State of Charge Estimation of the Lithium-Ion Battery Pack Based on Two Sigma-Point Kalman Filters

Nowadays, the lithium-ion battery pack (LiB) is used as the main power supply for electric vehicles (EV). The remaining energy of LiB is the very important parameter determined continuously by estimating LiB's state of charge (SoC). SoC estimation is one of the main functions of the battery management systems (BMS). This article presents the use of two sigma-point Kalman filters (SPKF) to estimate accurately the SoC of the LiB based on the second-order model of the cell. The LiB's average SoC and the zero bias of the current measurement through the LiB are estimated by the first SPKF, while the second filter is applied to calculate the SoC differences between LiB's average SoC and the modules' SoC in the LiB. To improve the SoC accuracy of the LiB modules, a second-order RC equivalent circuit model (SECM) of the cell is used, and the influences of temperature, voltage hysteric, measurement errors, and zero bias of current measurement on the SoC estimation of the LiB are taken into account. To verify the method, the experimental test is conducted in the LiB with cells connected in parallels and series. The simulation and experimental results are analyzed to prove that the SoC estimation of the modules in the LiB is higher accuracy, and the LiB's average SoC errors are less than 1.5% at different temperatures ranging from - 5 to 45 degrees C. The calculation time consuming is shorter, and the calculation complex is reduced significantly.