State estimation using physics-based equivalent circuit models of a Li-ion cell and Kalman filter

We have previously proposed a novel parameter identification method for the Li-ion battery equivalent circuit model (ECM) considering the electrochemical properties, with which a second order ECM resistances/capacitances values can be obtained from electrochemical parameters. In this paper, we apply this physics-based ECM with Kalman filter for cell SOC estimation. To implement the Kalman filter to nonlinear systems, we adopted both extended Kalman filter, which has most widely been used to linearize the nonlinear system models with Taylor-series expansion, and sigma-point Kalman filter, in which the nonlinear equations are directly used to propagate the sigma points through the system state equation and the observation equation so that the means and the covariance of the state vector can be estimated better than by the EKF. Simulation results show EKF/SPKF can give accurate estimation with error bounds, which verify the applicability of our method for cell SOC estimation. Besides, EKF and SPKF results are compared.