Scholarlink

<p>Carbon steel is widely used in automotive structural components due to its favorable mechanical properties and cost-effectiveness. However, its susceptibility to corrosion in complex service environments undermines vehicle safety, durability, and esthetics, leading to significant economic losses annually. Traditional corrosion assessment methods, such as salt spray and electrochemical tests, are limited in replicating real-world conditions and fail to capture the nonlinear interactions among multiple environmental factors. To overcome these limitations, this study introduces a corrosion big data approach integrating in situ monitoring, IoT sensing, and cloud computing. Multi-dimensional data including environmental parameters, material-state signals, and operational profiles are collected and fused. Using machine learning techniques such as random forest and deep learning, a dynamic corrosion–environment map is constructed, multi-factor coupling mechanisms are analyzed, and a high-accuracy corrosion rate prediction model is developed. Focusing on Q235 carbon steel, this work aims to establish a corrosion big data platform that reveals corrosion dynamics under realistic service conditions. The outcomes provide technical support for developing corrosion-resistant automotive steels, designing intelligent protection systems, and extending vehicle service life.</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

Atmospheric Corrosion Research of Carbon Steel for Automobile Based on Corrosion Big Data Technology

  • 期刊论文

  • 发表日期：

    2026年2月6日

查看全文
  • Zhang Junhua,
  • Yuan Rui,
  • Zhang Xujie,
  • Tang Yu,
  • Yang Xiaojia

摘要

Carbon steel is widely used in automotive structural components due to its favorable mechanical properties and cost-effectiveness. However, its susceptibility to corrosion in complex service environments undermines vehicle safety, durability, and esthetics, leading to significant economic losses annually. Traditional corrosion assessment methods, such as salt spray and electrochemical tests, are limited in replicating real-world conditions and fail to capture the nonlinear interactions among multiple environmental factors. To overcome these limitations, this study introduces a corrosion big data approach integrating in situ monitoring, IoT sensing, and cloud computing. Multi-dimensional data including environmental parameters, material-state signals, and operational profiles are collected and fused. Using machine learning techniques such as random forest and deep learning, a dynamic corrosion–environment map is constructed, multi-factor coupling mechanisms are analyzed, and a high-accuracy corrosion rate prediction model is developed. Focusing on Q235 carbon steel, this work aims to establish a corrosion big data platform that reveals corrosion dynamics under realistic service conditions. The outcomes provide technical support for developing corrosion-resistant automotive steels, designing intelligent protection systems, and extending vehicle service life.