An accurate SOC estimation system for lithium-ion batteries by EKF with dynamic noise adjustment

Based on an SOC (State of Charge) estimation system for Lithium-ion battery by EKF (Extended Kalman filter), we have improved the accuracy by dynamic adjusting of noises in the battery model. Firstly, the battery model and the SOC (State of Charge) estimation algorithm by EKF are explained. Then, static optimization for the process noise and observation noise in the battery model is discussed. Afterward, an adaptive noise tuning method is proposed and its accuracy is evaluated by some examinations. The error by the static method is 0.97% and 1.21% in different test patterns. The average of the SOC estimation errors with the adaptive noise tuning is 0.87% and 1.16%. Significant improvement of accuracy has been achieved.