Towards a smarter battery management system: A critical review on deep learning-based state of charge estimation of lithium-ion batteries

An accurate state of charge (SOC) estimation of lithium-ion batteries underpins a safe and optimized operation of the system. In recent years, deep learning-based SOC estimation has made significant progress. In order to provide researchers in this rapidly advancing field a comprehensive overview of...

Full description

Saved in:
Bibliographic Details
Main Authors: Jiaqi Yao, Julia Kowal
Format: Article
Language:English
Published: Elsevier 2025-09-01
Series:Energy and AI
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S266654682500117X
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849236157050650624
author Jiaqi Yao
Julia Kowal
author_facet Jiaqi Yao
Julia Kowal
author_sort Jiaqi Yao
collection DOAJ
description An accurate state of charge (SOC) estimation of lithium-ion batteries underpins a safe and optimized operation of the system. In recent years, deep learning-based SOC estimation has made significant progress. In order to provide researchers in this rapidly advancing field a comprehensive overview of the state of the art, this paper carries out a structured review on deep learning-based SOC estimation of lithium-ion batteries. A detailed taxonomy of SOC estimation approaches and popularly used public datasets is provided as an introduction to the technical background. A systematic walk-through of the existing deep learning-based SOC estimation approaches, together with the frequently applied optimization strategies, is presented, where we also appeal for a standardized evaluation protocol in this field. As highlight, the current trends and emerging perspectives are pointed out and discussed in detail, including physics-informed neural networks (PINNs), multi-task learning (MTL), few-shot learning, and continual learning. We believe this work could not only provide the researchers and practitioners new to this topic with a clear and detailed manual to start with, but also point out the emerging perspectives for further cutting-edge studies towards a smarter battery management system.
format Article
id doaj-art-bb85ea370604489ea75dab45af8a95f8
institution Kabale University
issn 2666-5468
language English
publishDate 2025-09-01
publisher Elsevier
record_format Article
series Energy and AI
spelling doaj-art-bb85ea370604489ea75dab45af8a95f82025-08-20T04:02:27ZengElsevierEnergy and AI2666-54682025-09-012110058510.1016/j.egyai.2025.100585Towards a smarter battery management system: A critical review on deep learning-based state of charge estimation of lithium-ion batteriesJiaqi Yao0Julia Kowal1Corresponding author.; Department of Electrical Energy Storage Technology (EET), Technische Universität Berlin, Einsteinufer 11, 10587, Berlin, GermanyDepartment of Electrical Energy Storage Technology (EET), Technische Universität Berlin, Einsteinufer 11, 10587, Berlin, GermanyAn accurate state of charge (SOC) estimation of lithium-ion batteries underpins a safe and optimized operation of the system. In recent years, deep learning-based SOC estimation has made significant progress. In order to provide researchers in this rapidly advancing field a comprehensive overview of the state of the art, this paper carries out a structured review on deep learning-based SOC estimation of lithium-ion batteries. A detailed taxonomy of SOC estimation approaches and popularly used public datasets is provided as an introduction to the technical background. A systematic walk-through of the existing deep learning-based SOC estimation approaches, together with the frequently applied optimization strategies, is presented, where we also appeal for a standardized evaluation protocol in this field. As highlight, the current trends and emerging perspectives are pointed out and discussed in detail, including physics-informed neural networks (PINNs), multi-task learning (MTL), few-shot learning, and continual learning. We believe this work could not only provide the researchers and practitioners new to this topic with a clear and detailed manual to start with, but also point out the emerging perspectives for further cutting-edge studies towards a smarter battery management system.http://www.sciencedirect.com/science/article/pii/S266654682500117XLithium-ion batteriesBattery management systemsState of charge estimationDeep learningNeural networks
spellingShingle Jiaqi Yao
Julia Kowal
Towards a smarter battery management system: A critical review on deep learning-based state of charge estimation of lithium-ion batteries
Energy and AI
Lithium-ion batteries
Battery management systems
State of charge estimation
Deep learning
Neural networks
title Towards a smarter battery management system: A critical review on deep learning-based state of charge estimation of lithium-ion batteries
title_full Towards a smarter battery management system: A critical review on deep learning-based state of charge estimation of lithium-ion batteries
title_fullStr Towards a smarter battery management system: A critical review on deep learning-based state of charge estimation of lithium-ion batteries
title_full_unstemmed Towards a smarter battery management system: A critical review on deep learning-based state of charge estimation of lithium-ion batteries
title_short Towards a smarter battery management system: A critical review on deep learning-based state of charge estimation of lithium-ion batteries
title_sort towards a smarter battery management system a critical review on deep learning based state of charge estimation of lithium ion batteries
topic Lithium-ion batteries
Battery management systems
State of charge estimation
Deep learning
Neural networks
url http://www.sciencedirect.com/science/article/pii/S266654682500117X
work_keys_str_mv AT jiaqiyao towardsasmarterbatterymanagementsystemacriticalreviewondeeplearningbasedstateofchargeestimationoflithiumionbatteries
AT juliakowal towardsasmarterbatterymanagementsystemacriticalreviewondeeplearningbasedstateofchargeestimationoflithiumionbatteries