Scholarlink

Lithium-ion batteries are widely used in electric vehicles and aerospace due to their excellent characteristics. However, the accurate estimation of their State of Health (SOH) is difficult. Existing methods have limitations such as high cost or poor generalizability. This paper proposes an SOH estimation method combining Electrochemical Impedance Spectroscopy (EIS), SHapley Additive exPlanations (SHAP), and Gaussian Process Regression (GPR). SHAP selects key EIS feature frequencies, and GPR estimates SOH efficiently. Results show that with a training - test ratio of 8:2, the maximum error is 3.084% and the Mean Absolute Error is 0.364%, indicating that this method can accurately estimate SOH.

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

Electrochemical Impedance Spectroscopy-Based SOH Estimation of Lithium-Ion Batteries Using SHAP and GPR

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Yang Zhou,
  • Jinlei Sun,
  • Xuliang Zhang,
  • Xiangrui Meng,
  • Huiying Ling,
  • Ruoyu Wang,
  • Yuhao Wu,
  • Shuhang Wang

摘要

Lithium-ion batteries are widely used in electric vehicles and aerospace due to their excellent characteristics. However, the accurate estimation of their State of Health (SOH) is difficult. Existing methods have limitations such as high cost or poor generalizability. This paper proposes an SOH estimation method combining Electrochemical Impedance Spectroscopy (EIS), SHapley Additive exPlanations (SHAP), and Gaussian Process Regression (GPR). SHAP selects key EIS feature frequencies, and GPR estimates SOH efficiently. Results show that with a training - test ratio of 8:2, the maximum error is 3.084% and the Mean Absolute Error is 0.364%, indicating that this method can accurately estimate SOH.