MULTI-OBJECTIVE OPTIMIZATION OF A BI-METAL HIGH TEMPERATURE RECUPERATOR FOR APPLICATION IN CONCENTRATING SOLAR POWER

In recent years, supercritical CO2 closed-cycle Brayton cycles have become a major candidate for future power cycle designs in concentrating solar power (CSP) applications, with many of these designs including partial recompression and regeneration to increase thermal efficiency. This increase in efficiency, combined with potential miniaturization of heat transfer equipment and turbomachinery, could help significantly decrease the cost of energy generated by CSP plants. The hightemperature recuperator in these designs plays an integral role in these cycles and must operate and high temperatures and pressures. Printed circuit heat exchangers (PCHEs) have become a leading technology for these recuperators due to their size advantage over traditional shell and tube heat exchangers. However, PCHEs for high-temperature recuperators often must be built from costly nickel alloys to accommodate the extreme operating conditions. One potential solution to this cost problem is to tailor the material of the heat exchanger body to its operating conditions, rather than needing to choose a single material. This could be accomplished by using additive manufacturing to create a multi-material unibody heat exchanger, with a high-performance nickel alloy being used only where temperature and pressure dictate its use. Specifically, powder bed fusion (PBF) would be used to create the low-temperature portion of the recuperator in stainless steel 316L, then the high temperature region would be added directly to the low-temperature portion in Inconel 625 using directed energy deposition (DED). This methodology would have the additional benefit of being able to manufacture the heat exchanger headers at the same time as the core. In this project, a 1-D model of such a heat exchanger is devised which models the variability of both fluid and solid properties. The design of the heat exchanger core is based on existing PCHE core designs. While optimizing the core design, a number of different channel shapes and fin configurations are considered. Arrays of airfoil fins appear to have comparable heat transfer performance with reduced pressure drop when compared to other core designs. A multi-objective optimization of a small-scale heat exchanger is then performed using the 1-D model in order to determine the dimensional parameters which simultaneously maximize the heat exchanger effectiveness and minimize its size. Two designs appear in the Pareto front resulting from this optimization. Analysis shows that the design with less heat transfer area achieves higher effectiveness by limiting axial conduction in the walls of the recuperator while also suffering much less pressure drop in both fluids.