Abstract
Battery ageing limits the performance, safety, and lifetime of lithium-ion battery systems, creating a critical need for diagnostic tools capable of guiding advanced battery management systems (BMS). Electrochemical impedance spectroscopy (EIS) is a powerful technique that is sensitive to the kinetic, interfacial, and transport processes underlying degradation. However, extracting physically interpretable, model-compatible information remains challenging. Here, we present a physics-based, EIS-informed parameter identification framework that supports a digital-twin-enabled BMS. Experimental EIS spectra are integrated into a comprehensive electrochemical model through nonlinear parameter estimation, yielding a set of kinetic, interfacial, and transport parameters that evolve with state of charge (SoC) and cycling-induced ageing. The extracted parameters are independently validated against time-domain galvanostatic cyclic simulations, predicting the state of health (SoH) with a root mean square error (RMSE) of approximately 1% compared to experimental data. The identified parameters capture trends associated with SoC and ageing, providing mechanistically interpretable indicators of degradation that remain consistent across frequency- and time-domain analyses. By enabling the extraction of ageing-sensitive parameters from EIS data, this approach supports model-informed and digital-twin-based BMS strategies, contributing to improved lifetime prediction, adaptive control, and the sustainable operation of lithium-ion batteries.

This publication has 33 references indexed in Scilit: