Application of Artificial Intelligence to State-of-Charge and State-of-Health Estimation of Calendar-Aged Lithium-Ion Pouch Cells

Abstract

Accurate State-of-Charge (SOC) and State-of-Health (SOH) estimation of lithium-ion batteries (LIBs) is essential for the battery management system (BMS). For the first time, a feed-forward artificial neural network (ANN) has been used to estimate SOC of calendar-aged lithium-ion pouch cells. Calendar life data has been generated by applying galvanostatic charge/discharge cycle loads at different storage temperature (35°C and 60°C) and conditions (fully-discharged and fully-charged). The data has been obtained at various C-rates for duration of 10 months at one-month intervals. In order to include LIB hysteresis effect, two separate ANNs have been trained for charge and discharge data. The ANN have achieved a Root Mean Square Error (RMSE) of 1.17% over discharge data and 1.81% over charge data, confirming the ability of the network to capture input-output dependency. The calendar-aged battery data at various degradation conditions has been employed to train a new ANN to estimate SOH. The ANN has shown RMSE of 1.67%, demonstrating the network capability to estimate SOH. This study highlights the importance of considering aging effects in SOC estimates and the ability of ANN to include these effects efficiently.