Remaining Useful Life Prediction for Lithium-Ion Batteries Based on Improved Mode Decomposition and Time Series

Accurately predicting the remaining useful life (RUL) of lithium-ion batteries holds significant importance for their health management. Due to capacity regeneration phenomenon and random interference during operation batteries, a single model may exhibit poor prediction accuracy generalization performance under scale signal. This paper proposes method RUL batteries. The is based on improved sparrow search algorithm (ISSA), which optimizes variational mode decomposition (VMD) long- short-term time-series network (LSTNet). First, this study utilized ISSA-optimized VMD decompose degradation sequence acquiring global trend components local recovery components, then ISSA–LSTNet–Attention ISSA–LSTNet–Skip were employed predict component component, respectively. Finally, results these different models integrated accurately estimate proposed was tested two public battery datasets; indicate root mean square error (RMSE) 2%, absolute (MAE) 1.5%, an correlation coefficient (R2) Nash–Sutcliffe efficiency index (NSE) both above 92.9%, implying high superior compared other models. Moreover, significantly reduces complexity series.