A two-step identification approach for an extended nonlinear double-capacitor model

Area of Science:

• Electrochemistry
• Computational Modeling
• Automotive Engineering

Background:

• Equivalent circuit models are standard for Li-ion cells in automotive applications.
• Existing models struggle with accurate discharge capacity prediction across diverse loads, dynamic tests, and parameterization.

Purpose of the Study:

• To propose an extended nonlinear double capacitor model for Li-ion cells.
• To improve accuracy, computational speed, and parameterization ease for automotive applications.

Main Methods:

• Increased model order and C-rate dependent parameters.
• Developed an identification procedure exploiting pseudo-linear problem characteristics.
• Utilized an analogy with the single particle model to reduce search space and enhance interpretability.

Main Results:

• The extended model accurately captures cell discharge capacity under various loads.
• Achieved a mean absolute average error of approximately 20 mV on LiFePO4 datasets and a realistic drive cycle.
• Demonstrated superior performance compared to a state-of-the-art model.

Conclusions:

• The proposed model extension effectively addresses limitations of traditional equivalent circuit models.
• Offers a robust and interpretable solution for Li-ion cell modeling in dynamic automotive environments.