Scholarlink

<p>Inhomogeneity in battery pack unit cells significantly impacts their lifetime, performance, and safety. Consistent electrochemical characteristics are essential to mitigate this issue, and minimizing manufacturing tolerances is critical. This study introduces a novel approach that integrates multi-parameter screening with k-means clustering to enhance battery pack design. In contrast to traditional single-parameter sorting, which limits precision and increases performance inconsistencies, the proposed method considers a wider range of factors, including open-circuit voltage, voltage change, capacity, and partial capacity, ensuring more balanced and consistent performance while effectively addressing outliers. An analysis of 800 commercial batteries revealed high correlation factors via statistical methods. Multiple regression analysis was used to quantify their effects, and a clustering algorithm was employed to group cells with similar characteristics. The effectiveness of the proposed method was validated through electrical characteristic tests by comparing the designed battery packs.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

High-accuracy cell discrimination based on multiple regression combined with k-means clustering algorithm for lithium-ion rechargeable cells

  • 期刊论文

  • 发表日期：

    2025年2月21日

查看全文
  • Pyeong-Yeon Lee,
  • Dongho Han,
  • Jonghoon Kim

摘要

Inhomogeneity in battery pack unit cells significantly impacts their lifetime, performance, and safety. Consistent electrochemical characteristics are essential to mitigate this issue, and minimizing manufacturing tolerances is critical. This study introduces a novel approach that integrates multi-parameter screening with k-means clustering to enhance battery pack design. In contrast to traditional single-parameter sorting, which limits precision and increases performance inconsistencies, the proposed method considers a wider range of factors, including open-circuit voltage, voltage change, capacity, and partial capacity, ensuring more balanced and consistent performance while effectively addressing outliers. An analysis of 800 commercial batteries revealed high correlation factors via statistical methods. Multiple regression analysis was used to quantify their effects, and a clustering algorithm was employed to group cells with similar characteristics. The effectiveness of the proposed method was validated through electrical characteristic tests by comparing the designed battery packs.