******************* This email from originates from outside Imperial. Do not click on links and attachments unless you recognise the sender. If you trust the sender, add them to your safe senders list https://spam.ic.ac.uk/SpamConsole/Senders.aspx to disable email stamping for this address. ******************* Hello, I want to perform an aeroacoustic simulation in which I am using APESolver (nektar-4.4.1). I am transferring the source terms from the fluid solver (OpenFoam) to nektar++. I need to interpolate the source terms from the finite volume mesh (which is relatively coarser) ino the spectral mesh which can be used by APESolver. Initially, I used Shepard's algorithm to interpolate the source terms which performed well for serial runs but, unfortunately, there are several numerical errors when I am passing the data using multiple processors. Recently I found out that there is an already available utility in nektar++ for field interpolation ( as post-processing utility). I wanted to inquire if it is possible to use the functionality for interpolating the data during runtime with multiple processors. Happy new Year to all of you!! regards, Nitish Arya Department of Aerospace Engineering IIT Kanpur, India
Hi Nitish, the FieldConvert modules which take point data use a Shepard interpolation as well, so if your problem is related to the algorithm itself, this wont help. Our implementation however should work fine on multiple processors, so it might be worth a try. In general field interpolation in aeroacoustics can be difficult, which is why Nektar++ 5.0 ships a coupling interface which uses a more robust interpolation, see [1] and [2] for more info. An OpenFOAM implementation for this interface and the AcousticsSolver is available at [3]. Besides, the AcousticsSolver in 5.0 is more mature than the APESolver, so this might help as well. Cheers, Kilian [1] https://arxiv.org/abs/1906.03489 [2] https://tuprints.ulb.tu-darmstadt.de/7611/ [3] https://github.com/mimove/Coupling-API/tree/master/OpenFOAM_Nektar_API Am 31.12.19 um 12:01 schrieb Nitish Arya:
This email from nitiesharya@gmail.com originates from outside Imperial. Do not click on links and attachments unless you recognise the sender. If you trust the sender, add them to your safe senders list <https://spam.ic.ac.uk/SpamConsole/Senders.aspx> to disable email stamping for this address.
Hello, I want to perform an aeroacoustic simulation in which I am using APESolver (nektar-4.4.1). I am transferring the source terms from the fluid solver (OpenFoam) to nektar++. I need to interpolate the source terms from the finite volume mesh (which is relatively coarser) ino the spectral mesh which can be used by APESolver. Initially, I used Shepard's algorithm to interpolate the source terms which performed well for serial runs but, unfortunately, there are several numerical errors when I am passing the data using multiple processors.
Recently I found out that there is an already available utility in nektar++ for field interpolation ( as post-processing utility). I wanted to inquire if it is possible to use the functionality for interpolating the data during runtime with multiple processors.
Happy new Year to all of you!!
regards, Nitish Arya Department of Aerospace Engineering IIT Kanpur, India
_______________________________________________ Nektar-users mailing list Nektar-users@imperial.ac.uk https://mailman.ic.ac.uk/mailman/listinfo/nektar-users
participants (2)
-
Kilian Lackhove
-
Nitish Arya