Scholarlink

<p>The experimental and ANN approaches were employed in this paper to explore the influence of stir casting parameters on the mechanical properties of ZrO₂ AZ91 magnesium composites. The AZ91–ZrO₂ composites were prepared according to the Taguchi L16 orthogonal array by changing the melt temperature, the content of reinforcement, stirring speed and time of stirring. The highest UTS of 222&#xa0;MPa and hardness of 81.68 Hv were achieved at a melt temperature of 650&#xa0;°C, stirring speed of 550&#xa0;rpm, wt % ZrO₂ of 6 wt % and stirring time of 10&#xa0;min. According to the ANOVA results, two significant factors affecting UTS are melted temperature (19%) and stirring speed (71%). The ANN model, developed using 16 datasets and divided into training, validation, and testing sets in a 70/15/15 ratio, demonstrated excellent predictive accuracy with an overall correlation coefficient of <i>R</i> = 0.9966 and minimal prediction error, confirming its suitability for regression analysis and process optimization. The enhanced mechanical properties were attributed to the effects of improved interfacial bonding, dislocation strengthening, load transfer and grain refinement. The findings demonstrate that ANN-assisted optimization can predict the optimum process conditions and simultaneously minimize experiment time for high-performance AZ91/ZrO2 composites.</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

Investigating stir casting process parameters for ZrO₂ reinforced magnesium alloy using backpropagation algorithm

  • 期刊论文

  • 发表日期：

    2026年3月17日

查看全文
  • E. M. Shaban,
  • A. R. Saravanan,
  • Vittel Rao Rajendranrao,
  • B. Muthuvel,
  • D. R. Rajkumar,
  • S. Revathi

摘要

The experimental and ANN approaches were employed in this paper to explore the influence of stir casting parameters on the mechanical properties of ZrO₂ AZ91 magnesium composites. The AZ91–ZrO₂ composites were prepared according to the Taguchi L16 orthogonal array by changing the melt temperature, the content of reinforcement, stirring speed and time of stirring. The highest UTS of 222 MPa and hardness of 81.68 Hv were achieved at a melt temperature of 650 °C, stirring speed of 550 rpm, wt % ZrO₂ of 6 wt % and stirring time of 10 min. According to the ANOVA results, two significant factors affecting UTS are melted temperature (19%) and stirring speed (71%). The ANN model, developed using 16 datasets and divided into training, validation, and testing sets in a 70/15/15 ratio, demonstrated excellent predictive accuracy with an overall correlation coefficient of R = 0.9966 and minimal prediction error, confirming its suitability for regression analysis and process optimization. The enhanced mechanical properties were attributed to the effects of improved interfacial bonding, dislocation strengthening, load transfer and grain refinement. The findings demonstrate that ANN-assisted optimization can predict the optimum process conditions and simultaneously minimize experiment time for high-performance AZ91/ZrO2 composites.