intelligent Automation of THErmodynamic cycles for a New energy Approach – ATHENA

Background and objectives:
Thermodynamic cycles (power cycles, refrigeration cycles, heat pumps) are essential for the energy transition, as they open up horizons for improving energy efficiency and decentralization (adaptation to local energy sources). There is no universal optimal cycle; optimality depends on precise specifications defining the objectives to be achieved and the constraints of the problem.

Problem:
Traditional heuristic approaches by experts are unable to explore non-intuitive configurations of processes that are often more efficient than classical structures. Superstructure optimization, which uses a predefined architecture with numerous possible unit operations and paths, offers an alternative for identifying non-intuitive process structures. However, this method is limited by the inductive bias of the predefined superstructure.

Innovation:
Increased computing capacity and advances in data science have popularized new algorithms for automatic process generation. Generative approaches, using deep learning algorithms, can generate new process structures, surpassing conventional optimization techniques.

Project objectives:
To develop high-performance tools for the synthesis of innovative thermodynamic cycles. The project combines a methodological study and a practical application to exploit the potential of deep learning in process synthesis. In particular, we aim, on the one hand, to develop process knowledge bases to feed generative methods and, on the other, to develop a tool for determining the optimal structure of any type thermodynamic cycle by optimizing a superstructure generated automatically by a deep learning approach.