改进SPBO优化BIRCH算法的退役动力电池等级划分  

Improved SPBO optimized BIRCH algorithm for retired power battery grade classification

作　　者：杨昌海 徐逸扬 杨婷婷 宋汶秦 王兴贵[2] YANG Changhai;XU Yiyang;YANG Tingting;SONG Wenqin;WANG Xinggui(Economic and Technical Research Institute of State Grid Gansu Electric Power Company,Lanzhou Gansu 730000,China;School of Electrical Engineering and Information Engineering,Lanzhou University of Technology,Lanzhou Gansu 730050,China)

机构地区：[1]国网甘肃省电力公司经济技术研究院,甘肃兰州730000 [2]兰州理工大学电气工程与信息工程学院,甘肃兰州730050

出　　处：《电源技术》2023年第4期469-473,共5页Chinese Journal of Power Sources

基　　金：国网甘肃省电力公司科技项目(W21FZ2730-244)。

摘　　要：针对退役动力电池存在数量庞大、一致性差、分类效率低等问题,将改进学生心理优化算法(SPBO)和BIRCH算法结合,对退役动力电池进行等级划分。基于BIRCH算法原理并结合退役动力电池参数,构建聚类特征树。利用改进后SPBO的优化特性,优化聚类特征树中节点的选取。分析仿真结果可知:与改进前相比,解决了聚类特征树中存在异常节点的问题;所提算法相对于传统k均值聚类,性能更为优良;使划分等级后的电池,拥有较高的一致性。Aiming at the problems of large number,poor consistency and low classification efficiency of retired power batteries,the improved student psychological optimization algorithm(SPBO)was combined with BIRCH algorithm to classify retired power batteries.Firstly,a clustering feature tree was constructed based on BIRCH algorithm and retired power battery parameters.Secondly,the selection of nodes in the clustering feature tree was optimized by using the optimization characteristics of the improved SPBO.The simulation results show that:compared with SPBO before improvement,the problem of abnormal nodes in the clustering feature tree is solved;the proposed algorithm has a better performance than traditional k-means clustering.The battery after the classification of has a higher consistency.

关 键 词：退役动力电池 SPBO BIRCH算法 电池等级划分 电池一致性 

分 类 号：TM912.9[电气工程—电力电子与电力传动]