期刊
CASE STUDIES IN THERMAL ENGINEERING
卷 45, 期 -, 页码 -出版社
ELSEVIER
DOI: 10.1016/j.csite.2023.102996
关键词
Annular microchannel heat sink; Relative efficiency index; Artificial neural network; Heat transfer management; Fin geometry
Thermal management of microelectronic circuits is a difficult challenge, but an artificial neural network can optimize the geometry of a finned-MCHS. The parameter t has a noticeable impact on the thermal and hydrodynamic properties of the device.
Thermal management of microelectronic circuits will be one of the most difficult challenges facing engineering processes in the near future. High operating temperatures in these devices can degrade the reliability of the components and reduce their life. Therefore, effective cooling technologies that can disperse the significant heat load from the surface of microelectronic equipment are required. An appropriate microchannel heat sink (MCHS) system with optimized geometry can be one of the reliable choices. In the current work, an artificial neural network (ANN) is exerted to optimize the geometry of a finned-MCHS. The distance of fins from the inlet in the second row (l), the distance of fins from the side walls in the first and third rows (t), and the angle of hexagons (theta) are the input parameters. According to the obtained results, the ANN model with a coefficient of determination of 0.999 performed well in predicting the Nusselt number (Nu) and pressure drop (Delta P). Among the investigated input parameters, the variations of the parameter of t affected the thermal and hydrodynamic properties of the device noticeably. Besides, the ANN model suggested that when the optimum values of input parameters (i.e., l = 7.636 mm, t = 4
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据