An adaptive kalman filter to estimate state-of-charge of lithium-ion batteries

Abstract

Fast and accurate estimation of battery state of charge (SOC) is a key technology in the battery management system. Based on the non-linear response characteristics of lithium batteries, an adaptive Kalman filter algorithm is put forward in this paper. It is known that the battery model parameters vary with SOC, battery temperature and battery aging. Moreover, the relationship between open circuit voltage (OCV) and SOC is nonlinear. To solve these issues, a piecewise linear approximation of the model parameters is proposed based on the SOC, and then the nonlinear battery model is turned into a piecewise linear one. On these bases, an adaptive Kalman filter can be implemented and thus the amount of computation can be reduced. In addition, we apply the Arrhenius equation to update internal resistance and the remaining capacity of battery which can reflect the aging state of battery. The algorithm achieves an adaptive SOC estimation and improves the estimation accuracy with a small amount of calculation. Finally, the simulation results show the accuracy and applicability of the algorithm.