Online Co-Estimation of the State-of-Health, State-of-Charge, and Remaining-Useful-Life of Lithium-Ion Batteries Using a Discrete Capacity Loss Model

Abstract

The main objectives of a battery management system (BMS) are to monitor the state-of-charge (SoC) and state-of-health (SoH) of lithium-ion batteries (LIBs). Due to their coupled nature, the SoC and SoH should be estimated simultaneously. In this article, an online co-estimation approach of the SoC, SoH, and remaining-useful-life (RUL) of an LIB has been proposed based on a novel discrete capacity loss (DCL) model. A particle filter (PF) has been used for battery capacity loss estimation using the DCL model to obtain its SoH. Parallelly, the estimated capacity loss was utilized for an online update of the capacity over battery aging. Thereafter, the updated capacity with the recursive least-squares (RLS) technique-based estimated equivalent circuit model (ECM) parameters was used for SoC estimation using an extended Kalman filter (EKF). Furthermore, the DCL model was used for RUL prediction using the capacity loss in terms of remaining ampere-hour throughput (AhT). The proposed result shows an SoH relative error (R.E.) band of ±0.05% and an SoC error band from 0% to −0.3% and 0% to −0.35% for a fresh and an aged battery, respectively, which outperforms the state-of-the-art co-estimation method. The RUL prediction error (P.E.) is just 20 Ah after using the first 900 Ah data for prediction.