State-of-Charge Estimation of Lithium-ion Batteries Using the Nesterov Accelerated Gradient Algorithm Based Bi-GRU

Electric vehicles have been developing rapidly in recent years, the State-of-Charge (SOC) of lithium-ion batteries can related to the safety and reliability of electric vehicles. With the development of the field of machine learning, data-driven methods often use machine learning platforms to obtain the relationship between battery measurement signals and SOC from them to achieve accurate estimation of SOC. To further explore the potential of machine learning platforms in SOC estimation, an improved model using the Nesterov Accelerated Gradient (NAG) algorithm based Bidirectional Gated Recurrent Unit (Bi-GRU) network is put forward in this paper. Furthermore, a well-recognized lithium-ion battery dataset from University of Maryland is applied to evaluate the validity and quality of the established NAG based Bi-GRU model. Finally, the results demonstrate that the established NAG based Bi-GRU model have a preferable precision of the SOC estimation at various ambient temperature, compared to state-of-the-art data-driven estimation methods.