SIMULATION AND THERMAL OPTIMIZATION OF A MANIFOLD MICROCHANNEL FLAT PLATE HEAT EXCHANGER

This paper describes a multi-objective optimization of single-phase, laminar flow inside a single element of a manifold microchannel flat plate heat exchanger. Approximation assisted optimization was used for the optimization process. The process uses metamodeling in conjunction with Computational Fluid Dynamic (CFD) simulation as a method to minimize the number of function evaluations and thereby obtain substantial reductions in computational time. Two optimization objectives were considered: a) maximizing heat density rate per temperature difference Q/(V Delta T) and minimizing pumping power density (PAT), and b) maximizing base heat transfer coefficient (h) and minimizing pumping power per base area (P/A(base)). Water and air were used as working fluids to compare the optimum solutions of the two fluids with very distinctive thermo-physical properties. The study shows that both optimization objectives result in similar optimum points. The behaviors of the optimum solutions for water and air are also discussed in detail. Additionally, as a case study using the optimization results, it was demonstrated that for an array of microchannels with volume as low as 4,250 mm(3) on one side, pumping power of 138 W and heat transfer rate of 56.7 kW can be achieved using water.