Study of battery state-of-charge estimation for hybrid electric vehicles

Aiming at the identification problem of battery state of charge (SOC) in hybrid electric vehicle, the neural network method is applied on the battery SOC estimation to build the power battery model based on radial basis function (RBF) after considering the nonlinear relationship between the battery state of charge and the observable external characteristics. Two simulation experiments were done under the situation of constant current charging and discharging mode and dynamic vehicle driving mode. The results show that the proposed SOC estimation method is feasible and of high estimation accuracy.