Battery-Involved Energy Management for Hybrid Electric Bus Based on Expert-Assistance Deep Deterministic Policy Gradient Algorithm

Energy management is an enabling technique to guarantee the reliability and economy of hybrid electric systems. This paper proposes a novel machine learning-based energy management strategy for a hybrid electric bus (HEB), with an emphasized consciousness of both thermal safety and degradation of th...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on vehicular technology Vol. 69; no. 11; pp. 12786 - 12796
Main Authors: Wu, Jingda, Wei, Zhongbao, Liu, Kailong, Quan, Zhongyi, Li, Yunwei
Format: Journal Article
Language:English
Published: New York IEEE 01.11.2020
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects:
Algorithms
Batteries
Battery health
deep deterministic policy gradient
Degradation
Energy management
expert assistance
hybrid electric bus
Hybrid systems
Lithium-ion batteries
Machine learning
Optimization
Product safety
Rechargeable batteries
Thermal degradation
Thermal management
thermal safety
Traction motors
ISSN:0018-9545, 1939-9359
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Energy management is an enabling technique to guarantee the reliability and economy of hybrid electric systems. This paper proposes a novel machine learning-based energy management strategy for a hybrid electric bus (HEB), with an emphasized consciousness of both thermal safety and degradation of the onboard lithium-ion battery (LIB) system. Firstly, the deep deterministic policy gradient (DDPG) algorithm is combined with an expert-assistance system, for the first time, to enhance the "cold start" performance and optimize the power allocation of HEB. Secondly, in the framework of the proposed algorithm, the penalties to over-temperature and LIB degradation are embedded to improve the management quality in terms of the thermal safety enforcement and overall driving cost reduction. The proposed strategy is tested under different road missions to validate its superiority over state-of-the-art techniques in terms of training efficiency and optimization performance.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0018-9545
1939-9359
DOI:10.1109/TVT.2020.3025627