State of Health Estimation for Lithium-Ion Batteries Based on Charging Curve and Machine Learning

Among the best crucial tasks of the BMS (battery management system) is estimating the state of health (SoH) which ensures the safe and reliable use of lithium-ion batteries. This paper shows an effective method for estimating the SOH of lithium ion batteries. In order to estimate SoH, a regression approach using the health factor and SoH is created through analyzing what battery's charging data analysis throughout the period of its service life. This method requires minimal learning and is easily applied in practical applications in chemistry or cell structure. Instead of using an entire charging voltage range, The time it takes to charge between two defined voltage levels is used to figure out the health parameter. Based on the regression model's goodness of fit, the voltage level is chosen for health prediction. The accuracy of the regression line is found to be over 99% when tested using a variety of test data.