Numerical Simulation of Temperature and Carbon Dioxide Distribution in Indoor Environment Using Two-Equation Turbulence Models

Heat transfer has a great effect on the energy saving of buildings. The main challenge is the description of turbulence effect. The present study conducts a computational fluid dynamics simulation of non-isothermal indoor airflow with contaminant diffusion to test the capability of four two-equation turbulence models. All two-equation turbulence models predict the similar velocity distribution, temperature distribution and concentration distribution. The SST k - ω model obtains larger turbulent viscosity far away from walls than k - ε models, meaning that the gradients of velocity, temperature and concentration is small. The SST k - ω model and the RNG k - ε model capture a large thermal plume above heaters. The computed temperature and carbon dioxide concentration are compared with the measurement. The standard k - ε model and the realizable k - ε model are capable of predicting a better temperature distribution. The realizable k - ε model has a good performance in simulating the concentration of CO 2 .