Multi-objective Optimizations Of Bionic Micro-channel Heat Sink Based On Genetic Algorithm And Investigations On Synergy Characteristics Between Heat And Flow

The development in microelectronic industry and semiconductor packaging technology has driven electronic chips to high integration density and high power density,which poses serious challenges on heat removal of electronic chips.Micro-channel heat sinks are being paid more and more attention due to its excellent cooling performance,miniature size and low noise,and how to arrange micro-channels in the confined space has become a hot topic for researchers in liquid cooling technology.After thousands years of natural selection,transport systems in mammals and plants has evolved into the finest configuration with excellent heat and mass transport performance.Inspired by fractal network in nature,in this paper,a Y-shaped?tree-like?fractal network is applied to micro-channel heat sinks?MCHS?.To make the bionic micro-channels applicative to heat removal of high power density integrated circuit,genetic algorithm is employed to optimize the configuration fractal network with four branching levels.First,N₁,b₁,r₁,r₂ are selected as the design variables,single-objective optimization is performed by separately optimize pumping power and thermal resistance of the bionic heat sink,and the hydraulically optimal model with minimal pumping power and thermally optimal model with minimal thermal resistance are obtained.The geometric parameters for hydraulically optimal model are?5,13.3,6.7,8.8?,with pumping power 13.6mW.The geometric parameters for thermally optimal model are?18,4.9,6.8,8.9?,with thermal resistance 0.121K/W.Numerical results show that though hydraulically optimal model presents excellent flow performance,its cooling performance is poor.While the thermally optimal model provides remarkable cooling efficiency,the pumping power it consumes is quite high.Therefore,a multi-objective optimization of flow-thermal resistance is needed.Results of multi-objective optimization show that in the effective zone of pumping power,constraint value of pumping power has significant impact on configuration of bionic micro-channels,the greater constraint value is,the more the initial branching,and more sophisticate the fractal network.Comparative study in heat transfer and flow characteristics between single and multi-objective optimal models demonstrates that the cooling performance of multi-objective optimal model is favorable and is similar to the thermally optimal model.Furthermore,Energy economy of multi-objective optimal model is almost the same to the hydraulically optimal model.Multi-objective optimal model can achieve mutual balance between thermal and hydraulic performance of the heat sink and shows excellent comprehensive performanceTo improve the temperature uniformity on heat source,double-layered and truncated double-layered heat sinks based on multi-objective optimal model are proposed.Results indicate that double-layered heat sink effectively suppresses the temperature rise on heat source though,a large temperature gradient can still be found near the inlet.The truncate double-layered heat sink,however,can alleviate the“overcooling”around the inlet,where the temperature uniformity improves by 24%?30% compared with the double-layered one.In addition,the truncate double-layered heat sink provides high cooling capacity which is very close to the double-layered one,about 1?2W/cm² lower than the double-layered heat sink.In the double-layered heat sink,the bottom layer shows advantage in heat transfer,when coolants are evenly distributed in two layers,heat exchange ratio between bottom and top layer increases from 1.22 to 1.33 with rising total flow rate,and it remains constant at 1.32 with rising heat flux(q_vtot=500 ml/min).While in the truncate double-layered one,heat exchange ratio increases from 1.31 to 1.39 with increasing total flow rate,and it remains at 1.38 in different heat fluxes(q_vtot=500 ml/min).In the end,to make the best of heat transfer advantage of the bottom layer,flow distribution between layers is optimized.It's found out that there is an optimal flow rate ratio?between the bottom and top layer,?_opt=2,where the outlet temperatures of two layers are equal to each other,and the cooling potential in each layer of fluids in each layer is fully brought out,leading to the lowest peak temperature.When?<2,heat exchange ratio Q_eff1/Q_eff2>?.While?>2,Q_eff1/Q_eff2<?.Only when heat distribution is accordant with flow distribution Q_eff1/Q_eff2=?=2,can the coolant in each layer be fully utilized,maximizing the cooling performance of truncate double-layered heat sink.