Multi-objective optimization of a microchannel heat sink combining cavities and longitudinal vortex generators based on CFD and NSGA-II genetic algorithm

Abstract

Microchannel heat sink (MCHS) is an innovative cooling device for high-powered integrated circuits. Cavities and longitudinal vortex generators (LVGs) are two kinds of effective heat transfer enhancement techniques for MCHS. It has been demonstrated that cavities can enhance the heat transfer but generate the dead zone of flow, while longitudinal vortex generators (LVGs) have the advantage of low pressure drop. In this paper, an integrated microchannel heat sink combing cavities and LVGs (MC-LVG-C) is proposed. The impacts of geometric parameters of cavities and LVGs on the thermal-hydraulic performance of proposed MCHSs are analyzed and evaluated by CFD when the Reynolds number (Re) ranges from 150 to 325. Further, the multi-objective configuration optimization based on Non-dominated Sorted Genetic Algorithm-II (NSGA-II) is performed at Re=325 to minimize the thermal resistance and pumping power of MC-LVG-C simultaneously. It is found that the proposed scheme (MC-LVG-C) has clear superiority in the thermal-hydraulic performance over MCHS with cavities only (MC-C) and MCHS with LVGs only (MC-LVG). Geometric parameters analysis reveals that the height (hl) and length (L) of LVGs and the spacing (L2) between LVGs and cavities have significant impact on the thermal-hydraulic performance. The NSGA-II optimization indicates that optimal thermal-hydraulic performance values for MC-LVG-C are divided into four design zones. For optimal solutions, better heat transfer can be obtained by increasing hl and L, while smaller resistance can be provided by increasing hl and L2. In practical design, the reasonable geometric parameters can be chosen from the proposed four design zones by the design requirement.