Analysis on SoC estimation of Lithium ion battery using EKF and LSTM

State of Charge (SoC) is an important parameter of a battery for its safety measures. The battery model consists of parameters like internal resistance and two RC stages. These parameters vary according to the internal temperature and SoC of the battery. So, it is necessary to know the characteristics of battery parameters with respect to SoC and internal temperature of the battery. In this paper, Extended Kalman Filter (EKF) is used to estimate the SoC for different temperatures under fixed and variable load conditions. To overcome the limitations of EKF, the data driven model, Recurrent Neural Network (RNN) based Long Short Term Memory (LSTM) model is adopted to estimate the SoC. It is observed from the simulation results that the performance of LSTM model is better than the EKF in terms of RMSE value.