Screening process-based modeling of the multi-cell battery string in series and parallel connections for high accuracy state-of-charge estimation

Abstract

This paper investigates a practical universal modeling of multi-cell battery strings in series and parallel connections to show high an accuracy SOC (state-of-charge) estimation based on the EKF (extended Kalman filter) if cell-to-cell variations are taken into account and settled by the screening process. Through the screening process for the selection of the cells that have similar electrochemical characteristics, this study describes an effort to provide each equivalent circuit model for multi-cell battery strings in series, parallel, and series/parallel connections. Three circuit-based multi-cell battery models are validated against the experimental data of the discharging/charging behavior in terms of the discharging/charging voltage, discharge capacity, OCV (open-circuit voltage), and internal resistances when compared with the experimental data of a single cell. According to the relation between a multi-cell battery string and a single cell, these models can be easily developed from a single cell model, the validity of which was demonstrated regarding its high accuracy in predicting cell performance. The proposed multi-cell battery model has been extensively validated by the model-based SOC estimation using the EKF for a Li-Ion cell. If the model parameters of a single cell are correctly measured and used in the multi-cell battery model, the accuracy in the SOC estimation of a multi-cell battery string could be significantly improved.