Lithium‐Ion Battery SOC Estimation Using Adaptive Sliding Mode Observer and Wavelet Transform

This paper proposes a new method for the state‐of‐charge (SOC) estimation of lithium‐ion (Li‐ion) batteries using an adaptive sliding mode observer and wavelet transform (ASMO‐WT). Unlike existing methods that rely on traditional adaptive gain sliding mode observers, the proposed method reduces measurement noise and estimation chattering. The noisy measurement signals of current and voltage are denoised using the wavelet transform (WT). Soft‐ and hard‐thresholding‐based denoising algorithms are employed to tune wavelet coefficients in order to achieve a good distinction between signal and noise. The dynamic behaviors of the Li‐ion battery are expressed in the state equation of a second‐order RC equivalent circuit model (ECM), and the model parameters are determined by using the hybrid pulse power characterization (HPPC) method. The results are compared to those of ASMO, extended Kalman filter (EKF), and unscented Kalman filter (UKF). The results show that the proposed ASMO‐WT method is effective at reducing chattering, and achieving a mean absolute error of 0.0242% and a root mean square error of 0.0274%, achieving high accuracy in SOC estimation at different temperatures. Moreover, ASMO‐WT requires less time to converge when an initial SOC estimation error is introduced, whereas ASMO, EKF, and UKF take longer time. The results suggest that the proposed algorithm reduces chattering while simultaneously improving estimation accuracy and performance.