Numerical design optimization of recuperator designs for sCO2 power cycle by CFD

As a part of my graduation internship in engineering and master’s degree, I work in the Helmholtz Zentrum Dresden-Rosseendorf institute for an internship of 6-months. I am a part of a team consisting of my supervisor, Dr. Sebastian Unger and a PhD student, Mrs. Malini Bangalore. The latter studies a thermal energy storage (TES). The storage would store electricity as heat from renewable energy sources when the supply in electricity is higher than the demand. In case of residual load demand, the stored thermal energy would be released to a power cycle and supply the electrical grid. Therefore, a heat exchanger , connecting the thermal storage system to the power cycle, needs to ben designed and here is where I intervene.
Before I came, it was already decided, that the power cycle uses supercritical fluid as working fluid and working fluid in the thermal energy storage circuit, more specifically the hot fluid, is CO2 at atmospheric pressure. It was also decided that the heat exchanger type is a printed circuit heat exchanger (PCHE), that the heat transfer rate is set to 10 MW and the hot fluid inlet temperature is 600 °C. However, to completely determine the geometry of the PCHE, the boundary conditions, i.e. temperature, pressure and mass flow rates in hot and cold channels, need to be determined.
To that end, a review of cycles and their modelling is made. It helps to calculate the boundary conditions for the cold side. Then, a cost-optimization analysis of the PCHE leads to determine the boundary conditions of the hot fluid. Indeed, as the hot fluid is CO2 at atmospheric pressure, even the smaller pressure drop has a significant impact on the operation cost of the PCHE. Therefore, a reference case may be determined.
Finally, CFD simulations were conducted on the reference case geometry. Furthermore, the behaviour of the PCHE is examined, when the mass flow rate is changed, e.g. because of different system operation. Finally, the CFD simulation results are compared to existing correlations.