Please use this identifier to cite or link to this item:
http://hdl.handle.net/10397/114259
Title: | Health-conscious energy management for fuel cell vehicles : an integrated thermal management strategy for cabin and energy source systems | Authors: | Jia, C Liu, W He, H Chau, KT |
Issue Date: | 1-Oct-2025 | Source: | Energy, 1 Oct. 2025, v. 333, 137330 | Abstract: | Operating temperature significantly affects the efficiency and durability of the fuel cell (FC) system and lithium-ion battery (LIB). However, existing energy management strategies (EMS) tend to ignore the electric-thermal coupling characteristics of energy source systems during decision-making, which limits the economic potential of fuel cell vehicles (FCV). To address this challenge, this paper proposes a novel health-conscious energy management paradigm that integrates comprehensive thermal management of the energy source systems and cabin, aiming to maximize the overall performance and economy of FCVs. Specifically, by constructing electric-thermal coupling lifespan models of the LIB and FC system, as well as a cabin dynamic thermal load model, we developed a comprehensive control framework for energy- and thermal coupling. The objectives of this framework are to optimize energy consumption, thermal health management of the energy source systems, and cabin comfort. On this basis, the state-of-the-art twin delayed deep deterministic policy gradient (TD3) algorithm is employed to achieve collaborative optimization of the onboard energy source systems and air conditioning system. This collaborative optimization can further optimize vehicle energy consumption, achieving the best balance between fuel economy, cabin comfort, and energy source systems durability. The results show that, compared with conventional TD3 EMS, the proposed EMS extends the lifespan of the LIB by 32.16 % and the FC system by 14.63 % in terms of energy source system health management. Additionally, in terms of total operational costs, the proposed EMS enhances the driving economy by 11.19 %. | Keywords: | Comprehensive thermal management Deep reinforcement learning Energy management strategy Energy source systems health management Fuel cell bus |
Publisher: | Pergamon Press | Journal: | Energy | ISSN: | 0360-5442 | EISSN: | 1873-6785 | DOI: | 10.1016/j.energy.2025.137330 |
Appears in Collections: | Journal/Magazine Article |
Show full item record
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.