Estimation of lithium-ion battery state of health using an ensemble model integrated with random forest and linear regression techniques

The demand for efficient battery management systems, especially lithium-ion batteries in electric vehicles, has increased with the transition to renewable energy. This study presents an ensemble model that combines Random Forest and linear regression to accurately estimate the state of health (SoH) of batteries. Using NASA data, divided into training, validation, and test sets, the model showed an error margin below 1%. An outlier in battery B0007 suggests potential future improvements in the model's generalization and applicability.