摘要
针对电池分选成组后的一致性问题,在分析电池容量衰减与放电电压曲线关系的基础上,采用模糊C均值聚类算法,提出一种基于放电电压平台的FCM电池分选方法。该方法选取单体电池放电电压平台的3个特征点作为样本,再将标准化后的样本集作为算法的分类对象,最后以聚类有效性函数判定最优分类结果。全寿命实验的结果表明,采用分选方法得到的电池组动态一致性好,循环寿命衰减率降低,在500次循环寿命测试后健康度仍保持在90%以上。所提出的分选方法分选效率高,适用于数量较大的电池样本,可有效识别组内电池样本的一致性和生产质量,实现电池的多场合利用和效率最大化。
To solve the consistency problem of batteries after being sorted into groups,on the basis of ana-lyzing the relation between battery capacity losses and discharge voltage and adopting fuzzy C-mean clustering algo-rithm ,a battery sorting scheme is proposed, taking advantage of the flatness of discharge voltage curve. With the scheme,three characteristic points on the flat segment of discharge voltage curve for each cell are chosen as samples and with normalized sample set as the sorting object of the algorithm,the optimal sorting results are finally judged by clustering validity function. The results of whole-life experiment show that the battery pack sorted by the scheme pro-posed has good dynamic consistency and lower cycle life decay rate,with its SOH remains above 9 0 % after 500 cy-cles of life tests. The battery sorting scheme proposed has high sorting efficiency and is suitable for large quantities of battery samples,and can effectively identify the consistency and production qualities of battery samples in a pack, achieving various occasion utilization and efficiency maximization of batteries.
出处
《汽车工程》
EI
CSCD
北大核心
2017年第8期864-869,878,共7页
Automotive Engineering
基金
国家重点研发计划(2016YFC0300100)资助
关键词
磷酸铁锂电池
模糊C均值聚类
放电电压平台
一致性
健康度
lithium iron phosphate battery
fuzzy C-means clustering
flat segment of discharge voltage curve
consistency
state of health
