Battery-degradation-involved energy management strategy based on deep reinforcement learning for fuel cell/battery/ultracapacitor hybrid electric vehicle

To acquire an optimal way to solve the energy management strategy (EMS) of fuel cell hybrid electric vehicles (FCHEVs), most of existing research focuses too much on the protection of fuel cell, while the degree of degradation of battery as an internal influence factor also plays an important role in EMS. In this paper, battery-degradation-involved hierarchical energy management framework utilizing an improved deep deterministic policy gradient (DDPG) algorithm is proposed for gaining the optimal EMS of FCHEV with three power sources. Firstly, to protect fuel cell and battery against the peak power, an adaptive fuzzy filter is employed to complete frequency-based decoupling of power demand for achieving the stratification of power. Then, the degradation model of battery is adopted according to the available data types of our test bench, and an adaptive multi-objective equivalent consumption minimization strategy model is constructed and solved by an improved DDPG-based algorithm. Finally, the simulation results show that, compared with the traditional DDPG, the proposed EMS can enhance the efficiency of fuel cell by 2.02% on average, and reduce the performance degradation of battery by 14.4% on average.