Multi-objective optimization of a double-layered microchannel heat sink with temperature-dependent fluid properties

Abstract

Multi-objective optimization of a microchannel heat sink with a rectangular wedge-shaped cross section was performed in this work. The optimization was performed by a multi-objective genetic algorithm using three-dimensional conjugate heat transfer analysis with variable thermo-physical properties of the coolant (water). Response surface approximation was used to approximate the objective function to reduce computing time. Two geometric variables related to the channel cross section and a ratio of flow rates in the upper and lower channels were selected as design variables for the optimization. Thermal resistance and pumping power were considered as the objective functions. The design space was explored through a parametric study, and Latin hypercube sampling was used for the selection of the design points. The optimizations were performed for three different heat fluxes: 1 × 106, 2 × 106, and 3 × 106 W/m2. Pareto-optimal fronts representing the trade-offs between the performance parameters were obtained for the selected heat fluxes, which yielded important results for the design of microchannel heat sinks.