Near-real-time parameter estimation of an electrical battery model with multiple time constants and SOC-dependent capacitance

A modified particle swarm optimization algorithm for conducting near-real-time parameter estimation of an electrical model for lithium batteries is presented. The model comprises a dynamic capacitance and a high-order resistor-capacitor network. The algorithm is evaluated on a hardware test bed with two samples of 3.3V, 40Ah, Lithium Iron Phosphate (LiFePO₄) battery driven under six different loading patterns. All intrinsic parameters together with the state-of-charge of the battery are estimated by firstly processing the 15-minute samples of the terminal voltage and current. Then, the voltage-current characteristics in the following 15 minutes are predicted. Results show that the extracted parameters can fit the first 15-minute voltage samples with high accuracy. Moreover, the electrical model can predict voltage-current characteristics in the following 15 minutes with the extracted parameters. The study lays foundation for the possibility of applying computational intelligence algorithms for parametric estimation of batteries.