An Ultrasonic Wave-Based Method for Efficient State-of-Health Estimation of Li-Ion Batteries
Kailong Liu, Yuhang Liu, S. P. Zhao, Xiaoyu Li, Qiao Peng
Abstract
Effective state-of-health (SoH) estimation is highly valuable for ensuring battery performance and safety. This article proposes an ultrasonic wave-based method for the accurate and rapid SoH estimation of lithium-ion (Li-ion) battery, enabled by combining the benefits of nondestructive ultrasonic detection and interpretable data-driven solution. In particular, an Li-ion battery is externally equipped with an ultrasonic sensor to promise fast real-time measurement of the ultrasonic signal with high spatial resolution and reproducibility. Following this endeavor, an explainable data-driven solution based on the enhanced interpretable generalized additive model with interactions (EI-GAMIs) is proposed to rapidly estimate battery SoH and explain the effects of the involved ultrasonic features of interest. Illustrative results disclose that with the guided ultrasound wave which can be obtained in microsecond level especially for a state-of-charge (SoC) range of 40%–60%, Li-ion battery SoH could be estimated precisely in real time with an <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$ R^{2}$</tex-math></inline-formula> above 0.99 via the proposed method. Moreover, the contributions of the main effect and pairwise interaction terms derived from six ultrasonic features could be quantified, while the global and local interpretations of their dynamic influences are well explained. This can help engineers to quickly obtain reliable information about battery health in real-world applications, which in turn will benefit the health management of Li-ion batteries.