Design of a Non-Linear Observer for SOC of Lithium-Ion Battery Based on Neural Network

This paper presents a method for use in estimating the state of charge (SOC) lithium-ion batteries which is based on an electrochemical impedance equivalent circuit model with controlled source. Considering that open-circuit voltage battery varies SOC, source proposed and current interact each other. On this basis, radial basis function (RBF) neural network adopted to estimate uncertainty online, non-linear observer RBF designed SOC batteries. It proved estimation error ultimately bounded by Lyapunov stability analysis, bound can be arbitrarily small. The high accuracy validity are verified experimental simulation results. results extended Kalman filter (EKF) compared method. improves convergence speed accuracy.