基于因子分析与K-means聚类的退役动力电池快速分选方法

针对目前退役动力电池数量多、快速分选方法匮乏的问题,提出一种基于脉冲功率测试(Hybrid PulsePower Characteristic,HPPC)、因子分析和聚类算法的退役动力电池快速分选与重组方法。根据电池管理系统(Battery Management System, BMS)记录的电池数据,计算单体电池电压数据得到电池最大可用容量。以HPPC一次放电脉冲提取的电池开路电压、欧姆内阻、极化电阻以及浓差电阻作为特征变量。特征变量数据经归一化算法与因子分析优化后,通过聚类算法完成电池分选与重组。实验结果表明:该方法下单体电池平均分选重组时间压缩在30 min以内,分组后一致性指标较好,在退役动力电池分选与重组中具有较好的实际意义。

A fast classification method based on factor analysis and K-means clustering for retired electric vehicle batteries

In order to solve the problem of a large number of retired electric vehicle batteries with a lack of quick sorting means, a fast classification and regroup approach of batteries is presented based on Hybrid Pulse Power Characteristic (HPPC), factor analysis and a clustering algorithm. The maximum available capacity of the battery is calculated by the voltage data of the battery cell in the Battery Management System (BMS). The primary discharge pulse of the HPPC is used to extract the open-circuit voltage, ohmic, polarization and concentration resistances of the battery as characteristic variables. After the characteristic variable data is processed by the normalization algorithm and factor analysis, the battery sorting and recombination are completed by the clustering algorithm. Experimental results show that the average separation and recombination time of a single battery is compressed within 30 min using the proposed method, which has practical significance for classification and regrouping of retired electric vehicle batteries. This work is supported by the National Natural Science Foundation of China (No. 51607004), the Collaborative Innovation Project of Anhui Universities (No. GXXT-2019-002) and the Natural Science Research Project of Anhui University (No. KJ2020A0509).