Scholarlink

<p>This study presents an integrated experimental and predictive investigation of friction stir welded (FSW) AA5052–AA6061 joints fabricated using a taper triangular tool. While FSW of aluminium alloys is well established, limited studies have systematically examined the combined influence of alloy pairing (similar and dissimilar joints) and tool geometry within a unified framework. In this work, three joint configurations (5052–5052, 6061–6061, and 5052–6061) were fabricated under controlled conditions and evaluated in terms of tensile strength, hardness, elongation, and microstructural characteristics. The results indicate that homogeneous joints exhibit superior mechanical performance due to improved material flow and grain refinement, whereas dissimilar joints show reduced strength and hardness due to interfacial segregation and incomplete bonding. The 6061–6061 joint demonstrated the highest strength (92.5&#xa0;MPa) and hardness (74.41 HBW), while the 5052–6061 joint exhibited the lowest performance. To extend beyond experimental observations, regression-based machine learning models were employed to capture non-linear process–property relationships. Among the models, support vector regression showed the best balance between accuracy and generalization, while polynomial regression indicated overfitting tendencies for limited datasets. The study establishes a process–structure–property relationship and provides a design-oriented framework for predicting weld performance and selecting suitable alloy combinations in lightweight structural applications.</p> Graphical abstract <p></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

Interactive design-oriented prediction of mechanical performance in friction stir welded AA5052–AA6061 joints

  • 期刊论文

  • 发表日期：

    2026年6月18日

查看全文
  • Y. B. Vrushali,
  • M. Rajanish,
  • G. S. Divya,
  • S. Chethan,
  • Y. A. Reena,
  • M. Jayashree,
  • A. Srinivaas,
  • Yuvaraja Naik,
  • H. B. Dileep,
  • Arunkumar,
  • L. C. Shashidhar,
  • Y. P. Ravitej

摘要

This study presents an integrated experimental and predictive investigation of friction stir welded (FSW) AA5052–AA6061 joints fabricated using a taper triangular tool. While FSW of aluminium alloys is well established, limited studies have systematically examined the combined influence of alloy pairing (similar and dissimilar joints) and tool geometry within a unified framework. In this work, three joint configurations (5052–5052, 6061–6061, and 5052–6061) were fabricated under controlled conditions and evaluated in terms of tensile strength, hardness, elongation, and microstructural characteristics. The results indicate that homogeneous joints exhibit superior mechanical performance due to improved material flow and grain refinement, whereas dissimilar joints show reduced strength and hardness due to interfacial segregation and incomplete bonding. The 6061–6061 joint demonstrated the highest strength (92.5 MPa) and hardness (74.41 HBW), while the 5052–6061 joint exhibited the lowest performance. To extend beyond experimental observations, regression-based machine learning models were employed to capture non-linear process–property relationships. Among the models, support vector regression showed the best balance between accuracy and generalization, while polynomial regression indicated overfitting tendencies for limited datasets. The study establishes a process–structure–property relationship and provides a design-oriented framework for predicting weld performance and selecting suitable alloy combinations in lightweight structural applications.

Graphical abstract