Open Access

ARTICLE

Research on a Model-Based and Data-Driven Fusion Estimation Method for Lithium Battery SOP

Qi Wang^1,2,3, Tao Zhu^1,*, Yibo Huang¹
1 School of Electrical and Information Engineering, Jiangsu University of Technology, Changzhou, China
2 Research and Development Department, Shuangdeng Group Co., Ltd., Taizhou, China
3 School of Automotive and Traffic Engineering, Jiangsu University, Zhenjiang, China
* Corresponding Author: Tao Zhu. Email:
(This article belongs to the Special Issue: New Energy and Energy Storage System)

Energy Engineering https://doi.org/10.32604/ee.2026.078633

Received 05 January 2026; Accepted 28 February 2026; Published online 24 March 2026

Abstract

Accurate SOP estimation is essential for energy management and safety control of lithium-ion batteries in new energy vehicles under complex operating conditions. However, strong nonlinearity and time-varying characteristics still limit the accuracy and robustness of conventional approaches. To address this issue, this paper proposes a model-based and data-driven fusion strategy for SOP estimation. A physics-based battery model is employed to describe the dynamic behavior of the battery, and an SOP estimation framework is constructed by jointly considering multiple practical constraints, including terminal voltage, SOC, and the intrinsic current limit, so as to enable online evaluation of available battery power. On this basis, a data-driven correction module is introduced to compensate for residual modeling errors and to capture nonlinear characteristics that are difficult to represent by the model alone. In this way, the proposed method combines the interpretability of model-based estimation with the flexibility of data-driven learning, achieving a balanced trade-off between accuracy and robustness. The proposed fusion approach is validated under the DST profile. The experimental results show that the method can achieve high estimation accuracy and good robustness under dynamic operating conditions, with the absolute SOP estimation error kept within 0.1 W. These results demonstrate that the proposed fusion framework provides a reliable and practical solution for SOP estimation in battery management systems.

---

Keywords

Lithium battery; SOP estimation; BP neural network; IBES optimization algorithm

---