Investigation of flow and heat transfer characteristics in a ribbed channel using grid-adaptive simulation method

To gain more insight into the flow and heat transfer mechanics of the ribbed channel, the grid-adaptive simulation (GAS) method, an advanced hybrid Reynolds-averaged Navier–Stokes (RANS) and large-eddy-simulation (LES) method based on Kolmogorov turbulence energy spectrum, is conducted to predict flow and heat transfer in ribbed channels. Comparisons are made between GAS and conventional hybrid RANS-LES methods. Results demonstrate that GAS-SST rapidly triggers unsteadiness and offers better flow and heat transfer predictions than scale-adaptive simulation (SAS) and delayed detached-eddy simulation (DDES). Based on the high-fidelity predictions provided by GAS method, the predominant turbulent and thermal spanwise wavelengths in the wake region are quantified by Hilbert transform. The evolution of the predominant turbulent spanwise wavelength and energy level is investigated. A negative correlation between the characteristic thermal spanwise wavelength and the local Nusselt number in the wake region is revealed. The smaller thermal spanwise wavelength could lead to stronger local heat transfer. The entropy generation rate (EGR) is used to analyze the energy loss. Both steady and unsteady effects contribute significantly to EGR, the unsteady portion of EGR is distributed mainly in the wake, and the steady portion of EGR is distributed mainly in wall-bounded regions.