Scholarlink

<p>Accurate estimation of the State of Health (SOH) is critical for ensuring the reliability and safety of lithium-ion battery management systems. While traditional power-law models are widely used, they often struggle to capture multi-stage degradation characteristics and accelerated aging behavior under varying thermal conditions. This study proposes a temperature-dependent double-exponential model (TDM) for SOH estimation. Unlike single-exponential or power-law approaches, the dual-exponential structure is able to describe both the stable solid electrolyte interphase (SEI) growth phase and the accelerated degradation stage associated with active material loss. To account for thermal effects, an Arrhenius-based temperature compensation term is incorporated into the exponential decay rates. A hierarchical two-stage identification algorithm is then developed, in which the linear parameters are updated by least squares and the nonlinear parameters are estimated by gradient-based iterations. Experimental results show that the proposed TDM improves fitting and prediction accuracy compared with conventional models in the considered validation cases. On the real NASA battery aging dataset, the proposed TDM achieves an RMSE of 0.0170, an MAE of 0.0140, and a MAPE of <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(1.95\%\)</EquationSource> </InlineEquation>.</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

Parameter estimation of a temperature-dependent double-exponential SOH model for battery management systems

  • 期刊论文

  • 发表日期：

    2026年6月23日

查看全文
  • Dongqi Han,
  • Jin Li

摘要

Accurate estimation of the State of Health (SOH) is critical for ensuring the reliability and safety of lithium-ion battery management systems. While traditional power-law models are widely used, they often struggle to capture multi-stage degradation characteristics and accelerated aging behavior under varying thermal conditions. This study proposes a temperature-dependent double-exponential model (TDM) for SOH estimation. Unlike single-exponential or power-law approaches, the dual-exponential structure is able to describe both the stable solid electrolyte interphase (SEI) growth phase and the accelerated degradation stage associated with active material loss. To account for thermal effects, an Arrhenius-based temperature compensation term is incorporated into the exponential decay rates. A hierarchical two-stage identification algorithm is then developed, in which the linear parameters are updated by least squares and the nonlinear parameters are estimated by gradient-based iterations. Experimental results show that the proposed TDM improves fitting and prediction accuracy compared with conventional models in the considered validation cases. On the real NASA battery aging dataset, the proposed TDM achieves an RMSE of 0.0170, an MAE of 0.0140, and a MAPE of \(1.95\%\) .