Battery remaining useful life prediction under coupling stress based on support vector regression

Batteries are used in many areas, such as the electronic, astronautics, aeronautics, automobile and energy, etc. However, there are many explosion and fire accidents caused by the battery aging, so accurately estimating its remaining useful life (RUL) is very critical. The research of battery life involves many disciplines. The aging process is affected by many factors, such as temperature, discharge depth, charge and discharge rate and operation condition. In this paper, the nonlinear capacity degradation of ternary Li-ion batteries is studied. By data-driven prediction methods, the RUL prediction model of ternary Li-ion batteries under coupling stress based on SVR is established by 6 sets of coupling stress experimental data, and through another set of coupled stress test verifies the correctness of the proposed model.By the model,the RUL can be estimated accurately,which can guarantee the safe operation of the locomotive and so on. Copyright (C) 2018 Elsevier Ltd. All rights reserved.