چكيده به لاتين
The aim of present work is to examine the effects of interaction between turbu-lence and thermal radiation in the presence of combined mixed convection–radiation heat transfer in channels. The flow is fully developed in the channel length and ther-mo-physical properties of fluid are variable with temperature. Many industrial appli-cations are facing with flow and heat transfer at high temperature and also in the presence of temperature difference in channels. Due to existing high temperature and considerable temperature differences, natural convection, thermal radiation and vari-able thermo-physical properties become important. Ignoring any of the mentioned phenomena can lead to errors in the simulation of heat transfer. Therefore, effects of all of these phenomena should simultaneously be investigated. In many of the previ-ous studies, thermo-physical properties are kept constant or thermal radiation is not considered.
The vertical and horizontal channels under study are formed by differentially heated flat parallel plates. Large eddy simulation and the low Mach number approach are employed to solve the governing equations. Also, the radiative transfer equation is solved using the method. The simulations have been performed using the cur-rently developed buoyantPimpleFoam solver in OpenFOAM.
The main focus is to find out whether neglecting turbulence-radiation interaction (TRI) is a valid assumption for such flows under consideration. At first, attention is paid to the validity of the Boussinesq approximation without radiation when the temperature differences are enhanced. Results display that an increase in the walls’ temperature difference, applying the Boussinesq approximation and neglecting the dependence of thermal conductivity and dynamic viscosity on temperature, can result in relatively large errors. Deviations of up to 33% and 45% in the prediction of fric-tion coefficient and Nusselt number, respectively, are shown. Also, in the non-Boussinesq conditions, the overall fluid temperature across the channel height in both channel configurations is more than those of the Boussinesq conditions. In the hori-zontal channel case, larger departures from the Boussinesq conditions result in more asymmetric velocity and temperature profiles.
Finally, the present results of turbulence-radiation interaction show that, in both configurations, the maximum values of emission TRI and incident TRI are 2% and 3%, respectively. Therefore, neglecting TRI in the non-reactive flows has almost no effect on the heat transfer flows, even with the existence of variable properties and non-Boussinesq conditions. Also, the turbulence is damped through the entire do-main of both horizontal and vertical channels by increasing the optical thickness.
