A fast estimation method for state-of-health of retired batteries based on health features

The disposal of retired batteries has become a major concern with the development of the new energy industry. Traditional methods estimate the state of health (SOH) of batteries mainly by collecting the characteristic parameters at specific state of charge (SOC); however, the SOC of the retired battery at the time of testing is unknown. Moreover, it is difficult to accurately obtain the SOC of the battery without knowing its capacity. This paper proposes a novel sorting strategy for retired batteries that can effectively reduce the sorting time of retired batteries with unknown SOC. Furthermore, this paper introduces the concept of resistive remaining capacity (RRC) and the cross-validation strategy of SOH, as the traditional assessment definition cannot fully describe the true SOH of the battery. The cross-validation strategy can evaluate the matching of internal resistance and remaining capacity. By combining the extracted feature parameters of retired batteries, the proposed methods in this paper can save the discharging time compared to the ampere-hour integration method of SOH acquisition. In addition, the method in this paper provides a more comprehensive assessment of retired batteries than the traditional method. The error in estimating the SOH of 11 retired batteries is less than 6 %.

Automated summary

The disposal of retired batteries is a major concern in the new energy industry. This paper proposes a novel sorting strategy for retired batteries with unknown state of charge, reducing the sorting time. It also introduces the concept of resistive remaining capacity and cross-validation strategy for a more comprehensive assessment of retired batteries.