SOC Estimation of Li-ion Batteries with Learning Rate-Optimized Deep Fully Convolutional Network

No Thumbnail Available
Hannan M.A.
How D.N.T.
Hossain Lipu M.S.
Ker P.J.
Dong Z.Y.
Mansur M.
Blaabjerg F.
Journal Title
Journal ISSN
Volume Title
Institute of Electrical and Electronics Engineers Inc.
Research Projects
Organizational Units
Journal Issue
In this letter, we train deep learning (DL) models to estimate the state-of-charge (SOC) of lithium-ion (Li-ion) battery directly from voltage, current, and battery temperature values. The deep fully convolutional network model is proposed for its novel architecture with learning rate optimization strategies. The proposed model is capable of estimating SOC at constant and varying ambient temperature on different drive cycles without having to be retrained. The model also outperformed other commonly used DL models such as the LSTM, GRU, and CNN on an open source Li-ion battery dataset. The model achieves 0.85% root mean squared error (RMSE) and 0.7% mean absolute error (MAE) at 25 �C and 2.0% RMSE and 1.55% MAE at varying ambient temperature (-20-25 �C). � 1986-2012 IEEE.
Charging (batteries); Convolution; Convolutional neural networks; Deep learning; Ions; Learning systems; Lithium compounds; Lithium-ion batteries; Long short-term memory; Temperature; Battery temperature; Convolutional networks; Learning rates; Lithium ions; Novel architecture; Open sources; SOC estimations; State of charge; Battery management systems