Data-driven state-of-charge estimation of lithium-ion batteries
Yuanliang Fan, Jing Wu, Zitao Chen, Han Wu, Jianye Huang, Binqian Liu
Abstract
Accurate estimation of state-of-charge (SOC) is essential for battery management system. This paper proposes a data-driven method for estimating the SOC of lithium-ion batteries. Long and short-term memory (LSTM) neural networks are designed for estimation of SOC, in which the currents and temperatures of the battery are defined as the inputs of the neural network, while the output of the neural network is considered as the SOC. Basing on these input and output data, the neural network is trained, which is further used as a model for estimating the SOC. The simulation results verify that the proposed method can meet the accuracy requirements about estimation of the SOC for battery management system.