Model Predictive Control Applied to Hybridised Modular Fuel Cell Systems

This paper proposes a predictive control strategy to minimize the hydrogen consumption of a modular fuel cell system hybridized with a battery. This approach is compared with a reference solution obtained using optimal control. By increasing the prediction horizon, performances tends to the reference ones, minimizing the consumption and aging of the modules while maintaining a solution that can be implemented online. It is shown that the gains in energy consumption and module aging are achieved at the expense of battery aging in both cases.