EUVIMEDEUVIMED entdecken
Crossref

Uncertainty-Compensated SOH Estimation for Lithium-Ion Batteries via Collaborative Bi-Ensemble Feature Selection and Dual-Channel Hybrid Architecture

Yuxin Cui, Zhao Zhang, Yunxiao Shan, Siqi Wang, Jinlian Liang, Di Zhou, Shu Li, Minghua Chen

IEEE Transactions on Instrumentation and Measurement · 2026 · 75 · 2512913-2512913

DOI: 10.1109/tim.2026.3704237 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Instrumentation and Measurement, 2026. Verfasst von Yuxin Cui und 7 weiteren Autor:innen. Die Arbeit nennt 60 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of Heilongjiang Province of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurately assessing the state of health (SOH) of batteries is crucial for optimizing their management systems and ensuring their safe operation. However, the inherent complexity and diversity of health features (HFs) introduce uncertainty into existing SOH estimation methods. Consequently, selecting an accurate and adaptive HF combination is critical for SOH estimation. This article proposes an uncertainty-compensated SOH estimation framework that integrates a collaborative Bi-Ensemble feature selection (CBFS) strategy with a dual-channel hybrid model, thereby enhancing its applicability across diverse battery datasets. First, an extensive collection of HFs is conducted, followed by the proposed CBFS strategy to screen for optimal HF combinations. This model-independent strategy compensates for uncertainty by identifying key HFs reflecting battery degradation trends, utilizing solely the intrinsic characteristics of the dataset. Second, a dual-channel deep hybrid model is developed, with hyperparameters optimized by the whale optimization algorithm (WOA). Finally, comprehensive experiments are carried out on the Oxford, NASA, MIT, and self-constructed battery datasets to explore the universality of the framework under different materials and complex working conditions. The results demonstrate that the uncertainty-compensated framework exhibits stronger universality and higher SOH prediction accuracy, offering a new and effective approach for preventive maintenance decision-making in battery health management.

Förderung

  • National Natural Science Foundation of China
  • Natural Science Foundation of Heilongjiang Province of China

Zitieren

APA

Cui, Y., Zhang, Z., Shan, Y., Wang, S., Liang, J., Zhou, D., Li, S., & Chen, M. (2026). Uncertainty-Compensated SOH Estimation for Lithium-Ion Batteries via Collaborative Bi-Ensemble Feature Selection and Dual-Channel Hybrid Architecture. IEEE Transactions on Instrumentation and Measurement, 75, 2512913-2512913. https://doi.org/10.1109/tim.2026.3704237

Vancouver

Cui Y, Zhang Z, Shan Y, Wang S, Liang J, Zhou D, et al. Uncertainty-Compensated SOH Estimation for Lithium-Ion Batteries via Collaborative Bi-Ensemble Feature Selection and Dual-Channel Hybrid Architecture. IEEE Transactions on Instrumentation and Measurement. 2026;75:2512913-2512913. doi:10.1109/tim.2026.3704237

BibTeX
@article{Cui2026,
  author = {Cui, Yuxin and Zhang, Zhao and Shan, Yunxiao and Wang, Siqi and Liang, Jinlian and Zhou, Di and Li, Shu and Chen, Minghua},
  title = {Uncertainty-Compensated SOH Estimation for Lithium-Ion Batteries via Collaborative Bi-Ensemble Feature Selection and Dual-Channel Hybrid Architecture},
  journal = {IEEE Transactions on Instrumentation and Measurement},
  year = {2026},
  volume = {75},
  pages = {2512913--2512913},
  doi = {10.1109/tim.2026.3704237}
}
Kostenlos starten →