- Nektar-users - imperial.ac.uk

Re: [Nektar-users] FW: Questions regarding APE Solver
by Amitvikram Dutta 04 Nov '18

04 Nov '18

Thanks a lot Killian, Is there a tentative release date for Nektar 5? On Fri, Oct 26, 2018, 3:58 AM Amitvikram Dutta < amitvikram.dutta(a)uwaterloo.ca> wrote: > > > > ________________________________________ > From: nektar-users-bounces(a)imperial.ac.ukOn Behalf OfKilian Lackhove > Sent: October 26, 2018 3:58:07 AM (UTC-05:00) Eastern Time (US & Canada) > To: Amitvikram Dutta; nektar-users > Subject: Re: [Nektar-users] Questions regarding APE Solver > > Hi Amitvikram, > > 1. Yes, only the lamb vector was used, so the RHS of the pressure > equation is zero. > > 2. It seems the userguide is out of date. In Nektar++ 4.4 the RHS are > specified using forcings (see the userguide). See e.g. > APE_2DPulseAdv_WeakDG_MODIFIED.xml example how this is set up in the xml > file. In short, you would have something like this: > > <CONDITIONS> > ... > <FUNCTION NAME="Source"> > <E VAR="p" VALUE="0"/> > <E VAR="u" VALUE="define lamb vector here"/> > <E VAR="v" VALUE="define lamb vector here"/> > <E VAR="w" VALUE="define lamb vector here"/> > </FUNCTION> > </CONDITIONS> > <FORCING> > <FORCE TYPE="Field"> > <FIELDFORCE> Source <FIELDFORCE/> > </FORCE> > </FORCING> > > 3. The next Nektar++ major release will ship a fieldfromstring module, > which should be able to compute the source terms without the pts detour. > Until then you can just use "FieldConvert foo.xml foo.chk bar.pts" to > obtain a pts file with the values at the quadrature points. > > In general, Nektar++ 5.0.0 will contain many changes and improvements to > the acoustics solver, in particular with regard to hybdrid CAA > cosimulations, so you might want to give it a try once its released. > > Cheers, > > Kilian > > > Am 26.10.18 um 04:55 schrieb Amitvikram Dutta: > > Hi everyone, > > > > I'm trying to use the APE Solver in nektar++ to predict sound generation > > from a particular industrial flow. The work is similar in scope to that > > published in this recent paper : > > http://www.iccfd.org/iccfd10/papers/ICCFD10-248-Paper.pdf. > > I'm particularly interested in how to define the source term. If I > > understand the work correctly, in a flow dominated by coherent > > structures, the Lamb vector becomes the dominant source. > > This leads to my first question : > > 1.In the APE-4 formulation there are 2 coupled equations with seperate > > source terms - one for the pressure fluctuations and one for the > > velocity fluctuations. In the above paper when the authors state that > > only the lamb vector was considered as a source, was the source for the > > RHS of the pressure fluctuations set to zero? > > 2. If this was the case, how was this specified in the .xml file? The > > only example I have come across is in the user guide which shows > > > > <FUNCTION NAME="Source"> > > <E VAR="S" VALUE="0" /> > > </FUNCTION> > > > > If one wanted to specify the pressure and velocity source terms > > seperately how would one do it in the .xml file? > > > > 3. From older posts in the mailing list I have seen suggestions of a > > workflow where the .chk/.fld files are written into a .pts file and then > > post processed to obtain the acoustic source. Is there a relevant filter > > or an extension the FieldConvert module that allows direct conversion > > from a .fld/.chk to a .pts format? I know of the interppoints extension > > that allows export to a .dat format but not the .pts? > > > > Sorry for the lengthy post. > > Thanks in advance! > > -- > > > > *Amitvikram Dutta* > > > > Graduate Research Assistant > > > > Fluid Mechanics Research Lab > > > > Multi-Physics Interaction Lab > > > > University of Waterloo > > > > > > _______________________________________________ > > Nektar-users mailing list > > Nektar-users(a)imperial.ac.uk > > https://mailman.ic.ac.uk/mailman/listinfo/nektar-users > > > > -- > Dr.-Ing. Kilian Lackhove > Fachgebiet für Energie- und Kraftwerkstechnik > L1|08 114 > Technische Universität Darmstadt > Otto-Berndt-Straße 3 > D-64287 Darmstadt > Germany > > Tel.: +49 6151 16 - 28915 > Fax: +49 6151 16 - 6555 > e-mail: lackhove(a)ekt.tu-darmstadt.de > > _______________________________________________ > Nektar-users mailing list > Nektar-users(a)imperial.ac.uk > https://mailman.ic.ac.uk/mailman/listinfo/nektar-users > -- *Amitvikram Dutta* Graduate Research Assistant Fluid Mechanics Research Lab Multi-Physics Interaction Lab University of Waterloo

