Thermal Runaway Detection of Lithium-ion Battery Based on SE-Res-LSTM

Monitoring the thermal runaway of lithium-ion battery, especially to detect the thermal runaway fault of lithiumion battery in the overcharge state is of great importance. This paper, therefore, proposes a battery state detection method based on squeeze and excitation residual (SE-Res) net combined with long short-term memory neural network (LSTM) to estimate the battery expansion force signal, a residual threshold is set to accurately determine the lithium-ion battery thermal runaway in real time. The accuracy of the scheme is verified by using real battery data under normal cycle and overcharge operation. The experimental results show that this method can accurately detect the thermal failure of lithium-ion battery under overcharge state.