点击次数：

影响因子：5.6

DOI码：10.1016/j.measurement.2026.121578

发表刊物：Measurement

关键字：Lithium-ion battery; State of charge-state of health; Collaborative estimation; Hybrid network; SOC changing threshold

摘要：The low accuracy and weak robustness of the state of charge-state of health (SOC-SOH) collaborative estimation for onboard lithium-ion batteries lead to fire and explosion of electric vehicles. However, the internal electrochemical reaction of lithium-ion batteries is complex, and the external application scenarios are random and changeable, making SOC-SOH collaborative estimation challenging. This paper proposes a SOC-SOH collaborative estimation framework based on the hybrid network and the SOC changing threshold using real-world data. Firstly, a Convolutional Neural Network-Gated Recurrent Unit-Attention (CNN-GRU-Attention) hybrid network for SOC estimation is constructed. Then, a SOC changing threshold concept is introduced for online capacity estimation using discharging data, avoiding the difficulty of obtaining some features when using charging data for SOH estimation, due to the random charging start point and the charging end point. Finally, the capacity convergence correction and the temperature correction methods are developed to achieve accurate and robust SOC-SOH collaborative estimation results. The experimental results indicate that the proposed framework can reach ideal SOC-SOH collaborative estimation results in real-world applications, including the northern and southern regions of China. The Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) of SOC estimation results are all below 0.95% and 1.10%, respectively. The MAE and RMSE of SOH estimation results are all below 0.95% and 1.05%, respectively.

备注：中科院2区Top

合写作者：Sihan Cheng,Siyi Liu,Yuan Chen,Heng Li

第一作者：Muyao Wu

论文类型：期刊论文

通讯作者：Li Wang

论文编号：121578

学科门类：工学

文献类型：J

ISSN号：0263-2241

是否译文：否

发表时间：2026-04-26

收录刊物：SCI、EI