators. One for explainability in Building AI and one for explainability in Grid AI. We developed these demonstrators in a multi-disciplinary joint effort between Defense, Safety and Security (Sophie van Gent and Jasper van der Waa), ICT, Strategy & Policy (Devin Diran, Ioannis Tolios and Dennis Starrenburg) and Energy and Materials Transitions.

Next steps

This project results in demonstrators which generate actionable knowledge on the value, possibilities and necessity of XAI for demand side flexibility. We aim to utilize the developed demo’s with actors such as DSO’s and Building Energy Management system providers to increase the explainability of smart systems to unlock demand side flexibility in a trustworthy manner. Moreover, we seek partners with explainability needs for smart/AI systems and look to develop demonstrators for new use-cases in the future.

Resources

Please see the links below to go to the demonstrators:

Contact

  • Devin Diran, TNO Vector, e-mail: devin.diran@tno.nl