State of Charge Estimation for Lithium-Ion Batteries NCR 21700 Using Linear Methods

In recent years, the adoption of electric vehicles (EVs) has significantly increased, leading to growing attention toward battery management systems. One of the most widely used battery cells in EVs is the NCR 21700. Accurate estimation of the State of Charge (SoC) is crucial for ensuring battery safety, longevity, and performance. Therefore, developing reliable SoC estimation methods has become an urgent research priority. This study aims to compare the performance of two machine learning algorithms–Extreme Gradient Boosting (XGBoost) and Linear Regression–for SoC estimation of the NCR 21700 battery cell. The dataset used includes voltage, current, and temperature values obtained from standard battery testing. Both models were trained and evaluated using the same dataset to ensure a fair comparison. The performance of each method was assessed based on the Mean Absolute Error (MAE). The results show that the XGBoost model achieved an MAE of 0.48 and a percentage MAE of 0.96%, whereas the Linear Regression model produced a higher MAE of 4.13 and a percentage MAE of 8.24%. These findings demonstrate that the XGBoost algorithm provides a significantly more accurate estimation of the SoC compared to Linear Regression. In conclusion, the XGBoost method is a more effective approach for SoC estimation in modern battery systems, particularly for EV applications utilizing the NCR 21700 cell.