2 1

Nektar++ installation in a cluster
by Nitish Arya 30 Oct '18

30 Oct '18

Hello. I am planning to use APE solver included in the nektar framework. I have to configure the MPI using Intel MPI which is available in the cluster, while dependencies are installed using an internet connection. The internet connection is only available at the head node in which I am unable to invoke the Intel MPI. So it means that I have to compile the dependencies separately. I tried compiling the ThirdParty dependencies separately but when I compiled Nektar, it was again searching for the dependencies via http or https ( both of the protocols are also not working ). Is there any way I can install Nektar without the installer searching for the dependencies in the internet. Sincerely, Nitish

1 0

Re: [Nektar-users] FW: FW: Questions regarding APE Solver
by Amitvikram Dutta 30 Oct '18

30 Oct '18

Hi Killian, I was hoping to get your take on my setup for the APESolver with forcing from pts files and baseflow from fld files The appropriate sections in my .xml file are as follows * <BOUNDARYCONDITIONS> <FUNCTION NAME="Baseflow">  <F VAR="*" TIMEDEPENDENT="1" FILE="nzs1_%4.fld"/> </FUNCTION> <FUNCTION NAME="Source"> <E VAR="p" VALUE="0"/> <F VAR="u" TIMEDEPENDENT="1" FILE="nzs1_vort_ptslamb_%4.pts:L_x"/> <F VAR="v" TIMEDEPENDENT="1" FILE="nzs1_vort_ptslamb_%4.pts:L_y"/> <F VAR="w" TIMEDEPENDENT="1" FILE="nzs1_vort_ptslamb_%4.pts:L_z"/> </FUNCTION> </BOUNDARYCONDITIONS> <FUNCTION NAME="InitialConditions"> <F VAR="*" FILE="nzs1_1500.fld" /></FUNCTION></CONDITIONS><FORCING> <FORCE TYPE="Field"> <FIELDFORCE> Source <FIELDFORCE/> </FORCE> </FORCING>* Where the base flow is taken from files from an incompressible simulation that was run earlier (named as nzs1_1500.fld - nzs1_1503.fld), and the forcing from .pts files which contain the lamb vector fluctuations (named as nzs1_vort_ptslamb_1500.pts-nzs1_vort_ptslamb_1503.pts). The solver fails with a segmentation fault (text shown below), which seems rather strange to me as none of the files involved are particularly large (10 Mb or so). *[nia0412:347383:0:347383] Caught signal 11 (Segmentation fault: address not mapped to object at address 0x10)==== backtrace ==== 0 0x000000000019b649 Nektar::MultiRegions::ExpList::GetBoundaryCondition() ???:0 1 0x000000000016cac5 Nektar::MultiRegions::DisContField2D::GenerateBoundaryConditionExpansion() ???:0 2 0x00000000001730d5 Nektar::MultiRegions::DisContField2D::DisContField2D() ???:0 3 0x000000000013ea91 Nektar::MemoryManager<Nektar::MultiRegions::DisContField2D>::Allocate<boost::shared_ptr<Nektar::LibUtilities::SessionReader>, boost::shared_ptr<Nektar::SpatialDomains::MeshGraph>, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >() ???:0 4 0x00000000001360c6 Nektar::SolverUtils::EquationSystem::v_InitObject() ???:0 5 0x000000000017b6c3 Nektar::SolverUtils::UnsteadySystem::v_InitObject() ???:0 6 0x00000000004378ac Nektar::APE::v_InitObject() ???:0 7 0x000000000043ada6 Nektar::APE::create() ???:0 8 0x000000000011608a Nektar::LibUtilities::NekFactory<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, Nektar::SolverUtils::EquationSystem, boost::shared_ptr<Nektar::LibUtilities::SessionReader> const&, Nektar::LibUtilities::none, Nektar::LibUtilities::none, Nektar::LibUtilities::none, Nektar::LibUtilities::none, Nektar::LibUtilities::none>::CreateInstance() ???:0 9 0x0000000000115062 Nektar::SolverUtils::Driver::v_InitObject() ???:010 0x000000000012651e Nektar::SolverUtils::DriverStandard::create() ???:011 0x000000000042f926 Nektar::LibUtilities::NekFactory<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, Nektar::SolverUtils::Driver, boost::shared_ptr<Nektar::LibUtilities::SessionReader> const&, Nektar::LibUtilities::none, Nektar::LibUtilities::none, Nektar::LibUtilities::none, Nektar::LibUtilities::none, Nektar::LibUtilities::none>::CreateInstance() ???:012 0x0000000000423b99 main() ???:013 0x0000000000021c05 __libc_start_main() ???:014 0x0000000000426dcd _start() ???:0===================Segmentation fault* Any insight would be greatly appreciated. Sincerely, On Fri, Oct 26, 2018 at 7:58 AM Amitvikram Dutta < amitvikram.dutta(a)uwaterloo.ca> wrote: > *From:* nektar-users-bounces(a)imperial.ac.ukOn Behalf OfAmitvikram Dutta > *Sent:* October 26, 2018 7:57:47 AM (UTC-05:00) Eastern Time (US & Canada) > *To:* nektar-users > *Subject:* Re: [Nektar-users] FW: Questions regarding APE Solver > > Thanks a lot Killian, > > Is there a tentative release date for Nektar 5? > > On Fri, Oct 26, 2018, 3:58 AM Amitvikram Dutta < > amitvikram.dutta(a)uwaterloo.ca> wrote: > >> >> >> >> ________________________________________ >> From: nektar-users-bounces(a)imperial.ac.ukOn Behalf OfKilian Lackhove >> Sent: October 26, 2018 3:58:07 AM (UTC-05:00) Eastern Time (US & Canada) >> To: Amitvikram Dutta; nektar-users >> Subject: Re: [Nektar-users] Questions regarding APE Solver >> >> Hi Amitvikram, >> >> 1. Yes, only the lamb vector was used, so the RHS of the pressure >> equation is zero. >> >> 2. It seems the userguide is out of date. In Nektar++ 4.4 the RHS are >> specified using forcings (see the userguide). See e.g. >> APE_2DPulseAdv_WeakDG_MODIFIED.xml example how this is set up in the xml >> file. In short, you would have something like this: >> >> <CONDITIONS> >> ... >> <FUNCTION NAME="Source"> >> <E VAR="p" VALUE="0"/> >> <E VAR="u" VALUE="define lamb vector here"/> >> <E VAR="v" VALUE="define lamb vector here"/> >> <E VAR="w" VALUE="define lamb vector here"/> >> </FUNCTION> >> </CONDITIONS> >> <FORCING> >> <FORCE TYPE="Field"> >> <FIELDFORCE> Source <FIELDFORCE/> >> </FORCE> >> </FORCING> >> >> 3. The next Nektar++ major release will ship a fieldfromstring module, >> which should be able to compute the source terms without the pts detour. >> Until then you can just use "FieldConvert foo.xml foo.chk bar.pts" to >> obtain a pts file with the values at the quadrature points. >> >> In general, Nektar++ 5.0.0 will contain many changes and improvements to >> the acoustics solver, in particular with regard to hybdrid CAA >> cosimulations, so you might want to give it a try once its released. >> >> Cheers, >> >> Kilian >> >> >> Am 26.10.18 um 04:55 schrieb Amitvikram Dutta: >> > Hi everyone, >> > >> > I'm trying to use the APE Solver in nektar++ to predict sound generation >> > from a particular industrial flow. The work is similar in scope to that >> > published in this recent paper : >> > http://www.iccfd.org/iccfd10/papers/ICCFD10-248-Paper.pdf. >> > I'm particularly interested in how to define the source term. If I >> > understand the work correctly, in a flow dominated by coherent >> > structures, the Lamb vector becomes the dominant source. >> > This leads to my first question : >> > 1.In the APE-4 formulation there are 2 coupled equations with seperate >> > source terms - one for the pressure fluctuations and one for the >> > velocity fluctuations. In the above paper when the authors state that >> > only the lamb vector was considered as a source, was the source for the >> > RHS of the pressure fluctuations set to zero? >> > 2. If this was the case, how was this specified in the .xml file? The >> > only example I have come across is in the user guide which shows >> > >> > <FUNCTION NAME="Source"> >> > <E VAR="S" VALUE="0" /> >> > </FUNCTION> >> > >> > If one wanted to specify the pressure and velocity source terms >> > seperately how would one do it in the .xml file? >> > >> > 3. From older posts in the mailing list I have seen suggestions of a >> > workflow where the .chk/.fld files are written into a .pts file and then >> > post processed to obtain the acoustic source. Is there a relevant filter >> > or an extension the FieldConvert module that allows direct conversion >> > from a .fld/.chk to a .pts format? I know of the interppoints extension >> > that allows export to a .dat format but not the .pts? >> > >> > Sorry for the lengthy post. >> > Thanks in advance! >> > -- >> > >> > *Amitvikram Dutta* >> > >> > Graduate Research Assistant >> > >> > Fluid Mechanics Research Lab >> > >> > Multi-Physics Interaction Lab >> > >> > University of Waterloo >> > >> > >> > _______________________________________________ >> > Nektar-users mailing list >> > Nektar-users(a)imperial.ac.uk >> > https://mailman.ic.ac.uk/mailman/listinfo/nektar-users >> > >> >> -- >> Dr.-Ing. Kilian Lackhove >> Fachgebiet für Energie- und Kraftwerkstechnik >> L1|08 114 >> Technische Universität Darmstadt >> Otto-Berndt-Straße 3 >> D-64287 Darmstadt >> Germany >> >> Tel.: +49 6151 16 - 28915 <+49%206151%201628915> >> Fax: +49 6151 16 - 6555 <+49%206151%20166555> >> e-mail: lackhove(a)ekt.tu-darmstadt.de >> >> _______________________________________________ >> Nektar-users mailing list >> Nektar-users(a)imperial.ac.uk >> https://mailman.ic.ac.uk/mailman/listinfo/nektar-users >> > -- > > *Amitvikram Dutta* > > Graduate Research Assistant > > Fluid Mechanics Research Lab > > Multi-Physics Interaction Lab > > University of Waterloo > -- *Amitvikram Dutta* Graduate Research Assistant Fluid Mechanics Research Lab Multi-Physics Interaction Lab University of Waterloo

