Hello,

I am looking for works on reactive turbulent flows in which there is comparison between two or more differents approachs:

- Monté-Carlo /PDF (problility density functions) - Beta function/mixture fraction - Eddy contact/multi-fluids - DNS (Direct Numerical Simulation)

a lot of thanks mounir

Could you give us some background information about each simulation method for turbulent reacting flows? And also why they are being pursued in so many different ways. Is this field almost hopeless?

One the one reasons for using particle based method is the simplicity of modelling several types of particles which interacts together trough collisions. One can easily understand that it is very simple to model a chemical reaction with that type of approach (particle 1+particle 2 -> particle 3).

One other reason is the simplicity of the update algorithm which is linear. The approach is different from PDE solvers where you try to solve directly NS equations. In a particle method,you can come back to NS trough the update equation. Actually, one industrial code (see www.exa.com) uses this method for external and internal subsonic aerodynamics. It is a very promising field but unfortunately unknown for a majority of CFD people (as it is based on statiscal mechanics and gas kinetic theory)...

Fred

Fred

Hello,

i think we have to made a distinction between different particle methods!

For turbulent reactive flows you can solve the transport equation for the joint probability density function (PDF) of velocity and composition. The PDF can be expressed as a sum of delta-peaks, which can said to be stochastic particles. The evolution of this particles can be calculated. By averaging them you can get the means and then moments of the PDF.

The great advantage of this method is, that reaction appears in closed form.

About the Lattice-Boltzmann-Method i don't know anything. But after a short look on www.exxa.com, i think its a different method

Greets

I'm a mechanical engineer. I want to improve myself in this area