Premium
Data‐Driven Fast Clustering of Second‐Life Lithium‐Ion Battery: Mechanism and Algorithm
Author(s) -
Ran Aihua,
Zhou Zihao,
Chen Shuxiao,
Nie Pengbo,
Qian Kun,
Li Zhenlong,
Li Baohua,
Sun Hongbin,
Kang Feiyu,
Zhang Xuan,
Wei Guodan
Publication year - 2020
Publication title -
advanced theory and simulations
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.068
H-Index - 17
ISSN - 2513-0390
DOI - 10.1002/adts.202000109
Subject(s) - cluster analysis , sort , leverage (statistics) , computer science , battery (electricity) , throughput , lithium (medication) , relevance (law) , algorithm , machine learning , power (physics) , telecommunications , medicine , physics , quantum mechanics , wireless , information retrieval , endocrinology , political science , law
While electrical vehicles (EVs) are expanding rapidly and getting more and more popular in the market, researchers have started to leverage the remaining capacity of used or to‐be‐retired batteries for their second‐life applications. It is crucial to develop a fast and efficient technology to first sort them and then extend their life while delivering energy, waste reduction, and economic benefits. In this work, a pulse clustering model embedded with improved bisecting K‐means algorithm is developed to effectively sort retired batteries with life cycles ranging from new to an end‐of‐life state. The relevance of selected variables is rigorously validated, reaching the accuracy as high as 88% compared with the traditional full charge–discharge test. To note, the test time has largely reduced from hours to minutes. This data‐driven clustering modeling with fast pulse test is a promising approach for clustering lithium‐ion batteries, which is demonstrated with a home‐built and high throughput intelligent clustering machine. In general, the technology opens a new generation of battery clustering, improving the efficiency and accuracy over the past semiempirical approaches.