@inproceedings{e4bdb8bb01e74669b4207ac30ce87ef6,
title = "Hierarchical Energy Management Strategy for Off-Road Hybrid Electric Vehicles",
abstract = "The frequently varying off-road conditions present new challenges for the energy management strategy (EMS) of off- road hybrid vehicles (HEVs). This study aims to propose a hierarchical EMS that fully utilizes the characteristics of power components to meet the requirements of both dynamic performance and economy. The strategy combines wavelet transform with a deep Q-network (DQN), where wavelet transform is used to extract various frequency components of the demand power to fit the characteristics of the engine-generator set, battery, and ultracapacitor. The DQN is employed to allocate the low-frequency demand power with the goals of enhancing fuel efficiency, preserving battery's state of charge (SOC), and extending the life of battery. The superiority of the proposed EMS has been confirmed by contrasting its simulation results with dynamic programming.",
keywords = "EMS, HEV, Reinforcement Learning, Wavelet Transform",
author = "Xuan Zhou and Lijin Han and Ningkang Yang and Mingxing Qiu",
note = "Publisher Copyright: {\textcopyright} 2024 IEEE.; 7th International Conference on Mechatronics and Computer Technology Engineering, MCTE 2024 ; Conference date: 23-08-2024 Through 25-08-2024",
year = "2024",
doi = "10.1109/MCTE62870.2024.11117635",
language = "English",
series = "2024 7th International Conference on Mechatronics and Computer Technology Engineering, MCTE 2024",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "1507--1513",
booktitle = "2024 7th International Conference on Mechatronics and Computer Technology Engineering, MCTE 2024",
address = "United States",
}