Scholarlink

Accurate modeling of Li-ion batteries is essential for optimizing performance and safety. Given the complexity of electrochemical models, parameter identification is challenging. This study presents a sensitivity-driven optimization framework designed to address uncertainties and improve the accuracy of the model. The current research introduces an innovative method for parameterizing the Newman pseudo two dimensional (P2D) model, involving two main steps. First, a sensitivity analysis was conducted using the Adaptive Metamodel of Optimal Prognosis technique to enhance model accuracy and significantly reduce computational time. Second, unknown and uncertain parameters were optimized using the non-linear programming using the Quadratic Lagrangian (NLPQL) approach. The optimized model strongly agreed with experimental data for three discharge rates (0.1 C, 0.5 C, 1 C) at 25 \(^{\circ }\) C, achieving a root mean square error (RMSE) of 28 mV in voltage and 0.61 K in temperature at 1 C rate.

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

Navigating the Unknown and Uncertain: A Sensitivity-Driven Optimisation Framework for High-Fidelity Battery Modeling

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Pragya Tripathi,
  • Rishi Relan

摘要

Accurate modeling of Li-ion batteries is essential for optimizing performance and safety. Given the complexity of electrochemical models, parameter identification is challenging. This study presents a sensitivity-driven optimization framework designed to address uncertainties and improve the accuracy of the model. The current research introduces an innovative method for parameterizing the Newman pseudo two dimensional (P2D) model, involving two main steps. First, a sensitivity analysis was conducted using the Adaptive Metamodel of Optimal Prognosis technique to enhance model accuracy and significantly reduce computational time. Second, unknown and uncertain parameters were optimized using the non-linear programming using the Quadratic Lagrangian (NLPQL) approach. The optimized model strongly agreed with experimental data for three discharge rates (0.1 C, 0.5 C, 1 C) at 25 \(^{\circ }\) C, achieving a root mean square error (RMSE) of 28 mV in voltage and 0.61 K in temperature at 1 C rate.