Optimization of double layer counter flow (DLCF) micro-channel heat sink used for cooling chips directly

Double layer counter flow (DLCF) microchannel heat sinks with rectangular channels have been modeled using FEM. The dimensions of DLCF such as height of channel, H/sub C/, width of channel, W/sub c/ and width of fin, W/sub f/ are variables selected for optimization in determining minimal thermal resistance using genetic algorithms (GA) and box optimization method for a fixed pumping power. Though both optimization techniques are widely used in constrained optimization problems such as the optimization of microchannel heat sinks, no direct comparison between the two techniques have been made. It is found GA performed marginally better than the box optimization method in terms of thermal performance. However GA required far less iteration time compared to box optimization method to complete the optimization. Later the GA is used to minimize thermal resistance of DLCF microchannel heat sink and compared with minimized thermal resistance of single layer counter flow (SLCF) microchannel heat sink and single layer parallel flow (SLPF) microchannel heat sink. It is found DLCF microchannel heat sink produced less thermal resistance compared to SLCF and SLPF microchannel heat sink.