Guided wave based state-of-charge inspection of lithium-ion batteries using frequency-wavenumber domain features

Ultrasonic inspection has been recognised as one of the promising non-destructive evaluation (NDE) techniques for assessing the condition of Lithium-ion batteries (LiBs). However, the limited size of batteries causes guided waves to overlap, reducing the precision of conventional time of flight (ToF) and amplitude-based indicators. In this paper, a frequency–wavenumber ( f-k ) based index E f − k is proposed to address this limitation. In addition, a numerical finite element model incorporating lithiation-induced property variations is developed to simulate guided wave propagation across different states of charge (SoCs), further revealing the underlying mechanism during lithiation. An experimental study shows that in pouch-type LCO batteries, the wavenumber k P of the anti-symmetrical mode (A mode) decreases with battery charging. This trend is primarily associated with a significant increase in the shear modulus of the anode and a reduction in its compacted density. Simultaneously, the decrease in E f − k reflects increased wave attenuation due to elevated fluid content and enhanced microstructural heterogeneity in both the anode and cathode. A comparative analysis of ToF, amplitude, k P and E f − k index reveals distinct characteristics: ToF exhibits a clear but weakly sensitive trend with relatively low variability, whereas amplitude shows high sensitivity but suffers from substantial variability due to wave overlap. The wavenumber k P demonstrates moderate sensitivity with high reliability, particularly in the mid-to-high SoC range. E f − k index achieves an optimal balance between sensitivity and reliability, highlighting its potential for guided-wave-based LiB diagnostics.