Hi guys

Can we do far field noise prediction using any of the CFD codes. If yes which code is it and how can we set up the boundary conditions to solve light-hill equations. Thanks in advance for you reply.

Usman

> Can we do far field noise prediction using any of the CFD codes.

Using a CFD code to get to the far field is bonkers. However, all compressible Navier-Stokes codes will predict acoustic waves, although not accurate ones in most cases, and so could be used to do this both inaccurately and at huge computational expense.

> If yes which code is it and how can we set up the boundary conditions to
: solve light-hill equations.

I presume you are referring to Lighthill's acoustic analogy? In which case, several commercial CFD codes can be used to provide source terms for various acoustic analogies. Whether this is useful in your case will depend on what you are trying to simulate.

There are also other approaches based on solving wave equations which may be more suitable for your problems.

Yes i was referring to Lighthill's acoustic analogy. Thank you very much for answer.

Regards Usman

the only way at present is to run a transient case with an LES model(for better accuracy) to get pressure data on the surfaces. Then use any CAA solver to get the sound intensity levels.

Thank you Mayur. I will look into CAA softwares.

Usman

Sysnoise is the most popular.

Thank you Mayur!

Regards Usman

Can you tell us what type of flow problem you are considering and what type of information you want to know about the acoustic field? My answer was unavoidably vague because of this and Mayur may or may not have made an incorrect guess about the type of flow you are considering.

Hi Andy

I talked to my supervisor about it this morning. I think we will stick to predicting near field noise because we are some what limited on computing resources at the moment. For now we are just considering pressure fluctuations in gas turbine combustor. Thanks for your help.

Usman

> I think we will stick to predicting near field noise because we are some
: what limited on computing resources at the moment.

In your case, getting to the far field is not a big problem except, possibly, along the axis of the jet. To do it you would place a surface near the outside of your CFD solution region enclosing all the acoustic sources and in the free stream (apart from part of the downstream jet). The CFD solution variables can be used to determine the strength of the acoustic waves on this surface and since one has uniform conditions and no acoustic sources outside this surface one can use fast and accurate anaytical means to get to the far field.

Of course, you still have a large CFD problem.

> For now we are just considering pressure fluctuations in gas turbine
: combustor.

Are these diffusion or premixed flames? Afterburners, combustor noise, acoustic instabilities, exhaust rumble or what?

1D methods can get a handle on a fair few real world problems. 3D unsteady CFD simulations are actively being researched in a number of groups but they are not being performed in large numbers in industry in the way of, say, the compressor and turbine folk.

Getting involved with a combustion research group might be a way forward but it depends on what problems you are trying to address and the resources your supervisor and above consider prudent to tackle them.