Enabling Practicality of Sustainable EV Battery Reuse

The growing number of retired electric vehicle (EV) batteries, often retaining around 80% of their initial capacity, presents opportunities for reuse to reduce resource consumption and environmental impact. This study proposes a machine learning–based diagnostic method that estimates battery static capacity using 3‐min partial charge and discharge data under 1 and 0.5C conditions. The model achieved root mean square errors (RMSEs) of 75.224 mAh (1.881%) and 84.508 mAh (2.113%) for discharge, and 37.159 mAh (0.929%) and 35.315 mAh (0.883%) for charge data, respectively. Compared to full discharge diagnostics, the proposed method is approximately 60 times faster while maintaining high accuracy. This enables near real‐time classification of batteries for reuse or recycling, improving operational efficiency, and reducing safety risks during storage. The approach offers a scalable, cost‐effective solution for battery screening, and supports sustainable resource management. It provides a practical foundation for circular battery systems and contributes to broader goals in energy efficiency and environmental conservation.