Estimation of state of health and remaining useful life of Li-ion batteries using LSTM, SVR, and GRN models

In this study, the state of health (SOH) and remaining useful life (RUL) of lithium-ion batteries are estimated using the most widely used and accurate machine-learning methods, including long short-term memory (LSTM), gated residual network (GRN), and support vector regressor (SVR). This study is done to prepare a numerical comparison about SOH and RUL estimation of lithium-ion batteries using machine learning methods. This kind of batteries are widely consumed in different industrial applications and it is necessary for the user to have an accurate estimation of SOH and RUL. NASA data was used to train the models. Based on the results, LSTM has the highest accuracy in estimating the SOH and RUL of lithium-ion batteries.

کلیدواژه ها:

lithium-ion batteries ، SOH ، RUL ، LSTM ، GRN ، SVR