Durability enhancement of fuel cell electric vehicles by exploring multi-level learning based control – DEAL

Fuel cell (FC) technology has been showing its advantages as an alternative power source in electric vehicles. Nowadays, most of the FCs in service still suffer from unsatisfactory durability performance, especially for transportation applications. It has been found that the main technical barrier lies in the system control level. A proactive control design dedicated to the enhancement of FC durability is still a fresh topic.
Several difficulties are to be overcome aiming at achieving durability enhancement control. First, the complex effects of operating parameters on FC degradation have to be quantified. Second, the FC system dynamics have to be modelled with adaptation to the FC degradation and operating uncertainty. Third, the durability enhancement control problem has to be formulated properly and feasible control approaches have to be proposed. Inspired by the results that the project team has obtained in the recent research on FC prognostics and health management and the recent remarkable development in machine learning, this project will be dedicated to exploring a multi-level learning based control strategy for FCs used in electric vehicles, especially FC buses. The control strategy will be treated using emerging machine learning methods and in a model predictive control framework. A holistic durability enhancement control framework will be originally formed. The FC durability in electric bus applications is expected to be improved significantly and reach the European 2023 target thanks to the control strategy.
The research developed in this project will be carried out by jointing the advantages of two research teams, i.e., LIS (UMR 7020)/PECASE in Marseille and FEMTO-ST (UMR 6174)/SHARPAC/FCLAB in Belfort.