Remaining useful life prediction of lithium battery based on capacity regeneration point detection

Lithium batteries have been widely used in various electronic devices, and the accurate prediction of its remaining useful life (RUL) can prevent the occurrence of sudden equipment failure. Battery capacity is a commonly used indicator to represent the health status of lithium batteries. However, the capacity regeneration is usually unavoidable due to the impact of battery rest time between two cycles, which leads to inaccurate prediction of the RUL. To solve this problem, this paper combines the particle filter (PF) and Mann-Whitney U test (PF-U) to detect the capacity regeneration point (CRP). In this light, the autoregressive (AR) model and PF algorithm are adopted for RUL prediction. The predicted capacity through AR model is used to update the degradation model parameters of PF algorithm, and the validation of our approach is verified through the lithium battery dataset of NASA. In comparison, our proposed method exhibits the highest precision and provides a platform to detect the points with capacity regeneration, and further reduce the RUL prediction error. (c) 2021 Elsevier Ltd. All rights reserved.

Automated summary

This study combines particle filter and Mann-Whitney U test to detect the capacity regeneration point of lithium batteries, using autoregressive model and PF algorithm for RUL prediction. The method is validated through experiments, showing improved prediction accuracy and reduced error rate.