Pseudo-Steady-State Reduced-Order-Model Approximation for Constant-Current Parameter Identification in Lithium-Ion Cells

The challenges of parameter identification for a lumped-parameter, physics-based model of a lithium-ion cell motivate a closed-form approximation that can be used inside an optimization routine. Present reduced-order models of the lithium-ion cell do not achieve the desired speed and fidelity for the parameter-identification application when applied to a constant-current test. This paper introduces a novel approximation to the cell internal and terminal-voltage dynamics that is specialized for constant-current applications and incorporates a model of solid and electrolyte diffusion, solid and electrolyte potential, and the kinetics of the solid-electrolyte interphase layer. The approximation leverages non-time-varying profiles for electrolyte-level quantities under a pseudo-steady-state assumption coupled with a nonlinear approximation to the lithium stoichiometry at the electrode surface. The proposed approximation achieves significantly improved speed and accuracy over a comparable reduced-order model simplified for constant current when evaluated against a full-order-model simulation using the true parameter values.