Êtes-vous un étudiant de l'EPFL à la recherche d'un projet de semestre?
Travaillez avec nous sur des projets en science des données et en visualisation, et déployez votre projet sous forme d'application sur Graph Search.
The design of efficient energy systems, through the development of new technologies and the improvement of current ones, requires the use of rigorous process synthesis methods for generating and analysing design alternatives. We introduce a digital twin of process and energy system design that interactively translates needs and preferences of decision makers into an optimization-based model and generates meaningful solutions. The Interactive Digital Twin (InDiT) assists decision makers in steering the exploration of the solution space and guiding them towards relevant system design decisions, taking into account multiple aspects such as the impact of uncertainties and multi-criteria analysis. InDiT enhances step-by-step communication with the decision maker, relying on visual aids to keep the communication during solution generation and exploration intuitive and flexible. In this way, decision makers are guided towards relevant solutions and improve their understanding of relations between the problem definition and system design decisions, while InDiT builds on the decision makers’ preferences and can, after training, suggest solutions that are best-suited to their interests. The novelty of this work lies in the holistic approach of addressing both (i) the systematic generation and exploration of solutions with the assistance of a digital consultant, which translates the decision maker’s needs into machine language and vice versa, and (ii) the interactive step-by-step technique on filtering and evaluating solutions intuitively. This guarantees that the decision maker does not only get solutions based on the design specifications made, but that personal preferences are taken into account during the solution synthesis step, and that the solution space can easily be explored under different criteria. The proposed methodology is demonstrated and applied to the design case of an integrated multi-product biorefinery.