International Journal of Mechanical & Mechatronics Engineering IJMME-IJENS Vol:19 No:03 10 190903-2727-IJMME-IJENS © June 2019 IJENS I J E N S Study of the Aerodynamic Structure in an Indoor Environment Occupied by a Human Body Hasna Abid* , 1,2 , Zied Driss 1 , Jamel Bessrour 2 1 Laboratory of Electro-Mechanic Systems (LASEM), National School of Engineers of Sfax (ENIS), University of Sfax, B.P. 1173, km 3.5 Road Soukra, 3038 Sfax, TUNISIA 2 National School of Engineers of Tunis (ENIT), University of Tunis El manar, BP 37, le Belvedere, 1002 Tunis, TUNISIA Corresponding authors: E-mail address: * abidhasna@ymail.com (HasnaAbid) Abstract-- This paper presents the CFD study of the aerodynamic structure in an indoor environment occupied by a human body. Reynolds Averaged Navier-Stokes equations were used and different turbulence models were evaluated such as the RNG k-ε turbulence model, the standard k-ε turbulence model, the standard k-ω turbulence model and the SST k-ω turbulence model. The results of our simulations developed using the software ''ANSYS Fluent 17.0'' were compared with findings from anterior experimental measurements. This comparison affirms that the turbulence model has a direct effect on the aerodynamic structure in an indoor environment. In fact, the SST k-ω turbulence model presents a good agreement with the anterior study and it is recommended for the related type of application. These numerical results highlight a thermal plume above the human body due to the convection heat transfer. Finally, indoor thermal comfort analyses are conducted at varying supply velocity 1. INTRODUCTION The indoor environment has a huge impact on the health of the occupant and the quality of the air [1-2]. In this context, several studies have been conducted to study the airflow and the temperature distribution inside buildings. In fact, research work is divided into two main categories. The first category is interested in the experimental study. However, the second type of research is focused on the validation of these results through the numerical method. In this respect, the occupant shape is one of the investigated parameters in the experimental analysis as shown in the relevant experience done by Brohus et al. [3]. They examined the flow of air around three simple heaters: a solid rectangular torso, a rectangular torso with a leg, a rectangular torso with a box head. Munds [4] used a heated cylinder. Then, Bjorn et al. [5] used a thermal mannequin. Over the past few years, Computational Fluid Dynamics (CFD) has been widely used to study indoor problems as this alternative is less costly. Hence, several CFD application studied the indoor air behavior in the indoor environment [6-10]. A number of simulations were conducted by Deevy [11] in order to evaluate the human body geometry effect. It is observed that the results were not adequate next to the body for the simple geometry. In addition, numerical analyzes were carried out with different occupant position. On one hand, the researches focused on the seated human body.Sorensen et al. [12] represented a numerical study of the flow around a seated mannequin inside a room with a displacement ventilation system. The heat transfer was taken place by convection and radiation. On the other hand, several researchers assessed the airflow around a standing occupant. Murakami et al. [13] investigated the airflow around a standing person by a low-Reynolds number k–ε model taking into account the thermal radiation effects. Indeed, it is worth mentioning that CFD results are very sensitive to the boundary conditions, computing grid qualities and the choice of turbulence models [14]. In this context, Sideroff and Dang [15] studied the grid resolution effect. In addition, they applied the ν 2 - f model and the Smagorinsky dynamic model (DSM). As a result, they found that the DSM did not offer an advantage compared to ν 2 – f model. Deevy et al. [16] investigated the influence of the turbulence model to compare between a detached-eddy simulation (DES) and an Unsteady Reynolds-averaged approach (URANS). They noticed that the DES model showed the best results in terms of temperature and velocity distributions. Taghinia et al. [17] applied the dynamic Smagorinsky model (DSM), the SST-SAS model and the RAST (Rahman–Agarwal– Siikonen–Taghinia) one equation model (OEM). Their study showed that the RAST OEM produced the most reasonable results. In a recent work, Taghinia et al. [18] simulated a simple shape of the human body applying three LES SGS (sub-grid scale) and one DES models. Furthermore, the effect of radiation from the occupant's body was considered in some cases. Comparisons with the experimental data proved that the ZEM and WALE models provided an accurately prediction of velocity and temperature profiles. Based on these previous studies, it is clear that the use of an adequate turbulence model is crucial for the study of aerodynamic characteristics inside the building. Furthermore, the boundary conditions variations have a significant impact on the numerical simulation. In this respect, Mohammed [19] used the zero equation model to study the effect of the inlet temperature and the velocity on the indoor air quality and the thermal comfort. He confirmed that a low inlet velocity can provide a very good condition in the environment. Martinho et al. [20] investigated the inlet boundary