EUVIMEDEUVIMED entdecken
Crossref

Uncertainty-Aware Deep Reinforcement Learning for Trainable Equivalent Consumption Minimization Strategy of Fuel Cell Hybrid Electric Tracked Vehicle

Qicong Su, Ruchen Huang, Zhendong Zhang, Yiwen Shou, Hongwen He

IEEE Transactions on Transportation Electrification · 2025-08 · 11 (4) · 10310-10321

DOI: 10.1109/tte.2025.3563203 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Transportation Electrification, 2025-08. Verfasst von Qicong Su und 4 weiteren Autor:innen. Die Arbeit wurde bisher 18-mal zitiert (Crossref) und nennt 35 Literaturangaben. Gefördert von: Ministry Support Project of China, National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Energy management strategies (EMSs) are pivotal in optimizing energy efficiency for vehicles equipped with hybrid electric powertrains. In spite of the growing adoption of deep reinforcement learning (DRL)-based approaches, challenges persist in achieving satisfactory optimization performance and maintaining reliable control. Motivated by this, this article introduces a novel trainable equivalent consumption minimization strategy (ECMS) framework for fuel cell hybrid electric tracked vehicles (FCHETVs) with uncertainty-aware control. First, the proposed framework employs a DRL algorithm to dynamically determine and optimize the equivalent factor in the ECMS method, facilitating improved fuel economy. Then, the soft actor-critic (SAC) algorithm is formulated for efficient policy learning. In order to further enhance control reliability, an ensembled policy network method is incorporated to measure uncertainty and mitigate suboptimal actions, thereby improving decision-making robustness. Simulation results reveal that the SAC-based trainable ECMS achieves significant fuel economy improvements, outperforming the traditional SAC and adaptive ECMS (A-ECMS) methods by 2.93% and 5.15%, respectively. The ensemble model, moreover, ensures reliable and effective control, with online testing results indicating an additional 2.08% improvement in fuel economy. These findings underscore the effectiveness of integrating learning-based and optimization-based approaches in EMS design, offering a robust pathway to reducing energy consumption and promoting sustainable transportation solutions.

Förderung

  • Ministry Support Project of China
  • National Natural Science Foundation of China

Zitieren

APA

Su, Q., Huang, R., Zhang, Z., Shou, Y., & He, H. (2025). Uncertainty-Aware Deep Reinforcement Learning for Trainable Equivalent Consumption Minimization Strategy of Fuel Cell Hybrid Electric Tracked Vehicle. IEEE Transactions on Transportation Electrification, 11(4), 10310-10321. https://doi.org/10.1109/tte.2025.3563203

Vancouver

Su Q, Huang R, Zhang Z, Shou Y, He H. Uncertainty-Aware Deep Reinforcement Learning for Trainable Equivalent Consumption Minimization Strategy of Fuel Cell Hybrid Electric Tracked Vehicle. IEEE Transactions on Transportation Electrification. 2025;11(4):10310-10321. doi:10.1109/tte.2025.3563203

BibTeX
@article{Su2025,
  author = {Su, Qicong and Huang, Ruchen and Zhang, Zhendong and Shou, Yiwen and He, Hongwen},
  title = {Uncertainty-Aware Deep Reinforcement Learning for Trainable Equivalent Consumption Minimization Strategy of Fuel Cell Hybrid Electric Tracked Vehicle},
  journal = {IEEE Transactions on Transportation Electrification},
  year = {2025},
  volume = {11},
  number = {4},
  pages = {10310--10321},
  doi = {10.1109/tte.2025.3563203}
}
Kostenlos starten →