Building Simulation: An International Journal

Article Title

Evaluation of CFD turbulence models for simulating external airflow around varied building roof with wind tunnel experiment


turbulence, velocity, LDA, CFD, RANS


Detailed airflow information around a building can be crucial for the design of naturally ventilated systems and for exhaust air dispersion practices in agricultural buildings like greenhouses and livestock buildings. Full-scale measurements are cost-intensive and difficult to achieve due to varied wind conditions. A common method to gain insight of flow field under different wind conditions is the numerical simulation by using computational fluid dynamics (CFD). Still, evaluation of a CFD models’ performance and validation of its predictions with high quality experimental data is necessary before the model is used in practice. In this research three types of common agricultural buildings, arched-type, pitched-type and flat-type roof, were examined by conducting experiments in a wind tunnel with controlled airflow conditions, in order to validate different 3D turbulence models for predicting airflow patterns. The focus of this work was the detailed description of the external airflow field over the varied roof geometries and especially the velocity distribution and turbulent kinetic energy in the wake of each building. Experimental measurements of velocity were performed with a Laser Doppler Anemometer (LDA) and were compared with 3D RANS turbulence models’ simulation results. A reasonable agreement was found between experimental and simulation results concerning the velocity and the turbulence kinetic energy with CFD models slightly underestimating these magnitudes. The k-ε series turbulence models and especially the standard k-ε, RNG k-ε and Realizable k-ε models presented good agreement concerning velocity contours; however, high prediction error occurred over the roof of the buildings compared to the average values.


Tsinghua University Press