A method for state of energy estimation of lithium-ion batteries at dynamic currents and temperatures

The state of energy (SOE) of Li-ion batteries is a critical index for energy optimization and management. In the applied battery system, the fact that the discharge current and the temperature change due to the dynamic load will result in errors in the estimation of the residual energy for the battery. To address this issue, a new method based on the Back-Propagation Neural Network (BPNN) is presented for the SOE estimation. In the proposed approach, in order to take into account the energy loss on the internal resistance, the electrochemical reactions and the decrease of the open-circuit voltage (OCV), the SOE is introduced to replace the state of charge (SOC) to describe the residual energy of the battery. Additionally, the discharge current and temperature are taken as the training inputs of the BPNN to overcome their interference on the SOE estimation. The simulation experiments on LiFePO4 batteries indicate that the proposed method based on the BPNN can estimate the SOE much more reliably and accurately.