Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 28 of the 28 references it cites were found in a registry or an index.

A Novel Method for Estimating the State of Health of Lithium-Ion Batteries Based on Physics-Informed Neural Network

Open access
Deng, Yuxuan, Du, Changqing, Ren, Zhong · 2025 · Batteries

Abstract

An accurate state of health (SOH) assessment of lithium-ion batteries is essential for ensuring the reliability and safety of electric vehicles (EVs). Data-driven SOH estimation methods have shown promise but face challenges in generalizing across diverse battery types and variable operating conditions. To address this, this study integrates physical information into data-driven approaches, enabling physically consistent inferences and a rapid adaptation to different battery chemistries and usage scenarios. Specifically, physical features correlated with battery degradation, such as the link between incremental capacity (IC) peaks and SOH, are used as constraints to guide model learning. A fully connected layer within a back-propagation neural network (BPNN) is employed to capture battery aging dynamics effectively. Experimental results on two datasets show that the proposed model outperforms traditional neural networks, reducing the RMSE by at least 1.1% and demonstrating strong generalizability in both single-dataset and transfer learning tasks.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.3390/batteries11020049

Cites

28 of 28 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 The flowchart of the PINN training scheme for ba tt ery state-of-health estimation, illustrating data preprocessing, feature extraction, model training, and evaluation steps. Feature extraction was performed using the XJTU and Oxford datasets, and the training and test sets were divided according to a 6:4 ratio. Both the BPNN and PINN models were trained based on this division ratio and further adjusted using the loss function shown in the figure. Figure 1. The flowchart of the PINN training scheme for battery state-of-health estimation, illustrating data preprocessing, feature extraction, model training, and evaluation steps. Feature extraction was performed using the XJTU and Oxford datasets, and the training and test sets were divided according to a 6:4 ratio. Both the BPNN and PINN models were trained based on this division ratio and further adjusted using the loss function shown in the figure.
Figure 1 The flowchart of the PINN training scheme for ba tt ery state-of-health estimation, illustrating data preprocessing, feature extraction, model training, and evaluation steps. Feature extraction was performed using the XJTU and Oxford datasets, and the training and test sets were divided according to a 6:4 ratio. Both the BPNN and PINN models were trained based on this division ratio and further adjusted using the loss function shown in the figure. Figure 1. The flowchart of the PINN training scheme for battery state-of-health estimation, illustrating data preprocessing, feature extraction, model training, and evaluation steps. Feature extraction was performed using the XJTU and Oxford datasets, and the training and test sets were divided according to a 6:4 ratio. Both the BPNN and PINN models were trained based on this division ratio and further adjusted using the loss function shown in the figure.CC-BY · From the PDF
Figure 2 Oxford ba tt ery dataset capacity degradation. The aging experiment was conducted at 40 °C with constant-current charging (CC mode), and the discharging process was carried out under dynamic operating conditions. Further details are provided above. Figure 2. Oxford battery dataset capacity degradation. The aging experiment was conducted at 40 ◦ C with constant-current charging (CC mode), and the discharging process was carried out under dynamic operating conditions. Further details are provided above. lected, with the operating conditions involving constant-current charging and constantcurrent discharging. The degradation trajectories of these four cells are shown in Figure 3.
Figure 2 Oxford ba tt ery dataset capacity degradation. The aging experiment was conducted at 40 °C with constant-current charging (CC mode), and the discharging process was carried out under dynamic operating conditions. Further details are provided above. Figure 2. Oxford battery dataset capacity degradation. The aging experiment was conducted at 40 ◦ C with constant-current charging (CC mode), and the discharging process was carried out under dynamic operating conditions. Further details are provided above. lected, with the operating conditions involving constant-current charging and constantcurrent discharging. The degradation trajectories of these four cells are shown in Figure 3.CC-BY · From the PDF
Figure 3 XJTU ba tt ery dataset capacity degradation. The aging experiment was at the room temperature with constant-current charging (CC mode), and the discharging process was constant-curFigure 3. XJTU battery dataset capacity degradation. The aging experiment was at the room temperature with constant-current charging (CC mode), and the discharging process was constantcurrent discharging. Further details are provided above.
Figure 3 XJTU ba tt ery dataset capacity degradation. The aging experiment was at the room temperature with constant-current charging (CC mode), and the discharging process was constant-curFigure 3. XJTU battery dataset capacity degradation. The aging experiment was at the room temperature with constant-current charging (CC mode), and the discharging process was constantcurrent discharging. Further details are provided above.CC-BY · From the PDF
Figure 4 Charging curve of Oxford ba tt ery dataset, which was conducted at 40 °C with constantcurrent charging (CC mode). Figure 4. Charging curve of Oxford battery dataset, which was conducted at 40 ◦ C with constantcurrent charging (CC mode).
Figure 4 Charging curve of Oxford ba tt ery dataset, which was conducted at 40 °C with constantcurrent charging (CC mode). Figure 4. Charging curve of Oxford battery dataset, which was conducted at 40 ◦ C with constantcurrent charging (CC mode).CC-BY · From the PDF
Figure 5 Charging curve of XJTU ba tt ery dataset, which was at the room temperature with constantcurrent charging (CC mode). Figure 5. Charging curve of XJTU battery dataset, which was at the room temperature with constantcurrent charging (CC mode).
Figure 5 Charging curve of XJTU ba tt ery dataset, which was at the room temperature with constantcurrent charging (CC mode). Figure 5. Charging curve of XJTU battery dataset, which was at the room temperature with constantcurrent charging (CC mode).CC-BY · From the PDF
Figure 6 Discharging curve of XJTU ba tt ery dataset, which was carried out under constant current conditions. Figure 6. Discharging curve of XJTU battery dataset, which was carried out under constant current conditions.
Figure 6 Discharging curve of XJTU ba tt ery dataset, which was carried out under constant current conditions. Figure 6. Discharging curve of XJTU battery dataset, which was carried out under constant current conditions.CC-BY · From the PDF
Figure 7 Discharging curve of Oxford ba tt ery dataset, which was carried out under dynamic operating conditions. Figure 7. Discharging curve of Oxford battery dataset, which was carried out under dynamic operating conditions.
Figure 7 Discharging curve of Oxford ba tt ery dataset, which was carried out under dynamic operating conditions. Figure 7. Discharging curve of Oxford battery dataset, which was carried out under dynamic operating conditions.CC-BY · From the PDF
Figure 8 The IC curves of Oxford ba tt ery dataset. Figure 8. The IC curves of Oxford battery dataset. second peak with more obvious changes.
Figure 8 The IC curves of Oxford ba tt ery dataset. Figure 8. The IC curves of Oxford battery dataset. second peak with more obvious changes.CC-BY · From the PDF
Figure 9 The IC curves of XJTU ba tt ery dataset. Figure 9. The IC curves of XJTU battery dataset.
Figure 9 The IC curves of XJTU ba tt ery dataset. Figure 9. The IC curves of XJTU battery dataset.CC-BY · From the PDF
Image 10
CC-BY · From the PDF
Figure 11 The layer configuration and learning mechanism of the PINN model.
Figure 11 The layer configuration and learning mechanism of the PINN model.CC-BY · From the PDF
Figure 12 The performance of neural networks on the Oxford ba tt ery dataset. Figure 12. The performance of neural networks on the Oxford battery dataset. Figure 12. The performance of neural networks on the Oxford ba tt ery dataset.
Figure 12 The performance of neural networks on the Oxford ba tt ery dataset. Figure 12. The performance of neural networks on the Oxford battery dataset. Figure 12. The performance of neural networks on the Oxford ba tt ery dataset.CC-BY · From the PDF
Figure 13 The error of neural networks on the Oxford ba tt ery dataset. Figure 13. The error of neural networks on the Oxford ba tt ery dataset. Figure 13. The error of neural networks on the Oxford battery dataset.
Figure 13 The error of neural networks on the Oxford ba tt ery dataset. Figure 13. The error of neural networks on the Oxford ba tt ery dataset. Figure 13. The error of neural networks on the Oxford battery dataset.CC-BY · From the PDF
Figure 14 The performance of neural networks on the XJTU ba tt ery dataset. Figure 14. The performance of neural networks on the XJTU battery dataset.
Figure 14 The performance of neural networks on the XJTU ba tt ery dataset. Figure 14. The performance of neural networks on the XJTU battery dataset.CC-BY · From the PDF
Figure 15 The error of neural networks on the XJTU ba tt ery dataset. Figure 15. The error of neural networks on the XJTU battery dataset.
Figure 15 The error of neural networks on the XJTU ba tt ery dataset. Figure 15. The error of neural networks on the XJTU battery dataset.CC-BY · From the PDF

