Abstract
The second-life utilization of retired batteries can not only extend their lifespan but also help alleviate energy crises and reduce environmental pollution. Technical challenges of battery reusing arise from the inherently low consistency of the retired batteries, especially when considering the lifespan. This paper proposes a meta-learning method for lifespan-based battery clustering. We first train a convolutional neural network that can classify batteries into different groups based on their lifespan, using data from the first three cycles only. Then, instead of directly predicting the lifespan, the well-trained network is utilized to predict if the lifespans of two batteries are similar, formulating an adjacent matrix. Finally, batteries with similar lifespans can be classified into the same group with the assistance of the adjacent matrix. In this way, the network trained with fresh cells can also be used to classify the retired batteries. Experimental results on 38 batteries indicate that our method can reduce the loss-of-lifespan by at least 20% compared with the cases of using the conventional capacity-resistance method to group batteries. This study demonstrates the potential of supervised learning in battery classification, offering new solutions to enhance the second-life utilization of retired batteries.
Original language | English |
---|---|
Title of host publication | Proceedings of 2024 IEEE 25th China Conference on System Simulation Technology and its Application, CCSSTA 2024 |
Publisher | Institute of Electrical and Electronics Engineers Inc. |
Pages | 591-594 |
Number of pages | 4 |
ISBN (Electronic) | 9798350366600, 9798350366594 |
ISBN (Print) | 9798350366617 |
DOIs | |
Publication status | Published - 21 Jul 2024 |
Event | 2024 IEEE 25th China Conference on System Simulation Technology and its Application (CCSSTA) - Tianjin, China, Tianjin, China Duration: 21 Jul 2024 → 23 Jul 2024 http://www.ccssta.org.cn/index |
Conference
Conference | 2024 IEEE 25th China Conference on System Simulation Technology and its Application (CCSSTA) |
---|---|
Abbreviated title | CCSSTA 2024 |
Country/Territory | China |
City | Tianjin |
Period | 21/07/24 → 23/07/24 |
Internet address |
Bibliographical note
Publisher Copyright:© 2024 IEEE.
Funding
This work is supported, in part, by the Research Seed Fund (103652) provided by LU Research Committee, National Natural Science Foundation of China (NSFC) under Grant numbers 52277223 and 51977131, and Shanghai Pujiang Programme (23PJD062).
Keywords
- battery aging
- battery screening
- classification
- second life usage