[tonky]

Hello everybody,

i am new at OpenFoam, but i read much theory of it and now I would like to implement the Acoustic Perturbation Equations (APE) in OpenFoam. I think it would not be very easy, so i hope that maybe somebody here have made experience with it.

Here is the relevante paper link:
http://www.sciencedirect.com/science...21999103001682

If you are a student, then it is mybe for free, if you log in with VPN of your university.

So if i get some experience with it i will share it here.

[Alhasan]

Dear Tonky,

By any chance did you have any luck in implementing APE in OpenFOAM.

Regards,
Hasan K.J

[tonky]

Hello Hasan,

after some practice with OF, I have started to work further on the implementation of APE-2 theory. But now I am facing some problems, which I am now trying to solve, also with the help of cfd-online.com.

I noticed that the equations of "APE-2" are similar to the equations of Seo and Moon in "Linearized Perturbed Compressible Equations":
http://www.cfd-online.com/Wiki/Hydro...stic_splitting

But my problem is now that i implemented the following three equations (see attached files) and I am solving them on the explicit way (diagonal solver) and the values of rho' and u' are increasing during the simulation without a limit. Now I am trying to implement them on the implicit way. Is this possible and if yes, how?

Hasan, did you have some experience with APE?

[tonky]

Hello Foamers,

I have read many posts here in cfd-online, but could not solve my problems with the iterations with the number of zero. I implemented the three equations in the following form:

solve(fvm::ddt(rhoS)+ rhoS * (fvc::div(averageU))
==
- rho * (fvc::div(Ua))
);

solve(fvm::ddt(Ua)+ fvc::grad(averageU & Ua)
==
- fvc::grad(pa));

solve(rho*fvm::ddt(pa)- c0*c0*fvc::ddt(rhoS)
==
- rho*fvc::ddt(pS));

The problem now is that the values of Ua and rhoS are rising until the simulation is breaking down, because of to high values. In my opinion it is because the solver cannot make more then one iteration and so could not converge. Am I right? And how can I solve this PDE's the right way?

[Zhi Cheng]

I'm getting in trouble with APE in OpenFOAM too...Thanks!

Originally Posted by tonky

Hello Foamers,

I have read many posts here in cfd-online, but could not solve my problems with the iterations with the number of zero. I implemented the three equations in the following form:

solve(fvm::ddt(rhoS)+ rhoS * (fvc::div(averageU))
==
- rho * (fvc::div(Ua))
);

solve(fvm::ddt(Ua)+ fvc::grad(averageU & Ua)
==
- fvc::grad(pa));

solve(rho*fvm::ddt(pa)- c0*c0*fvc::ddt(rhoS)
==
- rho*fvc::ddt(pS));

The problem now is that the values of Ua and rhoS are rising until the simulation is breaking down, because of to high values. In my opinion it is because the solver cannot make more then one iteration and so could not converge. Am I right? And how can I solve this PDE's the right way?

[xhan]

Hi,

another form of APE might be easier to implement in OpenFOAM. In the work of Maurerlehner et al., they introduced the scalar acoustic potential and solve the equation of it. Then acoustic pressure can be calculated in post-processing step.

Maurerlehner, P., Schoder, S., Tieber, J., Freidhager, C., Steiner, H., Brenn, G., Schäfer, K., Ennemoser, A., & Kaltenbacher, M. (2022). Aeroacoustic formulations for confined flows based on incompressible flow data. Acta Acustica.

https://1drv.ms/i/s!App_wgTsRcZ5gsgb...ofyhg?e=T99enV