Digital Twin Model for Lithium-Ion Battery SOC Estimation in Battery Swapping Station

As electric vehicles rapidly gain popularity, battery swapping stations have emerged as key infrastructure to enhance the convenience of electric vehicles. Accurately estimating the State of Charge (SOC) of batteries during the battery swapping process is critical for ensuring efficient operation an...

Full description

Saved in:
Bibliographic Details
Main Authors: Ruopeng Wu, Ying Shi, Qin Hu, Changjun Xie, Jicheng Yu, Boyang Ma, Shengzhong Liu
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10741523/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846163739162705920
author Ruopeng Wu
Ying Shi
Qin Hu
Changjun Xie
Jicheng Yu
Boyang Ma
Shengzhong Liu
author_facet Ruopeng Wu
Ying Shi
Qin Hu
Changjun Xie
Jicheng Yu
Boyang Ma
Shengzhong Liu
author_sort Ruopeng Wu
collection DOAJ
description As electric vehicles rapidly gain popularity, battery swapping stations have emerged as key infrastructure to enhance the convenience of electric vehicles. Accurately estimating the State of Charge (SOC) of batteries during the battery swapping process is critical for ensuring efficient operation and optimizing battery management. This paper proposes a digital twin framework and establishes a data-driven model for SOC estimation. The model, Ensemble Weighted Network (EWN), is embedded in the digital twin framework. It has three remarkable processes: multi-model ensemble, accuracy weight scaling, and time weight scaling. The multi-model ensemble combines five popular SOC estimating learners, to avoid the accuracy limitations of a single estimation model. The accuracy weight scaling and time weight scaling strategies are put forward to assign appropriate weights to each model, addressing significant fluctuations in estimation capability among different models across various periods. The results show the root mean square error, mean absolute error, and mean absolute percentage error of the model are less than 1.10%, 0.96%, and 1.92% respectively. Furthermore, we conduct experiments on public datasets to prove the proposed model has high estimation accuracy and robustness for different temperatures and battery types. Finally, the visualization of the digital twin is built, allowing operators to monitor and manage the batteries.
format Article
id doaj-art-13d72cb92c0f41048be42e4d8da9c33d
institution Kabale University
issn 2169-3536
language English
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-13d72cb92c0f41048be42e4d8da9c33d2024-11-19T00:03:05ZengIEEEIEEE Access2169-35362024-01-011216566316567710.1109/ACCESS.2024.349065510741523Digital Twin Model for Lithium-Ion Battery SOC Estimation in Battery Swapping StationRuopeng Wu0Ying Shi1https://orcid.org/0000-0003-3519-2308Qin Hu2https://orcid.org/0000-0002-1571-292XChangjun Xie3https://orcid.org/0000-0002-9626-0813Jicheng Yu4https://orcid.org/0000-0002-4525-2987Boyang Ma5Shengzhong Liu6School of Automation, Wuhan University of Technology, Wuhan, ChinaSchool of Automation, Wuhan University of Technology, Wuhan, ChinaSchool of Automation, Wuhan University of Technology, Wuhan, ChinaSchool of Automation, Wuhan University of Technology, Wuhan, ChinaChina Electric Power Research Institute, Wuhan, ChinaState Grid Tianjin Electric Power Company, Tianjin, ChinaState Grid Tianjin Electric Power Company, Tianjin, ChinaAs electric vehicles rapidly gain popularity, battery swapping stations have emerged as key infrastructure to enhance the convenience of electric vehicles. Accurately estimating the State of Charge (SOC) of batteries during the battery swapping process is critical for ensuring efficient operation and optimizing battery management. This paper proposes a digital twin framework and establishes a data-driven model for SOC estimation. The model, Ensemble Weighted Network (EWN), is embedded in the digital twin framework. It has three remarkable processes: multi-model ensemble, accuracy weight scaling, and time weight scaling. The multi-model ensemble combines five popular SOC estimating learners, to avoid the accuracy limitations of a single estimation model. The accuracy weight scaling and time weight scaling strategies are put forward to assign appropriate weights to each model, addressing significant fluctuations in estimation capability among different models across various periods. The results show the root mean square error, mean absolute error, and mean absolute percentage error of the model are less than 1.10%, 0.96%, and 1.92% respectively. Furthermore, we conduct experiments on public datasets to prove the proposed model has high estimation accuracy and robustness for different temperatures and battery types. Finally, the visualization of the digital twin is built, allowing operators to monitor and manage the batteries.https://ieeexplore.ieee.org/document/10741523/Battery swappinglithium-ion batterystate of chargedigital twinensemble model
spellingShingle Ruopeng Wu
Ying Shi
Qin Hu
Changjun Xie
Jicheng Yu
Boyang Ma
Shengzhong Liu
Digital Twin Model for Lithium-Ion Battery SOC Estimation in Battery Swapping Station
IEEE Access
Battery swapping
lithium-ion battery
state of charge
digital twin
ensemble model
title Digital Twin Model for Lithium-Ion Battery SOC Estimation in Battery Swapping Station
title_full Digital Twin Model for Lithium-Ion Battery SOC Estimation in Battery Swapping Station
title_fullStr Digital Twin Model for Lithium-Ion Battery SOC Estimation in Battery Swapping Station
title_full_unstemmed Digital Twin Model for Lithium-Ion Battery SOC Estimation in Battery Swapping Station
title_short Digital Twin Model for Lithium-Ion Battery SOC Estimation in Battery Swapping Station
title_sort digital twin model for lithium ion battery soc estimation in battery swapping station
topic Battery swapping
lithium-ion battery
state of charge
digital twin
ensemble model
url https://ieeexplore.ieee.org/document/10741523/
work_keys_str_mv AT ruopengwu digitaltwinmodelforlithiumionbatterysocestimationinbatteryswappingstation
AT yingshi digitaltwinmodelforlithiumionbatterysocestimationinbatteryswappingstation
AT qinhu digitaltwinmodelforlithiumionbatterysocestimationinbatteryswappingstation
AT changjunxie digitaltwinmodelforlithiumionbatterysocestimationinbatteryswappingstation
AT jichengyu digitaltwinmodelforlithiumionbatterysocestimationinbatteryswappingstation
AT boyangma digitaltwinmodelforlithiumionbatterysocestimationinbatteryswappingstation
AT shengzhongliu digitaltwinmodelforlithiumionbatterysocestimationinbatteryswappingstation