A State of Charge Estimation Method Based on Adaptive Unscented Kalman Filter for Lithium-ion Parallel-connected Battery System

Abstract

Due to state of charge (SOC) is a key parameter for the safe operation and control of a parallel-connected battery system (PBS), It is essential to estimate accurately SOC of a lithium-ion PBS composed of thousands of inconsistent cells when the noise statistics are unknown and/or time-varying, such as model noises and measurement noises. To resolve the problems, an equivalent circuit model of the PBS is firstly presented based on a model parameter regulator that can overcome the influence of cell-to-cell variation. An adaptive SOC estimation method based on adaptive unscented Kalman filter (AUKF), which is a time-varying unscented Kalman filter with a noise statistics estimator, is proposed for the PBS when the noise statistics are unknown. Compared with the UKF and EKF, the accuracy and effectiveness of the SOC estimation method is validated by the simulation results and experimental data.