A lithium-ion battery RUL prognosis method using temperature changing rate

Abstract

As a kind of complex electrochemical system, the performance of lithium-ion battery will degrade under continuous charging and discharging. It's particularly crucial to monitor the battery state of health and prognosis the battery remaining useful life (RUL). Considering the highly linear correlation between capacity and the changing rate of temperature (TR), a new RUL prediction approach is proposed which provides a better description on the capacity degradation based on the changing rate of battery temperature and cycle number N. First a binary linear regress model is proposed for battery state of health (SOH) and RUL prognosis. Then TR ratio which is extracted for TR prediction is predicted using the chosen historical data considering the similarity of different data sets. Finally, capacity is estimated sequentially based on the proposed model with the predicted TR and cycle number N. The results show that TR can not only indicate state-of-health more accurately, but also provide more precise and better robustness in RUL prediction and SOH monitoring. Furthermore, the regeneration of battery can be accurately predicted by our method.