<![CDATA[Degradation analysis of lithium-ion batteries is a critical aspect of understanding the aging behaviors and reliability of energy storage systems. Although purely data-driven approaches are widely utilized for their high short-term predictive accuracy, they often function as black-box models that fail to capture the underlying physicochemical mechanisms of capacity fade under dynamic operational conditions. To address these limitations, this study uses a Generalized Eyring-Weibull model for lithium-ion battery degradation analysis under multi-stressor conditions. The proposed analytical approach explicitly integrates three key operational stress variables simultaneously: temperature, State of Charge (SoC), and discharge current (C-rate). Model validation was performed using the NASA Prognostics Center of Excellence (PCoE) battery dataset across distinct discharge profiles (2A and 4A). Physical parameters—including activation energy (Ea) and stress exponents—were extracted via L-BFGS-B optimization, while the long-term capacity fade trajectory was tracked using Miner’s Rule. The evaluation results demonstrate that this generalized physics-based model accurately maps actual degradation trends, yielding high R2 values on State of Health (SoH) tracking. Furthermore, the model showcases superior robustness in capturing accelerated aging from high current loads using a single universal equation without requiring separate data retraining. By bridging empirical data with fundamental kinetic principles, this study provides an interpretable and robust analytical methodology for advanced predictive maintenance strategies.]]>
