Scholarlink

The adoption of electric vehicles (EVs) powered by lithium batteries has significantly impacted both environmental and economic aspects of transportation. Ensuring safety in EVs is crucial, which relies on accurately estimating the state of health (SOH) of lithium batteries. This study presents an SOH estimation method using a machine learning-based decision tree. The supervised decision tree model utilizes battery capacity, cycle count, and SOH as inputs to predict battery health. The results demonstrate a remarkable estimation accuracy of 99%, despite the nonlinear behavior of SOH caused by internal resistance variations. This high accuracy makes the method suitable for real-time applications, contributing to improved battery management and enhanced EV safety.

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

Enhancing EV Battery Safety: SOH Estimation with Machine Learning

  • 章节

  • 发表日期：

    2026年

查看全文
  • M. Arulmozhi

摘要

The adoption of electric vehicles (EVs) powered by lithium batteries has significantly impacted both environmental and economic aspects of transportation. Ensuring safety in EVs is crucial, which relies on accurately estimating the state of health (SOH) of lithium batteries. This study presents an SOH estimation method using a machine learning-based decision tree. The supervised decision tree model utilizes battery capacity, cycle count, and SOH as inputs to predict battery health. The results demonstrate a remarkable estimation accuracy of 99%, despite the nonlinear behavior of SOH caused by internal resistance variations. This high accuracy makes the method suitable for real-time applications, contributing to improved battery management and enhanced EV safety.