​Moving forward, increasing numbers of electric vehicles will have both a battery and a fuel cell. Once the battery and fuel cell have been selected and dimensioned, the next step is to come up with a control law to determine which energy supply should be used to power the vehicle at any given moment to ensure optimal efficiency. CEA-Liten engineers took a dynamic programming algorithm initially developed for managing logistics problems and modified and improved it for this particular use case.

The control law is already part of the MUSE multi-scale multi-physics modeling platform. Here, its formulation was modified to increase speed and streamline the iterative calculation process.

To try and avoid mathematical methods that take too much time to process, the researchers reworked the formulation of the optimization problem and adjusted the cell and battery models used.

The latest release of the software is three times faster than the previous version. It will be instrumental in a number of research projects, including the Gen-Z project to equip a race car with a hydrogen fuel cell for the 2023 Dakar Rally. Other upcoming projects will address use cases in aeronautics, including a personal aircraft.