Edge of Transfer Learning-Based Long Short-Term Memory Neural Networks in the Application of Battery Surface Temperature Prediction for Electric Vehicles

Lithium-ion (Li-ion) batteries are highly sensitive to operating conditions and temperature is one of the critical conditions that affect their performance. This article proposes a data-driven method for the prediction of the surface temperature of Li-ion batteries so that preemptive measures could be taken to maintain the temperature within the optimum range. LSTM-based neural network is one such method that helps in prediction using sequential data. In this article, a brief and effective comparison between G-LSTM and LSTM-TL is shown for the prediction of surface temperature. Theoretically, the TL method should reduce the computational burden and improve the prediction performance and the same has been observed in our experiment. This will make the system fault-tolerant. Moreover, the wide generalization and applicability of the developed model are shown through the temperature prediction on two different batteries that were not used in the training. The experimental results demonstrate that the G-LSTM model is capable of temperature prediction with root mean square error of 1.1578 $<^>{\circ }$C for battery 03 and 1.2101 $<^>{\circ }$C for battery 04. This error has been further reduced by around 40% to a value of 0.5012 $<^>{\circ }$C for battery 03 and 0.7480 $<^>{\circ }$C for battery 04 by using the LSTM-TL.