Yildiran, NisanurDikmen, Ismail CanKaradag, Teoman2026-04-042026-04-042024979-833153149-2https://doi.org/10.1109/IDAP64064.2024.10711165https://hdl.handle.net/11616/1080648th International Artificial Intelligence and Data Processing Symposium, IDAP 2024 -- 21 September 2024 through 22 September 2024 -- Malatya -- 203423In this study, State of Health (SoH) estimation of lithium-ion batteries, which are integral to our daily life, was carried out. Experimental data were taken from lithium titanate oxide (LTO) batteries with 18650 geometry, depending on the cycle aging process performed in constant current-constant voltage mode in the battery analyzer; 2000 cycles of charge and discharge data were recorded. The data obtained were extracted and pre-processed. The extracted data were subjected to differential voltage analysis, one of the data-driven methods used for SoH analysis. Differential Voltage Analysis (DVA) is a highly efficient method for accurate and detailed battery aging characteristics analysis. What sets this study apart is the use of machine learning algorithms, specifically Linear Regression (LR), Support Vector Machine (SVM), and Gaussian Process Regression (GPR), in conjunction with DVA. SoH prediction is considered a regression problem, and the innovative use of machine learning algorithms in this context is a key aspect of this research. As a result of the regression analysis performed in Matlab, the methods with the highest accuracy were determined. The SoH prediction with the highest accuracy was made by linear regression, and the error rate was 1.5005 RMSE. All findings obtained were evaluated comparatively. © 2024 IEEE.eninfo:eu-repo/semantics/closedAccessBatteryLithium-IonMachine Learning AlgorithmRegressionState of HealthConference Object10.1109/IDAP64064.2024.107111652-s2.0-85207909540N/A