Scholarlink

<p>Accurate estimation of the state of health (SOH) of lithium-ion batteries (LIBs) is essential for the safety of energy storage devices. Electrochemical impedance spectroscopy (EIS), which measures kinetic information in LIBs, is a powerful tool for explaining the degradation mechanisms of these batteries. To understand the internal state of coupled-measured LIBs, the distribution of relaxation times (DRT) method was first employed to reconstruct the EIS data and extract valuable health factors (HFs). Next, correlation analysis was used to identify the most pertinent degradation mode of LIBs at different states of charge (SOCs) and temperatures. Finally, four training test cases were designed using recurrent neural network to estimate SOH on LiCoO<sub>2</sub> battery datasets at 25 °C, 35 °C, and 45 °C. The model demonstrated a root mean square error (RMSE) of 0.67% and a mean absolute error (MAE) of 0.41%, while also exhibiting greater robustness and requiring less training time. This work demonstrates the excellent performance of time-domain information from AC impedance, offering a promising solution for battery SOH estimation.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

On advanced state-of-health assessment for lithium-ion batteries of degradation mode identification with time-domain AC impedance analysis

  • 期刊论文

  • 发表日期：

    2025年4月24日

查看全文
  • Minlei Xia,
  • Peng Ding,
  • Xuejuan Wang,
  • Qingwei Gao,
  • Hui Pan,
  • Wenyao Guo,
  • Penghui Shi,
  • Yulin Min

摘要

Accurate estimation of the state of health (SOH) of lithium-ion batteries (LIBs) is essential for the safety of energy storage devices. Electrochemical impedance spectroscopy (EIS), which measures kinetic information in LIBs, is a powerful tool for explaining the degradation mechanisms of these batteries. To understand the internal state of coupled-measured LIBs, the distribution of relaxation times (DRT) method was first employed to reconstruct the EIS data and extract valuable health factors (HFs). Next, correlation analysis was used to identify the most pertinent degradation mode of LIBs at different states of charge (SOCs) and temperatures. Finally, four training test cases were designed using recurrent neural network to estimate SOH on LiCoO2 battery datasets at 25 °C, 35 °C, and 45 °C. The model demonstrated a root mean square error (RMSE) of 0.67% and a mean absolute error (MAE) of 0.41%, while also exhibiting greater robustness and requiring less training time. This work demonstrates the excellent performance of time-domain information from AC impedance, offering a promising solution for battery SOH estimation.