Helical coil has drawn increasingly attention as the gas heater of supercritical Rankine cycles due to its high heat transfer rate and compact structure. This work focused on the thermal performance of supercritical CO2 helically coiled heater at a practical thermal boundary condition (i.e. convective boundary), under which the coil’s performance has been scarcely paid attention. Supercritical flow turbulence was solved by the Shear-Stress Transport k-ω model and heat transfer was dealt with in a solid-to-fluid conjugate manner. Influence of fluid temperature, mass flux, coil curvature and buoyancy was comprehensively analyzed over a wide range. Results reveal a significant self-regulation of local heat input in the region of Tb < Tpc, where heat flux gently varied along the coil axis. The mass flux and inlet temperature of heat source fluid had a strong impact on the local heat flux, the heat transfer coefficient and the thermal inhomogeneity, while coil curvature only had a weak influence. Comparison with results under uniform heat flux boundary shows that at convective boundary the influence of buoyancy on average heat transfer was weakened in the form of reduced affecting region and intensity, and local thermal inhomogeneity caused by buoyancy and centrifugal force was also relieved. Effect of thermal boundary condition is minimal in buoyancy-negligible cases like high mass flux condition. Further evaluation identifies that several Nusselt correlations obtained at uniform heat flux were capable of predicting average heat transfer at convective boundary, with a relative error within ±20%. Local heat transfer was characterized by a dimensionless number Ψ, which has three critical values of 0.1, 1 and 10 to quantitatively identify the relative importance of buoyancy and centrifugal effects. Finally, evaluation on the system design shows that helically coiled gas heater can contribute to improved compactness, reduced investment cost and a more flexible range of operating conditions for supercritical CO2 Rankine cycle.
Read full abstract