Data-Driven Online Health Estimation of Li-Ion Batteries Using A Novel Energy-Based Health Indicator
Wei Liu, Yan Xu
Abstract
Li-Ion batteries have been widely applied in power engineering. Aiming at online state of health (SOH) estimation of Li-Ion batteries, this letter develops a data-driven method using a novel energy-based health indicator (HI). The proposed HI is extracted from the discharge process considering that the discharge process is often less controllable than the charge process. Unlike previous works where only voltage sequences are considered, this HI incorporates both voltage sequences and discharge rates. Therefore, the developed HI enables online SOH estimation at different discharge rates from the offline training dataset. An open dataset is used for verification of the proposed method and very high accuracy is reported with an average RMSE of 1.23%.