*3.4. CFD Modeling*

CFD calculations were used to simulate the flow field inside the stack, comparing the model with the experimental results, and driving further measurements. The commercial software Ansys® Academic Fluent, Release 15.0 has been used for this calculation. It solves conservation equations for mass and momentum, according to an Eulerian approach, while an additional equation for energy conservation is solved to account for heat transfer processes. For flows involving species mixing or reactions, a species conservation equation is solved for each one and additional transport equations are also solved when the flow is turbulent.

In this model, the turbulent viscosity *μt*, assumed as isotropic, is computed by combining the turbulent kinetic energy *k* and its dissipation rate as follows:

$$
\mu\_t = \rho \mathcal{C}\_{\mu} \frac{k^2}{\epsilon},
\tag{4}
$$

where *ρ* is the density and *Cμ* is a constant.

A tetrahedral unstructured grid, consisting of 2.6 M cells, has been set up; the standard *k* model of turbulence has been adopted; the time-dependent simulations have been carried out using a fixed time step of 0.05 s. The composition, velocity and temperature of the gas stream have been set at the inlet, as boundary conditions for the numerical solver of the model equations. A homogeneous distribution of the tracer has been assumed on the inlet section at time zero conditions. Then, the average concentrations of the tracer on two straight lines corresponding, in the model grid, to the laser beam path lines in the real plant, have been computed for each time step, by averaging the tracer concentration calculated in each cell laying on these lines. These average concentrations are directly comparable with the detected signals.
