Scholarlink

<p>This study investigated the effects of three powder metallurgy parameters—compaction pressure (CP), sintering temperature (ST), and sintering time (St)—on the properties of pure magnesium. The relationships between these input parameters and the responses (density, porosity, and hardness) were modeled and analyzed. The influence of each parameter was quantified using experimental results fitted to&#xa0;response surface methodology (RSM)&#xa0;and&#xa0;artificial neural network (ANN)&#xa0;models. Analysis of variance (ANOVA) was used to identify significant parameters and validate the models. The optimal parameters, determined by the desirability function, were a CP of 650&#xa0;MPa, an ST of 525&#xa0;°C, and an St of 30 minutes. The ANN model demonstrated superior accuracy, achieving a correlation of approximately 99% with experimental data.&#xa0;The close agreement between the predicted R<sup>2</sup> and adjusted R<sup>2</sup> values (difference &lt; 0.2) confirms the model’s robustness and high predictive accuracy.&#xa0;The corrosion behavior of the optimized samples was examined in 0.25&#xa0;M HCl and half-strength Ringer’s solution (HSR) using potentiodynamic and cyclic polarization methods. Post-corrosion surface analysis was performed using field emission scanning electron microscopy (FESEM). The findings showed that the corrosion rate in Ringer’s solution was significantly higher than that in the acidic solution.&#xa0;This results from the development of a more complex and less protective corrosion product layer in the simulated physiological environment, along with the catalytic influence of bicarbonate (HCO<sub>3</sub><sup>−</sup>) and chloride (Cl<sup>−</sup>) ions in destabilizing the magnesium oxide/hydroxide film.</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

Experimental Investigation and Optimization of Powder Metallurgy Processes Using Response Surface Methodology and Artificial Neural Networks, and Corrosion Behavior of Pure Magnesium

  • 期刊论文

  • 发表日期：

    2026年1月27日

查看全文
  • Mohammed Ibrahim Fathi,
  • Borhen Louhichi,
  • Mohammed Ali Terres,
  • Jawdat Ali Yagoob

摘要

This study investigated the effects of three powder metallurgy parameters—compaction pressure (CP), sintering temperature (ST), and sintering time (St)—on the properties of pure magnesium. The relationships between these input parameters and the responses (density, porosity, and hardness) were modeled and analyzed. The influence of each parameter was quantified using experimental results fitted to response surface methodology (RSM) and artificial neural network (ANN) models. Analysis of variance (ANOVA) was used to identify significant parameters and validate the models. The optimal parameters, determined by the desirability function, were a CP of 650 MPa, an ST of 525 °C, and an St of 30 minutes. The ANN model demonstrated superior accuracy, achieving a correlation of approximately 99% with experimental data. The close agreement between the predicted R2 and adjusted R2 values (difference < 0.2) confirms the model’s robustness and high predictive accuracy. The corrosion behavior of the optimized samples was examined in 0.25 M HCl and half-strength Ringer’s solution (HSR) using potentiodynamic and cyclic polarization methods. Post-corrosion surface analysis was performed using field emission scanning electron microscopy (FESEM). The findings showed that the corrosion rate in Ringer’s solution was significantly higher than that in the acidic solution. This results from the development of a more complex and less protective corrosion product layer in the simulated physiological environment, along with the catalytic influence of bicarbonate (HCO3−) and chloride (Cl−) ions in destabilizing the magnesium oxide/hydroxide film.