An intelligent full-knowledge transferable collaborative eco-driving framework based on improved soft actor-critic algorithm

Ruchen Huang, Hongwen He and Qicong Su

Applied Energy, 2024, vol. 375, issue C, No S0306261924014612

Abstract: Eco-driving is a promising technology for fuel cell vehicles (FCVs) that simultaneously achieves safe driving and energy saving in the urban transport sector, particularly through the application of cutting-edge deep reinforcement learning (DRL). However, developing specific DRL-based eco-driving strategies for different FCVs is a laborious task, since repetitive training is required when encountering various FCV types. To tackle this challenge, this paper proposes an intelligent transferable collaborative eco-driving framework across FCV types. Firstly, the eco-driving problem in the vehicle-following scenario is formulated by collaboratively integrating adaptive cruise control (ACC) with energy management strategy (EMS), and then an improved soft actor-critic (I-SAC) algorithm is designed to solve this problem. After that, a source eco-driving strategy based on I-SAC is pre-trained for a light fuel cell hybrid electric vehicle (FCHEV). Finally, all learned knowledge in the source strategy is fully transferred and reused for a heavy-duty fuel cell hybrid electric bus (FCHEB) to get the target eco-driving strategy. Experimental simulations show that the proposed framework can expedite the development of the eco-driving strategy for FCHEB by 94.83% while reducing hydrogen consumption by 10.05%.

Keywords: Fuel cell vehicle (FCV); Eco-driving strategy; Deep reinforcement learning (DRL); Full-knowledge transfer; Improved soft actor-critic (I-SAC) (search for similar items in EconPapers)
Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (2)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0306261924014612
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:appene:v:375:y:2024:i:c:s0306261924014612

Ordering information: This journal article can be ordered from
http://www.elsevier.com/wps/find/journaldescription.cws_home/405891/bibliographic
http://www.elsevier. ... 405891/bibliographic

DOI: 10.1016/j.apenergy.2024.124078

Access Statistics for this article

Applied Energy is currently edited by J. Yan

More articles in Applied Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().