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 28 references

Research works printed but not linked: 2

  1. Christoph, R.B. (2017). Diagnosis and Prognosis of Degradation in Lithium-Ion Batteries, Department of Engineering Science University of Oxford.
  2. Birkl, C. (2017). Project-Physics-Informed Neural Network for Lithium-Ion Battery Degradation Stable Modeling and Prognosis. [Ph.D. Thesis, University of Oxford].

Cited by

See all 36

Metadata

Authors
Deng, Y., Du, C., & Ren, Z.
Affiliations
  • Wuhan University of Technology (CN)
  • Xihua University (CN)
Corresponding author
Du, Changqing
Venue
Batteries
Publisher
MDPI AG
Published
2025-01-26
Volume
11
Issue
2
Pages
49
ISSN
2313-0105 (ISSN-L)
OpenAlex
W4406866547
Cited by
32
References
28
Citations by year
2026: 23 · 2025: 9
FWCI
7.79
Type
journal article
Language
en
Open access
Locations
  • Batteries (publishedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
  • Key R&D project of Guangxi Province, China (AA23062058, XHRD2024-11233100-01)
  • Foshan Xianhu Laboratory of the Advanced Energy Science and Technology Guangdong Laboratory (AA23062058, XHRD2024-11233100-01)
Topics
  • Physical Sciences › Engineering › Automotive Engineering › Battery Modeling and Management
  • Physical Sciences › Engineering › Electrical and Electronic Engineering › Advancements in Battery Materials
  • Physical Sciences › Engineering › Automotive Engineering › Electric and Hybrid Vehicle Technologies
Keywords
lithium-ion batteries, state of health estimation, physics-informed neural networks, battery degradation, transfer learning, electric vehicles, data-driven battery modeling

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: A Novel Method for Estimating the State of Health of Lithium-Ion Batteries Based on Physics-Informed Neural Network. https://app.cochat.ai/papers/10.3390/batteries11020049
The paper
Deng, Y., Du, C., & Ren, Z. (2025). A Novel Method for Estimating the State of Health of Lithium-Ion Batteries Based on Physics-Informed Neural Network. Batteries. https://doi.org/10.3390/batteries11020049

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.