SoH Estimation of ASPILSAN INR18650A28 Cells Using Real-World Factory Data

DOI

10.2991/978-94-6239-668-5_43How to use a DOI?

Keywords

SoH Estimation; Machine Learning Algorithms; Regression; Lithium Ion Battery

Abstract

The estimation of the State of Health (SoH) of lithium-ion batteries is essential for ensuring the reliability and lifetime optimization of battery management systems (BMS). In this study, the long-term aging behavior of nickel-rich L i N i M n C o O 2 (NMC) INR18650A28 cells was analyzed using real-world production data from ASPILSAN Energy. Unlike previous works based on laboratory datasets, this study utilizes factory-origin, field-realistic data. Aging tests were performed with a WONIK PNE CC05–15 battery cycler at 23 ℃ under a constant current/constant voltage (CC/CV) protocol. The cells, with a nominal capacity of 2800 mAh, were charged at 0.5C (1.4 A) to 4.2 V and discharged at 1C (2.8 A) to 1.5 V for 1200 cycles. The raw voltage–capacity data were filtered and used for differential capacity ( d C / d V ) analysis, from which health-related electrochemical features were extracted as model inputs. These features and corresponding SoH values were employed to train regression-based models in MATLAB’s Regression Learner. Linear Regression, Robust Linear Regression, and Bagged Trees achieved the highest prediction accuracy. The Linear Regression model performed best with RMSE = 0.325 and MAE = 0.250, followed by the Robust model (RMSE = 0.378, MAE = 0.269) and Bagged Trees (RMSE ≈ 6.52, MAE ≈ 1.99). Results demonstrate that d C / d V -based features effectively represent battery degradation, enabling accurate SoH prediction with regression algorithms.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Nisanur Yildiran
AU  - Teoman Karadağ
PY  - 2026
DA  - 2026/05/14
TI  - SoH Estimation of ASPILSAN INR18650A28 Cells Using Real-World Factory Data
BT  - Proceedings of the International Conference on Current Problems in Engineering and Applied Sciences (ICCPEAS 2025)
PB  - Atlantis Press
SP  - 406
EP  - 417
SN  - 2352-5401
UR  - https://doi.org/10.2991/978-94-6239-668-5_43
DO  - 10.2991/978-94-6239-668-5_43
ID  - Yildiran2026
ER  -