This work developed and evaluated an acousto-ultrasonic system that measures battery state of charge and, in particular, its state of health using a built-in, low-profile piezoelectric transducer system. A diagnostic method was proposed that relates changes in the guided wave signals to the charge, discharge, and aging processes, via electrochemically-induced changes in mechanical properties. A matching-pursuit-based feature extraction scheme was developed to allow an efficient, in-operando decomposition of signals into a set of predictors correlated with battery states. A particle filter framework was established which provides state estimation and remaining life prognostics to allow ultrasonic features to be used as measurements. It was shown on off-the-shelf Li-ion batteries and battery-integrated structures that the ultrasonic technique significantly improved the prediction performance and contained uncertainties.
Reference
12th International Workshop on Structural Health Monitoring (IWSHM), Stanford University, USA, September 2019.
Bibtex
@article{ladpli2019health,
title={Health Prognostics of Lithium-ion Batteries and Battery-Integrated Structures},
author={LADPLI, PURIM and LIU, CHEN and KOPSAFTOPOULOS, FOTIS and CHANG, FU-KUO},
journal={Structural Health Monitoring 2019},
year={2019}
}