摘要
首先利用熵权法对退役电池各特征参数进行权重计算,依据计算结果选取电池剩余容量、开路电压和放电直流等效内阻作为电池聚类的分选因子。其次利用支持向量机算法预测退役电池剩余容量,最后利用K-均值聚类算法将电池分成4种等级,并通过实验证明了该方法的准确性。
In this paper,the entropy weight method is used to calculate the weight of each characteristic parameter of the retired battery firstly.According to the calculation result,the remaining capacity,open circuit voltage and discharge DC equivalent internal resistance of the battery are selected as the sorting factors of battery clustering.Secondly,the support vector machine algorithm is used to predict the remaining capacity of the retired battery.Finally,the K-means clustering algorithm is used to divide the battery into four levels,and the accuracy of the method is proved through experiments.
作者
李建林
王哲
李雅欣
刘若桐
闫湖
黄碧斌
Li Jianlin;Wang Zhe;Li Yaxin;Liu Ruotong;Yan Hu;Huang Bibin(Energy Storage Technology Engineering Research Center,North China University of Technology,Beijing 100144,China;State Grid Energy Research Institute Co.,Ltd.,Beijing 102209,China)
出处
《太阳能学报》
EI
CAS
CSCD
北大核心
2023年第1期418-425,共8页
Acta Energiae Solaris Sinica
基金
北京市自然科学基金(KZ202110009014)
国家电网有限公司总部科技项目资助:《面向电网企业的退役动力电池梯次利用辅助决策关键技术研究及软件开发》(项目编号:5419-201957216A-0-0-00)。
关键词
退役动力电池
电池分选
熵权法
容量预测
支持向量机
K-均值聚类算法
retired power battery
battery sorting
entropy weight method
capacity prediction
support vector machine
K-means clustering algorithm
