What are your methods and strategies to ensure a benefit your CFD simulations? What is your business or research segment?
It really depends on the sector and applications you're working with. In some sectors, quantitative predictions by means of CFD are possible, while in other applications, CFD predictions should focus on qualitative predictions, i.e. by comparing the impact of a change of geometry or boundary conditions on a complex system.
In the simulation of industrial combustion processes, for example, I think it is not really possible to accurately predict pollutant emission levels such as NOx quantitatively with reasonable certainty, for a number of reasons. It is however possible to predict how a change in a boundary condition like air excess ratio or a change of a burner geometry will affect the system and hence predict whether emissions will increase or not, and to what extent. So, by comparing a base case with geometry or boundary condition variations, you can improve the design of your product.
Dear Lukas, one of my research segments is fluid dynamics in burners combustion head and fan ventilation. I use CFD software, but results are often no confirmed by experimental data. I prefer write down numerical code (e.g. using Mathematica) for specific problems or models.
Gianluca
The CFD simulations provide the flow structure and heat transfer characteristics in the entire domain. It is not possible to obtain this information by performing experiments. Also experimentation is very costly. So, CFD is very helpful in industry considering this scenario. Also, instead of manufacturing a prototype every time for testing, modification in the CAD model and its simulation is easier in industrial scenario. So, CFD is becoming a essential tool in industry. But, you need some experimental data for validation of you CFD setup.
As a researcher;
-Some times the experimental work is not available ( high technique required, or very low dimensions, or the application on space, or the application on human body, or the some macro properties are invisible or not-measurable like internal properties of metal during solidification, or ...)
-Some times the experimental work is available but it is not easy to change the parameters of the problem to their effect during a process.
As an engineer;
- The quality improvement, the geometry development, the material enhancement, and the cost decrements, all of them need to theoretical and numerical parametric studies to avoid try and error process, to avoid stopping line of production, and to avoid wasting tools and materials and working time due to try and error processes.
Dear Lukas,
disregarding the application/research area on which the CFD tool is used, one should follow carefully some basic steps in order to produce reliable numerical results. Therefore, you have to bear in mind a couple of important issues, such as Validation and Verification. These are steps beyond convergence and mesh independency study (which also is a fundamental check) and they are related to experimental and mathematical accuracy of your modelling. You can find some literature from AIAA (AIAA Guide for the Verification and Validation of Computational Fluid Dynamics Simulations). This paper could also be of your interest (Article Validation Methodology in Computational Fluid Dynamics
.Good luck!
- Badges
- Science topic