Experimental Validation and Cross‐Scale Evaluation of Battery Capacity Degradation Techniques for Electric Vehicles

ABSTRACT Electric vehicles are key to reducing emissions and promoting sustainable mobility. In electric vehicles, accurate capacity degradation prediction is essential to ensure reliability, extend battery life, and reduce range anxiety under real‐world conditions. Accurate prediction of battery capacity degradation remains a challenging task due to nonlinear aging behavior, variability in operating conditions, and the complex electrochemical processes involved in lithium‐ion batteries. This research presents a unified, data‐driven approach for forecasting capacity degradation at both the pack and cell levels using experimental data from a 26 Ah two‐wheeler battery pack and its 18 650 cells, tested under identical thermal and electrical conditions. Three data driven techniques were evaluated which include machine learning, statistical learning and reliability‐based approaches. Each model used the same input experimental data of both cell and pack level testing. The models were implemented in MATLAB software, and their performance was compared using standardized evaluation metrics including RMSE, MAE, and R 2 . These challenges necessitate robust and reliable modeling approaches for effective battery health monitoring. Model‐specific novel architectures included SVR with an RBF kernel and optimized hyperparameters, GPR with a squared exponential kernel and Bayesian tuning, ANN with a feedforward network of 15 hidden neurons trained using the Levenberg–Marquardt algorithm, Polynomial Regression with second‐degree terms and ridge regularization ( λ = 0.1) and Weibull fitting with nonlinear least squares. Among all, SVR achieved the best performance ( R 2 = 0.9995 pack level, 0.9988 cell level) followed closely by GPR and ANN. Reliability models showed reduced accuracy at the cell and pack scale. These findings represent the potential of the proposed data‐driven framework for integration into real‐time Battery Management Systems (BMS), enabling accurate state‐of‐health monitoring and reducing range uncertainty in electric vehicle applications. These challenges necessitate robust and reliable modeling approaches for effective battery health monitoring.