摘要
Due to the fast-filling capability and zero-emission, hydrogen fuel cell vehicle (FCV) is a new hot spot for transportation electrification. However, the hydrogen filling process will cause the internal temperature of the hydrogen storage tank (HST) to rapidly increase, reaching safety limits, thereby limiting the state of charge (SoC) of the HST and the long-distance navigation of FCVs. To solve the conflict between temperature rise and SoC, this paper develops a parallel filling strategy with an adaptive dormant stage based on deep reinforcement learning. The results show that the proposed hydrogen filling strategy can achieve higher SoC under different initial conditions, and this strategy is more effective at low initial temperature and low initial SoC.
源语言 | 英语 |
---|---|
页(从-至) | 1 |
页数 | 1 |
期刊 | IEEE Transactions on Transportation Electrification |
DOI | |
出版状态 | 已接受/待刊 - 2023 |