Advancing state of health estimation for electric vehicles: Transformer-based approach leveraging real-world data

The widespread adoption of electric vehicles (EVs) underscores the urgent need for innovative approaches to estimate their lithium-ion batteries’ state of health (SOH), which is crucial for ensuring safety and efficiency. This study introduces SOH-TEC, a transformer encoder-based model that processe...

Full description

Saved in:
Bibliographic Details
Main Authors: Kosaku Nakano, Sophia Vögler, Kenji Tanaka
Format: Article
Language:English
Published: Elsevier 2024-12-01
Series:Advances in Applied Energy
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S266679242400026X
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850246616367235072
author Kosaku Nakano
Sophia Vögler
Kenji Tanaka
author_facet Kosaku Nakano
Sophia Vögler
Kenji Tanaka
author_sort Kosaku Nakano
collection DOAJ
description The widespread adoption of electric vehicles (EVs) underscores the urgent need for innovative approaches to estimate their lithium-ion batteries’ state of health (SOH), which is crucial for ensuring safety and efficiency. This study introduces SOH-TEC, a transformer encoder-based model that processes raw time-series battery and vehicle-related data from a single EV trip to estimate the SOH. Unlike conventional methods that rely on lab-experimented battery cycle data, SOH-TEC utilizes real-world EV operation data, enhancing practical application. The model is trained and evaluated on a real-world dataset collected over nearly three years from three EVs. This dataset includes reliable SOH labels obtained through periodic constant-current full-discharge tests using a chassis dynamometer. Despite the challenges posed by noisy EV real-world data, the model shows high accuracy, with a mean absolute error of 0.72% and a root mean square error of 1.17%. Moreover, our proposed pre-training strategies with unlabeled data, particularly SOH ordinal comparison, significantly enhance the model’s performance; using only 50% of the labeled data achieves results nearly identical to those obtained with the full dataset. Self-attention map analysis reveals that the model primarily focuses on stationary or consistent driving periods to estimate SOH. While the study is constrained by a dataset featuring repetitive driving patterns, it highlights the significant potential of transformer for SOH estimation in EVs and offers valuable insights for future data collection and model development.
format Article
id doaj-art-353ccf4394314ef491143dcf48081ef2
institution OA Journals
issn 2666-7924
language English
publishDate 2024-12-01
publisher Elsevier
record_format Article
series Advances in Applied Energy
spelling doaj-art-353ccf4394314ef491143dcf48081ef22025-08-20T01:59:09ZengElsevierAdvances in Applied Energy2666-79242024-12-011610018810.1016/j.adapen.2024.100188Advancing state of health estimation for electric vehicles: Transformer-based approach leveraging real-world dataKosaku Nakano0Sophia Vögler1Kenji Tanaka2Graduate School of Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, 1138656, Tokyo, Japan; Corresponding authors.School of Engineering and Design, Technical University of Munich, Arcisstrasse 21, Munich, 80333, Bavaria, GermanyGraduate School of Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, 1138656, Tokyo, Japan; Corresponding authors.The widespread adoption of electric vehicles (EVs) underscores the urgent need for innovative approaches to estimate their lithium-ion batteries’ state of health (SOH), which is crucial for ensuring safety and efficiency. This study introduces SOH-TEC, a transformer encoder-based model that processes raw time-series battery and vehicle-related data from a single EV trip to estimate the SOH. Unlike conventional methods that rely on lab-experimented battery cycle data, SOH-TEC utilizes real-world EV operation data, enhancing practical application. The model is trained and evaluated on a real-world dataset collected over nearly three years from three EVs. This dataset includes reliable SOH labels obtained through periodic constant-current full-discharge tests using a chassis dynamometer. Despite the challenges posed by noisy EV real-world data, the model shows high accuracy, with a mean absolute error of 0.72% and a root mean square error of 1.17%. Moreover, our proposed pre-training strategies with unlabeled data, particularly SOH ordinal comparison, significantly enhance the model’s performance; using only 50% of the labeled data achieves results nearly identical to those obtained with the full dataset. Self-attention map analysis reveals that the model primarily focuses on stationary or consistent driving periods to estimate SOH. While the study is constrained by a dataset featuring repetitive driving patterns, it highlights the significant potential of transformer for SOH estimation in EVs and offers valuable insights for future data collection and model development.http://www.sciencedirect.com/science/article/pii/S266679242400026XLithium-ion batteryElectric vehicleState of health estimationDeep learningTransformer
spellingShingle Kosaku Nakano
Sophia Vögler
Kenji Tanaka
Advancing state of health estimation for electric vehicles: Transformer-based approach leveraging real-world data
Advances in Applied Energy
Lithium-ion battery
Electric vehicle
State of health estimation
Deep learning
Transformer
title Advancing state of health estimation for electric vehicles: Transformer-based approach leveraging real-world data
title_full Advancing state of health estimation for electric vehicles: Transformer-based approach leveraging real-world data
title_fullStr Advancing state of health estimation for electric vehicles: Transformer-based approach leveraging real-world data
title_full_unstemmed Advancing state of health estimation for electric vehicles: Transformer-based approach leveraging real-world data
title_short Advancing state of health estimation for electric vehicles: Transformer-based approach leveraging real-world data
title_sort advancing state of health estimation for electric vehicles transformer based approach leveraging real world data
topic Lithium-ion battery
Electric vehicle
State of health estimation
Deep learning
Transformer
url http://www.sciencedirect.com/science/article/pii/S266679242400026X
work_keys_str_mv AT kosakunakano advancingstateofhealthestimationforelectricvehiclestransformerbasedapproachleveragingrealworlddata
AT sophiavogler advancingstateofhealthestimationforelectricvehiclestransformerbasedapproachleveragingrealworlddata
AT kenjitanaka advancingstateofhealthestimationforelectricvehiclestransformerbasedapproachleveragingrealworlddata