Simulation of spectators' aerodynamic drag using porous models approximation. (15th October 2020)

• Record Type: Journal Article
• Authors: Mahgoub, Ahmed Osama; Ghani, Saud; Rashwan, Mohammed M.; Ismail, Salman M.; ElBialy, Esmail A.
• Is Part Of: Building and environment. Volume 184(2020)
• Publication Date: 2020-10-15
• Subjects: Aerodynamics -- CFD -- Drag force -- Porous media model approximation -- Stadiums -- Thermal comfort; Buildings -- Environmental engineering -- Periodicals; Building -- Research -- Periodicals; Constructions -- Technique de l'environnement -- Périodiques; Electronic journals; 696
• DOI: 10.1016/j.buildenv.2020.107248
• Languages: English
• ISSNs: 0360-1323

Abstract

Abstract: Evaluation of the thermal comfort is essential for complex ventilation systems design. Assessment of thermal indices requires representative velocity and pressure fields' values. When simulating the air flow in large facilities such as stadium, the effect of crowds' geometrical features needs to be captured. Using porous models approximations to simulate the aerodynamic effect of detailed spectators' geometry reduces the required mesh size and associated processing time. This paper investigates the use of different porous media models approximations for capturing the effect of large crowds inside complex building systems, such as stadiums. Their efficiency of capturing the effect of spectators on the air flow were compared to the simulation of the exact spectators' geometry. The exact spectators' geometrical model was of a stadium tiers section with 28 spectators. Using a wind tunnel, the exact spectator's model results were validated against a 1:10 scaled physical model. The experiments included PIV and hot-wire velocity measurements. The results of the pressure drop were used to obtain the coefficients needed to utilize the porous models. Compared to the exact spectators' case, the three-dimensional porous volume model approximation yielded an average absolute error of 24.5% in velocity, while the two-dimensional porous jump model yielded results with an average error of 1.5%. In comparison to the exact model cooling load, the results yielded a difference of 6% for the 2D porous jump and 6.5% for the 3D porous volume. Nevertheless, both models yielded more representative results than the case of simulation of empty bleachers. Highlights: To optimize thermal simulations, spectator body drag must be accounted for. The 2D porous jump model acceptably simulated the spectators' drag effect with an average error of 1.5%. The 2D porous jump model yielded better results in comparison to the 3D porous volume model. The porous model approach reduced the mesh size and computational processing time. The usage of porous models affected the values of the evaluated thermal comfort indices.