Energy Management Strategy for Fuel Cell Hybrid Power System Considering Fuel Cell Recoverable Performance Loss

Abstract

In the development of fuel cell hybrid vehicles, energy management strategy (EMS) plays an important role. The traditional EMS studies focus on the optimization of energy distribution in order to minimize the hydrogen consumption and degradation cost of fuel cell. However, the phenomenon of fuel cell recoverable performance loss which can affect the efficiency of fuel cell is ignored. Therefore, it is necessary to consider recoverable performance loss in the EMS of fuel cell hybrid power system, which can better simulate practical fuel cell hybrid vehicle scenarios. Although recoverable performance loss can be reversed by performing recovery procedures, fuel cell operation is interrupted during the procedures, thus it is worth studying appropriate time to perform recovery procedures without affecting total power output. To solve these problems, this paper proposed an EMS framework by considering fuel cell recoverable performance. In the study, fuel cell recoverable performance loss is converted to equivalent hydrogen consumption, then an advanced deep reinforcement learning method, Deep Deterministic Policy Gradient (DDPG), is selected to obtain EMS and determine the optimal time for conducting recovery procedure. Results show that with the proposed method, the total cost can be reduced by about 10% compared to EMS without considering fuel cell recoverable performance.