Scholarlink

<p>Shear Assisted Processing and Extrusion (ShAPE), a solid-phase processing (SPP) route capable of directly recycling aluminum (Al) alloy scraps, exhibits a potential of tolerating higher impurity levels than conventional melt-based recycling processes. This capability primarily arises from the effective fragmentation and redistribution of impurities induced by intensive stirring and shear plastic deformation. In this work, tubes with a gradient composition were extruded via ShAPE from 6063 Al machining chips blended with iron (Fe) powder at weight fractions of 0.2–1.2 wt%, 0.2–4.7 wt%, and 0.2–9.2 wt%. The effects of Fe concentration, die rotational speed, and plunge velocity on processing responses, including torque and force, were evaluated using three machine learning regression models: gradient boosting, random forest, and neural network. Model performance was assessed using the coefficient of determination and root mean squared error, followed by feature importance and Shapley Additive Explanations analysis to identify the dominant variables and interpret their influence. The Fe particle diameters and the particle counts were quantified from micrographs of the extruded tubes. The results show that Fe concentration and die plunge velocity govern the steady-state torque and force, while rotational speed has a minor effect on force in the explored parametric space. Fe concentration also significantly influences impurity fragmentation, evidenced by a ~ 24% reduction in particle diameter interquartile range from 1 wt% to 4.5 wt% Fe and a further ~ 19% reduction at 9 wt%, alongside a peak in median particle count at 4.5 wt% Fe.</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

Data-driven analysis of impurity effects on solid-phase recycling of aluminum alloys

  • 期刊论文

  • 发表日期：

    2026年5月14日

查看全文
  • Akash Mukhopadhyay,
  • Xiao Li,
  • Shivakant Shukla,
  • Glenn J. Grant,
  • Tianhao Wang

摘要

Shear Assisted Processing and Extrusion (ShAPE), a solid-phase processing (SPP) route capable of directly recycling aluminum (Al) alloy scraps, exhibits a potential of tolerating higher impurity levels than conventional melt-based recycling processes. This capability primarily arises from the effective fragmentation and redistribution of impurities induced by intensive stirring and shear plastic deformation. In this work, tubes with a gradient composition were extruded via ShAPE from 6063 Al machining chips blended with iron (Fe) powder at weight fractions of 0.2–1.2 wt%, 0.2–4.7 wt%, and 0.2–9.2 wt%. The effects of Fe concentration, die rotational speed, and plunge velocity on processing responses, including torque and force, were evaluated using three machine learning regression models: gradient boosting, random forest, and neural network. Model performance was assessed using the coefficient of determination and root mean squared error, followed by feature importance and Shapley Additive Explanations analysis to identify the dominant variables and interpret their influence. The Fe particle diameters and the particle counts were quantified from micrographs of the extruded tubes. The results show that Fe concentration and die plunge velocity govern the steady-state torque and force, while rotational speed has a minor effect on force in the explored parametric space. Fe concentration also significantly influences impurity fragmentation, evidenced by a ~ 24% reduction in particle diameter interquartile range from 1 wt% to 4.5 wt% Fe and a further ~ 19% reduction at 9 wt%, alongside a peak in median particle count at 4.5 wt% Fe.