Physics-Based Modeling and Parameter Identification for Lithium Ion Batteries Under High Current Discharge Conditions

Abstract

Extreme scenarios of high discharge current must be understood for better battery management system design. Physics-based modeling can give a better insight into the battery response but can be challenging due to the large number of parameters. In this work, an electrochemical pseudo-2D model is developed and used in the parameter identification and validated under high current discharge conditions. Commercial 18 650 cells with maximum rated current of 20 A (13.3 C) are characterized with discharge rates up to 40 C under controlled thermal conditions. The proposed three-step parameter identification procedure starts with the open circuit voltage being used to estimate the equilibrium potentials. In a second step, kinetic parameters are identified under high current aided by a parameter sensitivity analysis and parameter optimization with an evolutionary algorithm. The third step is the verification by comparing simulation results with measurements resulting in root main square error under 89 mV for currents until 26.6 C. Limits of the model are explored in the 33.3 C case, where a parameter re-fit shows that polarization effects change for very high current.