Conclusion

    This Workshop gives us the opportunity to improve our skills in the use of a CFD tool, namely Fluent. More than a rigorous study of the Rayleigh Benard instabilities, it can be seen of an exercise dedicated to the study of these instabilities using a CFD tool.

The use of a CFD tool present several advantages for a Engineering study. First of all, it is quite easy to manage and is usually cheaper than test in nature scale. It provides also several values that can be computed everywhere in the domain. You can also save one computation and work on it after the first study without any costs.
For the negatives points, it can be very time costing in case of a complex geometry, but it's not the main difficulty. The centre of the problem is the numerical methods used in the code and the type of grid used. It seems that for a good accuracy of the calculus, lots of parameters have to be taken into account. The solution is sensitive to these parameters (for example the initial condition). To know exactly the reaction of the code and the way into the equation are solved is necessary for the user to be critical towards the results of this code. It is necessary for an engineer that is working on simulation that are not already very well known.

One of the goal of our formation at ENSEEIHT in MFN is to give us the ability to understand all these parameters. However, it appears that it takes a lot of time. To know Fluent, for example, we will need to work with it frequently and also to go deep into its coding. It seems not possible to try to achieve this task as there is several different codes in use at the moment, but we are aware of the fact that we are supposed to be able to "understand" as an advanced user any CFD tool.