摘要：AbstractThe State of Charge (SoC) is one of the important quantities estimated by the Battery Management System (BMS) of Lithium-ion cells. However, the hysteresis effect and flat SoC-OCV nature of Lithium Iron Phosphate (LFP) battery complicate the SoC estimation. This paper proposes a novel switched model to successfully capture the hysteresis phenomena and enhance the accuracy of SoC estimation of LFP cells. The model is switched between charge and discharge modes, where the current direction decides the mode. The model parameters are functions of SoC and the switched mode. The parameters are estimated from Pulse Charge Data (PCD) and Pulse Discharge Data (PDD) using a Successive Recursive Least Square (SRLS) technique. The SRLS algorithm ensures sufficiency of excitation by capturing only the transient response of each pulse. Using the proposed model, SoC estimation is carried out using the Extended Kalman Filter (EKF). The proposed approach is validated by a real drive cycle data which is widely used to test vehicle performance. The study has been carried out onLiFePO4pouch cell with a nominal capacity of 20Ah and a nominal voltage of 3.3V and experiments are performed using the Biologic (BCS-815) battery testing equipment.
关键词：KeywordsBattery modelExtended Kalman Filter (EKF)HysteresisLFP pouch cellSuccessive Recursive Least Square (SRLS)State of Charge (SoC)
