COBRAPRO: A MATLAB toolbox for Physics-based Battery Modeling and Co-simulation Parameter Optimization

Abstract

COBRAPRO (Co-simulation Battery Modeling for Accelerated Parameter Optimization) is a physics-based battery modeling software with the capability to perform closed-loop parameter optimization using experimental data. COBRAPRO is based on the Doyle-Fuller-Newman (DFN) model (Doyle et al., 1993), which is the most widely accepted high-fidelity model that considers the lithium-ion transport and charge conservation in the liquid electrolyte and solid electrodes, and kinetics at the solid and liquid interface during lithium intercalation and deintercalation. Such physics-based models have found applications in battery design (Couto et al., 2023; Dai & Srinivasan, 2016) and advanced battery management systems to ensure reliable and safe operation of electric vehicles (Kolluri et al., 2020). The DFN model relies on physical parameters such as geometric, stoichiometric, concentration, transport, and kinetic properties. Accurate parameter values are essential for the model to predict real battery behavior under various usage conditions. However, parameters obtained from cell teardown experiments cannot be directly applied (Ha & Onori, 2024; Li et al., 2022), as the DFN model simplifies a real battery, assuming perfectly spherical particles, neglecting electrode heterogeneity, and considering internal dynamics only in one dimension. Parameter identification is therefore crucial for developing models that accurately capture battery dynamics. COBRAPRO enables users to identify these parameters for any battery using standard current-voltage data from a battery cycler, optimizing DFN parameters by minimizing the error between simulated and experimental results through an integrated optimization routine.