A Study on Rapid Capacity Diagnosis of Used Batteries Using Partial Discharge Data and Polynomial Modeling

The demand for second-life utilization of used batteries (UBs) is rapidly increasing with the increasing adoption of electric vehicles and emphasis on circular economy policies.As of 2025, >3,500 UBs have accumulated domestically, with projections reaching several hundred thousand units by 2030, highlighting the requirement for efficient and scalable performance evaluation techniques for assessing the residual value of UBs.Conventional capacity diagnosis methods such as full-cycle testing in certification systems like Korea Certification require >180 min per unit, imposing significant time and resource burdens.To address this challenge, this study investigated a rapid diagnostic methodology that estimates capacity using partial discharge data and open-circuit voltage measurements.A high-order polynomial regression using the least-squares method was applied and validated through experiments on 20 second-life battery modules.The proposed model achieved a capacity estimation error within 2% compared with reference measurements, demonstrating reliable accuracy without requiring full charge-discharge cycles.This method enabled diagnosis within approximately 20 min under C/3 discharge conditions.This approach enhances the practicality of large-scale UB screening, supports efficient recycling and reuse operations, and provides a technological foundation for advancing second-life battery applications in energy storage and mobility sectors.