Multi-objective optimization of microchannel heatsink with wavy microtube by combining response surface method and genetic algorithm

This paper utilizes a combination of response surface method and genetic algorithm to provide an optimized model for microchannel heatsink with wavy microtube. The effect of flow rate in microchannel and microtube, location and diameter of microtube on the desired response surfaces, including hydrodynamic, thermal, and entropy parameters is examined. Then, the correlation between the variables is determined using a nonparametric regression model. The optimal design points are identified by multi-objective optimization and the Pareto diagram. The greatest effect is related to the nominal Reynolds number, with is directly related to the pumping power and entropy generation. The next effective parameter is the diameter of the microtube, which is inversely related to the pumping power, average surface temperature, and entropy generation. It is found that the maximum temperature difference on the hot surface is 3.1 K for optimal design HS3. For the geometry HS1, this value is less than 0.9 K by examining the optimal geometries of the micro heatsink based on the simultaneous study of the uniformity of CPU surface temperature and average CPU temperature. In addition, the flow velocity in the microtube of the design HS1 is much higher than that of the design HS3.

Automated summary

This paper combines response surface method and genetic algorithm to provide an optimized model for a microchannel heatsink with wavy microtube. The study examines the effects of various parameters on heatsink performance, determines the optimal design points, and compares the advantages and disadvantages of different designs.