A Multi-Variable Multi-Step Seq2seq Networks for the State of Charge Estimation of Lithium-Ion Battery

Abstract

As a key index of Lithium-ion battery, the state of charge (SOC) has attracted many researchers’ attention, which is significance for the performance optimization and safety of the Electric vehicles. Due to the complexity and changeable of lithium-ion batteries, we propose a multi-variable and multi-step Temporal neural network to cover this task. Specially, a novel multi-step training strategy is applied to deal with long time sequences, and multi-variables is added to supervise the prediction. In addition, the Seq2seq net with LSTM modules is employed to improve the prediction accuracy in various lithium-ion battery operating environments of 10℃, 25℃ and 40 ℃. The A123 lithium-ion batteries’ database collected by University of Maryland for the real scenarios has been used in the training process. The ablation studies indicate that the multi-step module is necessary under the RMSE, MAE and MAPE index. The experimental results indicated that our model obtains the higher estimation accuracy and better robustness. The prediction error of comparison experiment under our proposed algorithms is 0.45%@ RMSE, 0.30%@ MAE and 1.11%@ MAPE at 10 ℃, so that the proposed algorithm is excellent upon other models on the prediction of lithium-ion batteries’ SOC.

Keywords: Lithium-ion batteries, SOC, LSTM, Seq2seq, multi-variable, multi-step