This part of the project will be devoted to the study of three different meshes:

               *   The coarse structured mesh shown below:

               *   The coarse structured mesh refined close to the airfoil:

                 *   An unstructured mesh described below:

            We have tested those three meshes with exactly the same parameters. The following figure represents
the calculated residuals of the volumic mass evoluating with iterations.

            We can see that the precision for the structured mesh has a limit around 1000 iterations. Then for this case too much iterations are not usefull. The refined structured mesh improve the convergence of about three
order of precision. However, this second calculation was longer than the first one. The unstructured mesh also has a better precision.

        Therefore the third mesh seems to be more efficient. We will now look more precisely at results obtained
with each mesh. The following figures represent the evolution of mach number, after 10000 iterations, for the different meshes.

            We can see in those figures that, although the unstructured mesh gets a better precision with the residual study, it is less clear when we look at the ending shock on the airfoil. However, the refined structured mesh gives a better result. Therefore the use of unstructured mesh seems not to improve results although it improve
precision.