Advanced machine learning techniques for State-of-Health estimation in lithium-ion batteries: A comparative study

Simple item page
dc.contributor.authorSedlařík, Marekcs
dc.contributor.authorVyroubal, Petrcs
dc.contributor.authorCapková, Dominikacs
dc.contributor.authorOmerdic, Edincs
dc.contributor.authorRae, Mitchellcs
dc.contributor.authorMačák, Martincs
dc.contributor.authorŠedina, Martincs
dc.contributor.authorKazda, Tomášcs
dc.coverage.issue6cs
dc.coverage.volume524cs
dc.date.accessioned2025-07-17T08:59:23Z
dc.date.available2025-07-17T08:59:23Z
dc.date.issued2025-06-01cs
dc.description.abstractThe accurate modeling and prediction of the State-of-Health (SOH) of lithium-ion (Li-ion) batteries are crucial for extending their lifespan, ensuring reliability, and minimizing the costs associated with extensive laboratory testing. This paper investigates the SOH estimation of Li-ion batteries utilizing advanced machine learning (ML) techniques. Specifically, 600 cycles were performed on Samsung INR18650–35E cells using the Constant Current Constant Voltage (CCCV) protocol. The input data for the ML methods were extracted from both charging and discharging cycles to achieve the best possible results. Data-driven models with different methodological foundations were used to predict SOH: Gaussian Process Regression (GPR), Support Vector Regression (SVR), and from the field of Artificial Neural Networks (ANN), Feed-Forward Neural Network (FFNN) and Adaptive Neuro-Fuzzy Inference System (ANFIS), which utilizes fuzzy logic. The input features for the ML methods were analyzed using Pearson Correlation Analysis (PCA), and additional inputs for the ANFIS method were selected using Exhaustive Search (ES) to identify the optimal combination of inputs with the lowest Root Mean Square Error (RMSE). The individual ML methods were evaluated on datasets of various sizes using the features with the highest correlation to SOH and the full set of features to detect overfitting. Further experiments explored the dependency of RMSE on the amount of training data, and SOH estimation of one battery was performed using training data from another. Overall, experiments show that nearly all methods achieved RMSE below 0.5% for SOH estimation, with SVR proving the most stable technique and ANFIS excelling with meticulously optimized configurations.en
dc.formattextcs
dc.format.extent1-22cs
dc.format.mimetypeapplication/pdfcs
dc.identifier.citationELECTROCHIMICA ACTA. 2025, vol. 524, issue 6, p. 1-22.en
dc.identifier.doi10.1016/j.electacta.2025.145988cs
dc.identifier.issn1873-3859cs
dc.identifier.orcid0000-0003-4160-8031cs
dc.identifier.orcid0000-0002-3614-5629cs
dc.identifier.orcid0000-0002-2830-4888cs
dc.identifier.orcid0000-0002-2944-9499cs
dc.identifier.orcid0000-0003-1973-0292cs
dc.identifier.other198149cs
dc.identifier.researcheridE-8103-2018cs
dc.identifier.researcheridAAX-5947-2020cs
dc.identifier.scopus56376348100cs
dc.identifier.scopus57201313573cs
dc.identifier.scopus56574103900cs
dc.identifier.urihttps://hdl.handle.net/11012/255182
dc.language.isoencs
dc.publisherElseviercs
dc.relation.ispartofELECTROCHIMICA ACTAcs
dc.relation.urihttps://www.sciencedirect.com/science/article/pii/S0013468625003512cs
dc.rightsCreative Commons Attribution 4.0 Internationalcs
dc.rights.accessopenAccesscs
dc.rights.sherpahttp://www.sherpa.ac.uk/romeo/issn/1873-3859/cs
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/cs
dc.subjectState-of-HealthLi-ion batteryMachine learningSupport vector regressionGaussian process regressionFeed-forward neural networkAdaptive neuro-fuzzy inference systemen
dc.titleAdvanced machine learning techniques for State-of-Health estimation in lithium-ion batteries: A comparative studyen
dc.type.driverarticleen
dc.type.statusPeer-revieweden
dc.type.versionpublishedVersionen
eprints.grantNumberinfo:eu-repo/grantAgreement/MSM/EH/EH22_008/0004617cs
sync.item.dbidVAV-198149en
sync.item.dbtypeVAVen
sync.item.insts2025.07.17 10:59:22en
sync.item.modts2025.07.17 10:34:09en
thesis.grantorVysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií. Ústav elektrotechnologiecs
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
1s2.0S0013468625003512main.pdf
Size:
9.61 MB
Format:
Adobe Portable Document Format
Description:
file 1s2.0S0013468625003512main.pdf
Download
Collections
Ústav elektrotechnologie