Computational Fluid Dynamics (CFD) is a numerical approach to analyzing fluid flows in a given environment, particularly in building design. It provides information on air velocities, pressures and temperatures in and around building spaces. This method uses partial differential equations to numerically solve the phenomena, taking into account boundary conditions such as building aeraulics, internal heat gains and air-conditioning systems. CFD simulations are essential for optimizing ventilation and air conditioning in large spaces, guaranteeing optimum comfort.

Partial differential equations require boundary conditions to be solved. These are established on the basis of on-site measurement data and information from the project manager. For a steady-state study in a space open to the outside, we need to define the characteristics of the walls (material, physical properties, viscosity, temperature) as well as those of the surfaces exposed to the outside (flow direction, velocity, pressure, temperature, surface coefficients). It’s crucial to guarantee computational stability when defining these conditions, as the equations are solved iteratively to get closer to the solution.

The code solver used approximates the equations at each mesh node, respecting the fundamental principles of physics (conservation of mass and energy). It uses the standard k-epsilon turbulence model, which solves for two variables: turbulent kinetic energy and the kinetic energy dissipation rate. This model is widely used in industrial and HVAC applications because of its good convergence speed and acceptable memory requirements. For thermoaerulic studies, the effect of radiative exchanges between walls, thermal conduction, thermal draught and gravity are taken into account. The studies are carried out on the whole building, without establishing a symmetry section.