A fast sorting method for retired power batteries based on charging curve characteristics
Accurate and rapid sorting is crucialin the echelon utilization of retired power batteries.The charging curve and capacity of retired power batteries are obtained by charging and discharging test.The grey correlation analysis method is employed to determine the voltage interval with the best capacity correlation.Based on the battery aging mechanism, the charging capacity ΔQ, charging time T, main peak center voltage V1 and the ratio of charging capacity to interval voltage K corresponding to the optimal voltage interval are extracted as the characteristic parameters to characterize the inconsistency of the battery.The local outlier factor algorithm is employed to screen the abnormal aging batteries while the K-means clustering algorithm is adopted to complete the sorting of retired batteries.Meanwhile, a static and dynamic two-dimensional index system is proposed to evaluate the sorting consistency, and two sets of charge and discharge data of decommissioned batteries are used for verification.Our experimental results show the battery's static consistency is increased by 55％ and its dynamic consistency by 82％ after sorting, and the average test time of a single battery is reduced to 30 minutes.Compared with the K-means clustering algorithm, the static and dynamic consistency of sorting is increased by 50％ and 33％respectively after fusing the local outlier factor algorithm.Compared with the capacity increment method and the static parameter sorting method, the static consistency of our method is up by 28％ and 5％respectively, and the dynamic consistency jumps by 76％ and 61％ respectively.
retired power batteryconsistencyfast sortingK-meansLOF
国家科技期刊平台
NSTL
万方数据
