<p>The miniaturization of electronic devices necessitates the development of effective thermal management technology. This study computationally analyzed the influence of staggered delta rib arrangements on the thermal performance and flow behavior of microchannel heat sink within the ranges of attack angle (<i>θ</i> = 22.5°, 30°, 37.5°, 45°, 52.5°, and 60°) and Reynolds number (<i>Re</i> = 100, 400, and 700). As the distance between the point of flow separation from the delta ribs and the next rib increases, the flow zigzags between the ribs. As a result, the a-type exhibited the best thermal-hydraulic performance among the three different types. At <i>Re</i> = 700, for the a-type, as θ increases, the heat transfer rate also increases, but pressure drop also rises, leading to the highest performance evaluation criterion (<i>PEC</i>) of 1.3018 at <i>θ</i> = 30°. The optimal geometric parameters derived from the artificial neural network model reached a <i>PEC</i> of 1.3030 at <i>θ</i> = 27.7° and <i>Re</i> = 700 for the a-type, showing a high accuracy with an error of 0.0307 % compared with the CFD re-simulation. These results provide significant advantages not only for this study but also for the performance optimization of various thermal management systems.</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

Parameter optimization of staggered delta rib arrangements in microchannel heat sink based on artificial neural networks

  • Mun Su Lee,
  • Jeong Geun Gwon,
  • Seong Hwan Ahn,
  • Young Min Seo,
  • Seokho Kim,
  • Hoon Ki Choi,
  • Yong Gap Park

摘要

The miniaturization of electronic devices necessitates the development of effective thermal management technology. This study computationally analyzed the influence of staggered delta rib arrangements on the thermal performance and flow behavior of microchannel heat sink within the ranges of attack angle (θ = 22.5°, 30°, 37.5°, 45°, 52.5°, and 60°) and Reynolds number (Re = 100, 400, and 700). As the distance between the point of flow separation from the delta ribs and the next rib increases, the flow zigzags between the ribs. As a result, the a-type exhibited the best thermal-hydraulic performance among the three different types. At Re = 700, for the a-type, as θ increases, the heat transfer rate also increases, but pressure drop also rises, leading to the highest performance evaluation criterion (PEC) of 1.3018 at θ = 30°. The optimal geometric parameters derived from the artificial neural network model reached a PEC of 1.3030 at θ = 27.7° and Re = 700 for the a-type, showing a high accuracy with an error of 0.0307 % compared with the CFD re-simulation. These results provide significant advantages not only for this study but also for the performance optimization of various thermal management systems.