Predicting battery capacity from impedance at varying temperature and state of charge using machine learning

Predicting battery capacity from impedance at varying temperature and state of charge using machine learning Gasper et al. demonstrate prediction of battery capacity using electrochemical impedance spectroscopy data recorded under varying conditions of temperature and state of charge.A variety of methods for featurization of impedance data are tested using several machine-learning model architectures to rigorously investigate the limits of using impedance to monitor battery health.