Physics-informed hierarchical perception modulation network for lithium-ion battery health management

Shuai Hao, Jirou Feng, Jinrun Dong, Wenyue Cui, Jinhua Cheng and Maoguo Gong

Energy, 2025, vol. 335, issue C

Abstract: The rapid proliferation of electric vehicles has underscored the critical role of lithium-ion batteries in ensuring energy efficiency and operational safety. Accurate state of health (SOH) estimation of battery, remains a significant challenge due to the complexity and variability of real-world usage conditions. To address this issue, a novel physics-informed hierarchical perception modulation network is proposed to model battery degradation dynamics and predict SOH with enhanced accuracy and robustness. The proposed framework employs a convolutional neural network to capture local fine-grained electrochemical signatures. These local representations are subsequently modulated by degradation-aware parameters derived from a Transformer-based module, which models long-range temporal dependencies and encapsulates the global aging trajectory. This hierarchical modulation mechanism enables dynamic balancing between short-term micro-patterns and long-term degradation trends, thus enhancing the model’s resilience under varying operating conditions. Furthermore, a physics-informed loss function is introduced, derived from a discretized state-space degradation equation, to incorporate domain-specific knowledge and regularize the learning process. Experimental evaluations conducted on benchmark LiFePO4 battery datasets from MIT and HUST demonstrate the superior performance of the proposed method, achieving higher predictive accuracy and stronger generalization compared to state-of-the-art approaches.

Keywords: Lithium-ion battery; State of health; Physics-informed neural network; Deep learning (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S036054422503871X
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:335:y:2025:i:c:s036054422503871x

DOI: 10.1016/j.energy.2025.138229

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 ().