2 1

Regarding BiGlobal stability in Nektar++
by Abhishek Kumar 26 Oct '18

26 Oct '18

Hi, I am trying to understand the session file "PPF_R15000_3D_VCS.xml." I have two confusion: 1) Since modeType is HalfMode than why "HomModesZ" is set to 2. I think this should be 1. 2) Since the base flow is 2D2C, then is it necessary to include variable "w" in the base flow? In general, can we use a base flow generated from DNS which only have variables (uvp) in the session file having FIELDS = (uvwp)? I would be highly thankful for the comments on this issue. With regards Abhishek Kumar --------------------------------------------------------------------------------------- Abhishek Kumar Research Associate Applied Mathematics Research Centre Coventry University, Coventry CV1 5FB The United Kingdom Email # ac7600(a)coventry.ac.uk, abhishek.kir(a)gmail.com ---------------------------------------------------------------------------------------

2 4

Questions regarding APE Solver
by Amitvikram Dutta 26 Oct '18

26 Oct '18

Hi everyone, I'm trying to use the APE Solver in nektar++ to predict sound generation from a particular industrial flow. The work is similar in scope to that published in this recent paper : http://www.iccfd.org/iccfd10/papers/ICCFD10-248-Paper.pdf. I'm particularly interested in how to define the source term. If I understand the work correctly, in a flow dominated by coherent structures, the Lamb vector becomes the dominant source. This leads to my first question : 1.In the APE-4 formulation there are 2 coupled equations with seperate source terms - one for the pressure fluctuations and one for the velocity fluctuations. In the above paper when the authors state that only the lamb vector was considered as a source, was the source for the RHS of the pressure fluctuations set to zero? 2. If this was the case, how was this specified in the .xml file? The only example I have come across is in the user guide which shows <FUNCTION NAME="Source"> <E VAR="S" VALUE="0" /> </FUNCTION> If one wanted to specify the pressure and velocity source terms seperately how would one do it in the .xml file? 3. From older posts in the mailing list I have seen suggestions of a workflow where the .chk/.fld files are written into a .pts file and then post processed to obtain the acoustic source. Is there a relevant filter or an extension the FieldConvert module that allows direct conversion from a .fld/.chk to a .pts format? I know of the interppoints extension that allows export to a .dat format but not the .pts? Sorry for the lengthy post. Thanks in advance! -- *Amitvikram Dutta* Graduate Research Assistant Fluid Mechanics Research Lab Multi-Physics Interaction Lab University of Waterloo

