Numerical Optimization for Nanofluid Flow in Microchannels Using Entropy Generation Minimization

This study presents the numerical simulation of the three-dimensional incompressible steady and laminar fluid flow of a trapezoidal microchannel heat sink using nanofluids as a cooling fluid. Navier-Stokes equations with a conjugate energy equation are discretized by the finite-volume method. Numerical computations are performed for inlet velocity (W (in)=4m/s, 6m/s, and 10m/s), hydraulic diameter D ( h )=106.66 mu m, and heat flux (q ''=200kW/m(2). Numerical optimization is demonstrated as a trapezoidal microchannel heat sink design which uses the combination of a full factorial design and the genetic algorithm method. Three optimal design variables represent the ratio of upper width and lower width of the microchannel (1.2 <=alpha <= 3.6), the ratio of the height of the microchannel to the difference between the upper and lower width of the microchannel (0.5 <=beta <= 1.866), and the volume fraction (0 <=phi <= 4%). The dimensionless entropy generation rate of a trapezoidal microchannel is minimized for fixed heat flux and inlet velocity. Numerical results for the system dimensionless entropy generation rate show that the system dimensionless friction entropy generation rate increases with Reynolds number; on the contrary, the higher the Reynolds number, the lower the system dimensionless thermal entropy generation rate. The results below show that the two-phase model gives higher enhancement than the single-phase model assuming a steadily developing laminar flow.