Learning Forum

[CarolineSCi]

Dear all,

I am trying to perform a two-ways FSI simulation as I would like to investigate the deformation of a porous medium and the flow behavior at its interface. please see the picture attached. 

It is a very basic simulation where the flow should pass through a porous medium. I would like to use the pressure distribution obtained at the porous medium interface as parameter input for my static simulation, where then I would get the deformation of the porous medium. However, this problem is not an one-way FSI, as I also have to investigate the flow behavior just close to the interface, where the porous medium starts, after the porous medium is deformed. Therefore I have to use a coupling system. (please, correct me if I am wrong)

I have looked for two-ways FSI and watched tutorials, but unfortunately all of them do not indicate how to use the interface between the fluid and the porous medium as a region for data transfer. One possibility would be the use of a porous jump boundary condition in Fluent, however it did not work so far. Do you guys have any idea how I could solve my problem?

Best Regards,

Carol

 

 

[Rob]

You need to transfer the pressure to Mechanical for the deformation: there will be a surface at the joint between the open fluid & porous zones (probably interior:something). That can be moved, but I don't know how FSI will deal withthat as it's not a wall. The deformation will also effect the porous coefficients which isn't dealt with: you'll need to figure that out.  

Overall it's not going to be an easy calculation. 

[CarolineSCi]

Hallo rwoolhou,

thank you very much for your answer. In fact, I have tried to move the interfaces before and I have seen it is possible, however I still do not know how to deal with the interfaces. As the joint has two interfaces, it will be possible to use them both in the data transfer for a coupling simulation, however the coupling simulation do not work. I am wondering if I can make them as one interface. Regarding the deformation, I think it will not be a problem as I can modify the permeability during a transient simulation. If the FSI can not deal with this sort of simulation, do you think it might be still possible to do a coupling system where the deformed mesh obtained from the Ansys mechanical can still be imported to fluent at each timestep ?

Best regards

[Rob]

If you make the fluid zones conformal that should leave you with an interior surface rather than an interface: that'll probably be easier to deform. 

[CarolineSCi]

Dear rwoolhou,

you are right. In fact, the only problem I had was the two interfaces generated when I split the domain. The other thing I had to do was to combine both interfaces using Fluent. Now the coupling simulation is working so far. Thank you very much for your help.

Cheers