Lithium-ion battery SOH prediction based on multi-dimensional features and multi-model feature selector

Hao Li and Chao Chen

Energy, 2025, vol. 331, issue C

Abstract: With the extensive application of lithium-ion batteries in electronic devices and electric vehicles, accurately predicting their state of health (SOH) has become increasingly critical. To address this, we propose an SOH prediction model based on multidimensional feature extraction and a Multi-Model Feature Selector (MMFS). The model extracts key features from multiple stages and dimensions of battery operation cycles and enhances prediction performance through a comprehensive feature selection approach. In terms of feature extraction, a hierarchical multidimensional feature extraction method is employed to capture the dynamic changes in battery performance comprehensively. This method includes basic characterization features, second-order differential curve features, and third-order features such as distance and shape characteristics during the charging process. For feature selection, MMFS systematically evaluates the importance of features by integrating multi-model scoring and selecting the most valuable features. Experimental results demonstrate that the MMFS-based model achieves outstanding performance in key metrics such as root mean squared error (RMSE) and the coefficient of determination (R2). Notably, it exhibits significant advantages in complex models such as eXtreme Gradient Boosting (XGBoost) and Categorical Boosting (CatBoost), achieving a maximum RMSE of 0.1373%.

Keywords: Lithium-ion battery; SOH prediction; Machine learning; Feature engineering (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0360544225024867
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:energy:v:331:y:2025:i:c:s0360544225024867

DOI: 10.1016/j.energy.2025.136844

Access Statistics for this article

Energy is currently edited by Henrik Lund and Mark J. Kaiser

More articles in Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().