Electrochemical Impedance Spectroscopy of Rechargeable Li-Ion Batteries: Novel Insights and Improvements

Electrochemical impedance spectroscopy (EIS) is a widely used technique for battery health diagnostics, wherein a frequency-domain excitation signal is applied to the battery and the corresponding voltage response is analyzed to construct its impedance spectrum. Existing EIS methodologies typically model the battery as a linear time-invariant (LTI) system, assuming that under small perturbations and resting conditions, the impedance spectrum accurately reflects the battery’s behavior. However, this paper demonstrates that the presence of the battery’s open-circuit voltage (OCV) invalidates the LTI assumption, even when a zero-mean excitation signal is used on a fully rested battery. Specifically, we show that OCV variations introduce nonlinearity in the voltage response, significantly affecting the impedance spectrum in the low-frequency range. To address this limitation, we propose a novel approach that integrates a tailored excitation signal design and an observation model to robustly estimate and remove the OCV’s contribution from the measured response. This correction enables a more accurate impedance characterization by explicitly accounting for the inherent nonlinearity introduced by the OCV, thereby overcoming the limitations of traditional and state-of-the-art EIS methods. The proposed approach is validated using data obtained from a battery simulator and a cylindrical Li-ion battery cell with a high-precision EIS device. The approach does not need battery parameters and can be used across different chemistries, sizes, temperatures, and aging conditions.