2 1

Regarding base flow for HalfMode
by Abhishek Kumar 23 Oct '18

23 Oct '18

Hi, I am trying to do linear stability analysis using HalfMode, with the two-dimensional base flow and three-dimensional perturbation for the buoyancy-driven problem. I provide a base flow with (u,v,c,p), where c is a scalar. I am getting the following error: "For half mode, expect 2D2C base flow." Is there any solution to include a variable "c" in the base flow? With regards Abhishek --------------------------------------------------------------------------------------- Abhishek Kumar Research Associate Applied Mathematics Research Centre Coventry University, Coventry CV1 5FB The United Kingdom Email # ac7600(a)coventry.ac.uk, abhishek.kir(a)gmail.com ---------------------------------------------------------------------------------------

1 0

Regarding linear stability solver in parallel
by Abhishek Kumar 23 Oct '18

23 Oct '18

Hi, I am trying to test the linear stability solver in parallel. I did a simple test with the channel flow with a single processor and on 8 processors. I observed the following: Time taken per 1000 steps for single processor = 2.76512s Time taken per 1000 steps for 8 processors = 24.7811s I expected a reduction in the computational time while running parallel. Is there any issue in the parallel implementation of the linear stability solver? Is something related with the eigenvalue solver? I would be highly thankful for the comments on this issue. With regards Abhishek Kumar --------------------------------------------------------------------------------------- Abhishek Kumar Research Associate Applied Mathematics Research Centre Coventry University, Coventry CV1 5FB The United Kingdom Email # ac7600(a)coventry.ac.uk, abhishek.kir(a)gmail.com ---------------------------------------------------------------------------------------

2 2

Routine to create *chk
by WEI HE 19 Oct '18

19 Oct '18

Dear all, Does anyone know how to load *vtu files by. nektar++? Or the rules to export the binary check file *.chk? Thanks, Regards, Wei

4 4

Re: [Nektar-users] Problem with msh to xml conversion using NekMesh
by Oliviu Sugar 18 Oct '18

18 Oct '18

Hi Dave, I checked out the latest version of Master from the repository yesterday. I also ran the NekMesh regression tests including the gmsh version 4 files, and all test completed successfully. This is what led me to think I might have made a mistake with the geo file, but if this is the case it is not evident. Best,Oliviu On Wednesday, October 17, 2018, 9:00:25 PM GMT+1, David Moxey <D.Moxey(a)exeter.ac.uk> wrote: Hi Oliviu, The problem is likely gmsh 4, which was only recently released and changed the msh file format. Thanks to Julian, one of our PhD students, we have recently added support for this to NekMesh: https://gitlab.nektar.info/nektar/nektar/merge_requests/964 However you will need to upgrade/compile your own version of the latest Nektar++ master to get this functionality. Alternatively you can use Gmsh v3.x to generate your msh file instead. Many thanks, Dave > On 17 Oct 2018, at 20:43, Oliviu Sugar <oliviu_sugar(a)yahoo.com> wrote: > > Hi Stanislaw, > > All lines and surfaces in the gmsh script have been defined as transfinite, in order to force a structured mesh generation, and I have verified that all elements are quads. > > Best regards, > Oliviu > > On Wednesday, October 17, 2018, 8:38:33 PM GMT+1, Stanisław Gepner <sgepner(a)meil.pw.edu.pl> wrote: > > > Hi, a wild guess would be that you have a mix of triangles and quads in a composite. From what I remember this is not supported. > Pozdrawiam, > Stanisław Gepner > > Dnia 17 października 2018 21:32:19 CEST, Oliviu Sugar <oliviu_sugar(a)yahoo.com> napisał(a): > Dear all, > > I've run into some problems when trying to convert a msh file (generated using gmsh 4.0.2) into a xml file, using NekMesh. > The script geo file used with gmsh is attached. > The error encountered by NekMesh is: "Different types of elements in same composite!" > Any advice on what needs to be changed in the geo file to resolve this issue would be much appreciated. > > Best regards, > Oliviu > > _______________________________________________ > Nektar-users mailing list > Nektar-users(a)imperial.ac.uk > https://mailman.ic.ac.uk/mailman/listinfo/nektar-users

1 1

Problem with msh to xml conversion using NekMesh
by Oliviu Sugar 17 Oct '18

17 Oct '18

Dear all, I've run into some problems when trying to convert a msh file (generated using gmsh 4.0.2) into a xml file, using NekMesh.The script geo file used with gmsh is attached.The error encountered by NekMesh is: "Different types of elements in same composite!" Any advice on what needs to be changed in the geo file to resolve this issue would be much appreciated. Best regards,Oliviu

2 2