I want to use DNS method to calculate the flow field data first, eg. rms velocities, mean velocities. May i use fluent to import these data to do particle tracking using DPM? If it's available, someone can tell me the procedures? Thanks in advance.

forgot to say: it's a 2-D turbulent flow in duct.

Just solve your flow field and then define an injection

i mean i use another approach (not fluent) to get flow field. my question is how to import these flow field data into fluent for the further particle tracking? anyway, thanks, man.

You can interpolate data from a fluent simulation (say the x-component of the velocity vector for each cell) and edit the .ip file to exam the format the Fluent uses. But in doing things this way you will need to make sure that your other mesh (that you will be using in a different CFD code I presume) has a the same mesh resolution and geometry for importing the data into Fluent

thanks, man. do you know another cfd code which uses lagrangian equation to solve particle tracking problems?

you did not understand my reply.

If you are using another approach to get the flow field data as you said, then you have to make sure the data is in the appropriate format before attempting to interpolate the data into Fluent. This is possible.

Are there other codes that solve the Lagrangian particle force balance? Yes.

could you recommend me some? Fluent uses discrete random walk model to predict turbulent fluctuations. In my thesis, i'm supposed to use another models, e.g., continuous filter white-noise model, to get turbulent fluctuations. Then i can compare the simulations results. So, do you know some codes, solvers like that? thanks a lot.

Sorry Hao, I am only familiar with the DRW and particle cloud tracking models. It sounds like you may need to develop your own code in MATLAB or C++.