Co-estimation and definition for states of health and charge of lithium-ion batteries using expansion

Abstract

Accurate and efficient estimation of state of charge (SOC) and state of health (SOH) is crucial for the safe and stable operation of lithium-ion batteries (LIBs). However, achieving accurate SOC–SOH co-estimation remains a challenging task. Meanwhile, the states are primarily defined by capacity, which cannot be directly measured and requires extensive calculations. Hence, a SOH–SOC co-estimation method based on multi-time scale long short-term memory (LSTM) is proposed, which utilizes the expansion characteristics. Additionally, definitions of SOH and SOC are presented based on expansion for the first time, which can obtain the LIBs’ states by direct measurement. To evaluate the necessity of expansion measurements for battery state estimation and definition, the University of Michigan Battery Laboratory (UMBL) dataset is utilized. Then, the relationships between expansion characteristics and SOH/SOC are deeply analyzed. The experimental results demonstrate that the proposed co-estimation method estimates SOC and SOH with high accuracy. The root mean square error (RMSE) and mean absolute error (MAE) are within 0.68 % and 0.51 %, respectively. Moreover, compared with the traditional definitions based on capacity, the MAE/RMSE of new definitions for SOH and SOC are 0.59 %/0.78 % and 5.2 %/6.6 %, respectively.