Battery life prediction method based on DE-GWO-LSTM

Aiming at the problem of inaccurate prediction results of lithium-ion battery life, a lithium-ion battery life prediction model based on hybrid algorithm is designed. The position of grey wolf algorithm is updated by differential evolution algorithm, which improves the population diversity and avoids premature stagnation of the algorithm. The GWO-LSTM model and DE-GWO-LSTM model are compared and analyzed by using NASA data. The proposed DE-GWO-LSTM can well conduct global search and local search, and improve the prediction performance to a certain extent.