What do CFD?
Through my professional activity, I offer companies expertise in CFD, or numerical modeling in fluid mechanics. “And what is it?” I have already been told. Here is, for those who are not familiar with it, what my job consists of.
Fluid mechanics... With computers
CFD (Computational Fluid Dynamics) consists in studying physics of flows via numerical resolution of the equations governing the behavior of fluids. It thus makes it possible to calculate the temporal and/or spatial evolution of any fluid system, subject to a degree of approximation with respect to reality which is directly dependent on the chosen modeling hypotheses.
Broadly, a CFD problem is often dealt with an approach consisting of three main steps, of which here is a small overview.
The problem modelling
It is during this phase that the digital representation of the actual process is built. It is therefore necessary to:
✓ Define the geometry of the problem and the computational domain.
✓ Discretize this calculation domain so as to be able to solve the desired equations (resolution at points, in control volumes, or others, everything depends on the chosen resolution method), i.e. build a mesh. To be complete, it should be pointed out that this step is not necessary with certain approaches. However, most of the major commercial and open-source codes on the market are based on the so-called finite volume method, which does require the creation of a mesh.
✓ Define the equations to be solved according to the physics of the problem (presence of thermal phenomena, turbulence, phase changes, chemical reactions, etc…).
✓ Define the boundary conditions of the computational domain, and the initial conditions of the problem dealt with.
✓ Define the physical properties of the materials present.
The expertise of the CFD engineer must in particular allow him to find a good compromise between the degree of precision / the type of desired results, and the hypotheses that it is possible to adopt in the modelling. In particular, taking into account the right terms in the equations, and the correct correspondence of the boundary conditions / initial conditions with the real process are essential points for obtaining a relevant simulation.
The process numerical simulation
This second step calls for the use of CFD software which will, at any point (or volume) of the mesh and for any time step, solve the evolution of the model previously built. Depending on the complexity of the addressed problem, the nature of the used software and the available computer power, this resolution can take a few seconds or several months.
The CFD engineer must here use his numerical skills to ensure the smooth running of the simulation, by choosing appropriately:
✓ Spatial and temporal numerical schemes
✓ The resolution algorithms of the linear systems created
✓ The numerical parameters necessary for a good convergence of the calculation towards a physical solution, without resulting in a simulation that is too long or unnecessarily fine.
The obtained results interpretation
During this last post-processing step, it is first necessary to ensure the consistency of the results. As such, the existence of a reference case where it is possible to compare with experimental results often proves to be valuable. Failing this, the evolution of the quantities of interest (taking into account the expected physical behavior, conservation laws, etc.) must make it possible to validate the model built (or to make corrections to it). It is only at this time that it becomes relevant to use the results for the desired purpose (optimization, understanding, etc.).
So what's the point?
Wisely used, the CFD proves to be a major asset:
✓ It is possible to simulate systems before building them, which allows them to be optimized in a pre-project phase.
✓ It is possible to optimize a process via a multitude of parametric simulations, all at a very modest cost and timeframe compared to a test campaign.
✓ Access to all physical quantities, at any point of the constructed mesh and at any simulated time, allows access to a much greater quantity of information than experimentally, which promotes understanding of the physical phenomena involved .
CFD is therefore both a research tool and an industrial tool, which has its place in the world of engineering. Nowadays, it is present in a very large number of sectors, ranging from transport to meteorology, via chemistry, medicine or construction. The optimization it brings can also have ecological benefits.
And the CFD engineer, in all this?
So, working in the world of CFD requires a number of skills:
✓ Strong knowledge of fluid mechanics, physics, applied mathematics, programming and scientific computing.
✓ Mastery of the tools on which a CFD engineer can base his expertise: CFD software, meshing software, scientific visualisation software… But also scientific programming languages.
The CFD engineer has, somewhere, the role of interface between the physical world and the numerical world. This often involves a great deal of interaction with other trades (civil engineering, processes, IT, etc.), and with other areas of physics (electromagnetism, mechanics, chemistry, etc.). His expertise must allow him to correctly represent the desired physics, with the tools and computing power at his disposal. As the saying goes, he must make correct calculations, not just calculations!
This is obviously the objective that I pursue through my activity at NEMOSFLOW 🙂
If you liked this article, or if you want to discuss it, do not hesitate to show up!
Thank you for visiting our website!