Evaluation of models and methods to simulate thermal radiation in indoor spaces

The theoretical models of Surface-to-Surface (S2S) thermal radiation, including the Monte Carlo model, Discrete Transfer model, Modest model and Heat-Flux-Split approach, are evaluated in terms of the predictive accuracy and CFD computational cost when simulating indoor thermal flows. It is demonstrated that the inclusion of thermal radiation in the CFD model is vital as the air temperature in the lower levels can be underpredicted while the heater surface temperature can be significantly overpredicted if the radiative effects are ignored. In addition, the predicted temperature distribution on the heat-receiving solid surfaces is highly sensitive to the selected radiation model. The comparisons demonstrate that the Monte Carlo model and Discrete Transfer model have comparable predictive capabilities while the latter requires less CPU time and is more computationally efficient. The appropriate number of the representative photons and rays are also recommended for the Monte Carlo model and Discrete Transfer model, respectively.