Neural Network-Based State of Charge Observer Design for Lithium-Ion Batteries

A new method for the state of charge (SOC) estimation of lithium-ion batteries is proposed based on an inclusive equivalent circuit model in this brief. In particular, we propose to utilize the neural network to estimate the uncertainties in the battery model online. A radial basis function neural network-based nonlinear observer is then designed to estimate the battery's SOC. Following Lyapunov stability analysis, it is proved that the SOC estimation error is ultimately bounded and the error bound can be arbitrarily small. Experimental and simulation results illustrate the performance of the proposed approach. Furthermore, we compare the SOC estimation results of the extended Kalman filter with the proposed radial basis function neural network-based nonlinear observer. The proposed approach has faster convergence speed and higher precision.