Scholarlink

<p>Shrinkage porosity severely limits the performance and yield of A356 aluminum sand castings. This study presents an integrated computational framework combining finite element method simulation, Taguchi design, and response surface methodology to systematically optimize gating systems and minimize this defect. Eleven parameters were initially screened, identifying pouring speed (45.0% contribution), pouring temperature (18.5%), and number of ingates (7.3%) as the most significant factors. A split-plot RSM design accounting for hierarchical factor structure (hard-to-change geometric versus easy-to-change process parameters) yielded a highly accurate predictive model (R² = 0.968, Q² = 0.941). Numerical optimization identified a robust parameter set: tapered sprue and runner, 2–3 ingates, 725&#xa0;°C pouring temperature, 50&#xa0;cm/s pouring speed, and 100&#xa0;mm riser diameter. Empirical validation confirmed shrinkage porosity of 5.15%, achieving 71.5% reduction from baseline. Validation was reinforced through quantitative thermal analysis from in-mold thermocouples (R² = 0.92) and microstructural evaluation confirming defect suppression. Uncertainty analysis demonstrates process robustness (Cpk = 1.96), with three months production data showing 60% increase in monthly profit and sub-one-week payback. The methodology provides a practical, data-driven tool enabling foundries to replace costly trial-and-error with science-based virtual design.</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

Identification of critical gating system parameters and development of a predictive model for shrinkage defects in A356 sand castings using FEM and response surface methodology

  • 期刊论文

  • 发表日期：

    2026年3月29日

查看全文
  • Aragaw Mulu Muhaba

摘要

Shrinkage porosity severely limits the performance and yield of A356 aluminum sand castings. This study presents an integrated computational framework combining finite element method simulation, Taguchi design, and response surface methodology to systematically optimize gating systems and minimize this defect. Eleven parameters were initially screened, identifying pouring speed (45.0% contribution), pouring temperature (18.5%), and number of ingates (7.3%) as the most significant factors. A split-plot RSM design accounting for hierarchical factor structure (hard-to-change geometric versus easy-to-change process parameters) yielded a highly accurate predictive model (R² = 0.968, Q² = 0.941). Numerical optimization identified a robust parameter set: tapered sprue and runner, 2–3 ingates, 725 °C pouring temperature, 50 cm/s pouring speed, and 100 mm riser diameter. Empirical validation confirmed shrinkage porosity of 5.15%, achieving 71.5% reduction from baseline. Validation was reinforced through quantitative thermal analysis from in-mold thermocouples (R² = 0.92) and microstructural evaluation confirming defect suppression. Uncertainty analysis demonstrates process robustness (Cpk = 1.96), with three months production data showing 60% increase in monthly profit and sub-one-week payback. The methodology provides a practical, data-driven tool enabling foundries to replace costly trial-and-error with science-based virtual design.