State of Charge Estimation Using Unscented Kalman Filter in Electric Vehicles

Lithium-ion batteries are preferred more than other types of batteries in electric vehicle applications due to their inherent safety, fast charge capacity, and long life. To create an accurate battery model, it is important to be able to identify health factors such as charge and health. Using LA92 drive cycle experimental data, the typical lithium-ion battery charge state estimation algorithm has been improved. First, we created a mathematical model of an analogue circuit battery to accurately mimic the behavior of a lithium-ion battery. The Thevenin model consists of two RC branches and uses an extended Kalman filter to identify model parameters. The three-dimensional curve of SOC as a function of SOC and T using hybrid pulsed power characteristic (HPPC) test data achieved at 5 degrees from 40°C to −10°. To calculate. A comparison of the two methods is shown, showing that the UKF method for battery SOC assessment is more reliable than the traditional method CC (Coulomb counting). The error observed in the UKF results is less than 1%, indicating that the UKF can reliably estimate battery condition.