Scholarlink

<p>This paper introduces a robust and scalable system designed to assess and forecast the health of lithium-ion batteries, commonly used in electric vehicles. The system evaluates battery cells based on key metrics such as internal resistance (IR), open-circuit voltage (OCV), actual capacity (Ah), discharge rate (DR), and cycle life. It categorizes cells into performance grades using supervised machine learning techniques like Logistic Regression, Decision Tree, Support Vector Machine (SVM), Naive Bayes, and Random Forest. Moreover, the machine learning models are used to estimate the State of Charge (SoC) and remaining cycle life, facilitating real-time diagnostics and predictive maintenance. The framework considers temperature-induced noise to improve realistic optimization and cost-efficiency within existing battery management systems. Suitable for both new and second-life batteries, it enhances safety, reliability, and operational lifespan.</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

A robust machine learning-based system for battery grading and lifecycle prediction for electric vehicle applications

  • 期刊论文

  • 发表日期：

    2025年8月29日

查看全文
  • Appalabathula Venkatesh,
  • J. Vijaya Kumar,
  • S. Phani Kumar,
  • S. S. Kiran,
  • K. Ashwini,
  • L. Vijay,
  • B. Muralidhar

摘要

This paper introduces a robust and scalable system designed to assess and forecast the health of lithium-ion batteries, commonly used in electric vehicles. The system evaluates battery cells based on key metrics such as internal resistance (IR), open-circuit voltage (OCV), actual capacity (Ah), discharge rate (DR), and cycle life. It categorizes cells into performance grades using supervised machine learning techniques like Logistic Regression, Decision Tree, Support Vector Machine (SVM), Naive Bayes, and Random Forest. Moreover, the machine learning models are used to estimate the State of Charge (SoC) and remaining cycle life, facilitating real-time diagnostics and predictive maintenance. The framework considers temperature-induced noise to improve realistic optimization and cost-efficiency within existing battery management systems. Suitable for both new and second-life batteries, it enhances safety, reliability, and operational lifespan.