An SVM-based health classifier for offline Li-ion batteries by using EIS technology

This paper presents an offline testing framework and simulation to measure the aging situation of Li-ion batteries within the battery management system or laddering use for maintenance activities. It presents the use of electrochemical impedance spectroscopy as a non-destructive inspection method to detect battery states. Multiple cycles (charge and discharge) were done to gain EIS results in different conditions, such as temperature. Results were captured and digitalised through a suitable circuit model and mathematical methods for fitting. The state of health values were calibrated, and data were reshaped as vectors and then used as input for support vector machine. These data were then used to create a machine learning model and analyse the aging mechanism of lithium-ion batteries. The machine learning model is established, and the decision boundaries are visualised in 2D graphs. The accuracy of these machine learning models can reach 80% in the test cases, and good fitting in lifetime tracking. The framework allows more reliable SOH estimation in electric vehicles and more efficient maintenance or laddering operations.