Identifiability Study of Lithium-Ion Battery Capacity Fade Using Degradation Mode Sensitivity for a Minimally and Intuitively Parametrized Electrode-Specific Cell Open-Circuit Voltage Model

Abstract

When two electrode open-circuit potentials form a full-cell OCV (open-circuit voltage) model, cell-level SOH (state of health) parameters related to LLI (loss of lithium inventory) and LAM (loss of active materials) naturally appear. Such models have been used to interpret experimental OCV measurements and infer these SOH parameters associated with capacity fade. In this work, we first re-parametrize a popular OCV model formulation by the N/P (negative-to-positive) ratio and Li/P (lithium-to-positive) ratio, which have more symmetric and intuitive physical meaning, and are also pristine-condition-agnostic and cutoff-voltage-independent. We then study the modal identifiability of capacity fade by mathematically deriving the gradients of electrode slippage and cell OCV with respect to these SOH parameters, where the electrode differential voltage fractions, which characterize each electrode's relative contribution to the OCV slope, play a key role in passing the influence of a fixed cutoff voltage to the parameter sensitivity. The sensitivity gradients of the total capacity also reveal four characteristic regimes regarding how much lithium inventory and active materials are limiting the apparent capacity. We show the usefulness of these sensitivity gradients with an application regarding degradation mode identifiability from OCV measurements at different SOC (state of charge) windows.