To reduce the chattering phenomenon in conventional sliding mode observers(CSMOs)for state of charge(SoC)estimation of lithium battery and improve estimation accuracy

a novel improved adaptive sliding mode observer(IASMO)was proposed. First

based on a dual polarization(DP)equivalent circuit model

a nonlinear terminal sliding surface was introduced to ensure the convergence of the objective function to the equilibrium point in finite time

while a continuous control rate was designed to decrease chattering in the estimation results. Second

to address modeling uncertainties

an adaptive switching gain function was developed

which adjusts automatically according to the sliding surface function

further enhancing the estimation accuracy and robustness of the algorithm. The convergence of the observer was proven using Lyapunov stability theory. Finally

based on the self-build experimental platform

battery data under different working conditions were collected

and a simulation model was built for verification. The experimental results indicate that the proposed ISMO has better estimation accuracy and better robustness than the CSMOs and currently mainstream types of improved sliding mode observers.