SOC Estimation Method of Lithium-ion Batteries Based on Relaxation Effect and GRU-RNN

State of charge (SOC) of lithium-ion battery is one of the key parameters for battery management system (BMS). This paper proposes a SOC estimation method of lithium-ion batteries based on relaxation effect and gated recurrent unit recurrent neural network (GRU-RNN). Firstly, the SOC estimating model based on relaxation effect is proposed using GRU-RNN. Then, a periodic discharge experiment is designed according to the characteristics of relaxation effect, which is used to collect voltage relaxation dataset. Finally, the proposed model is trained using the voltage relaxation dataset, and the SOC estimation results of proposed model under different discharge currents are verified.