Hystimator: DRT-based hysteresis modelling for accurate SoC estimation in LFP battery cells
No Thumbnail Available
Author
Thenaisie, Guillaume
Brivio, Claudio
DOI
10.1049/rpg2.13130
Abstract
State of Charge (SoC) estimation for Lithium-Iron Phosphate (LFP) batteries is challenging due to a flat Open Circuit Voltage (OCV) curve and a well-known hysteresis effect. The authors built upon a previous study, which has shown that hysteresis in LFP is not an inherent characteristic but a very slow relaxation process when compared to other battery chemistries. Distribution of Relaxation Times (DRT) is used to deconvolve Electro-Impedance Spectroscopy (EIS) measurements and model the hysteresis effect. The extracted DRT parameters show good agreement at low frequencies with previous thermodynamic studies in both fresh and aged cell conditions. The proposed model, called hystimator, integrates the hysteresis characteristics into a physics-based Electro-Chemical Model (ECM). The validation results show a significant reduction in the Root Mean Square Error (RMSE) during real-world laboratory testing. This approach holds promise for SoC estimation in LFP battery cells, especially in embedded Battery Management System (BMS).
Publication Reference
IET Renewable Power Generation, 18 (S1), pp. 4387 - 4398
Year
2024