State-of-Charge Estimation of Lithium-ion Batteries Based on Deep Neural Network

摘要

The State of Charge (SOC) of the battery is one of the core functions of the battery management system (BMS). Accurate estimation of SOC is essential for the reliability and safety of battery systems. The model-based SOC estimation method requires the establishment of an accurate battery model and the use of complex adaptive filtering algorithms, which is difficult to implement in engineering. Lithium-ion batteries generate a large amount of data such as voltage, current and temperature during the working process, which combines pure data-driven deep learning algorithms with lithium-ion batteries SOC estimation. This paper is based on the deep neural network to estimate the SOC of lithium-ion batteries, using public data sets and model simulation data sets to compare the effectiveness of Gated Recurrent Unit Neural Network (GRU), Long Short-Term Memory Network (LSTM) and Recurrent Neural Network (RNN) on SOC estimation. It verifies that GRU is superior to LSTM and RNN in terms of network performance and estimation accuracy.