- Nektar-users - imperial.ac.uk

Nectar-Developer: Find each element's coefficients from ExpList
by ashok jallepalli 15 Feb '16

15 Feb '16

Hi everyone, I currently working on a project and would like to do the following task and need some guidance. (1) Read an .fld file. (2) Load it into vector fields into expansions (exp_u, exp_v, exp_w) (3) Loop through each element in the mesh and do some processing in it and store the result in a new expansion. I have a problem in step (3). I am able loop through each expansion but unable to understand how should I retrieve coefficients for each element. Array<OneD,NekDouble> gcoeff = exp_u->GetCoeffs() - This gets me the coefficients of whole expansion. How do I match coefficients which belong to each element. These two statements used in the code seem to reorder the coefficients in the expansion, Can someone explain what they are doing. Exp_u->GlobalToLocal() Exp_u->LocalToGlobal() Would any of these functions reorder coefficients elementwise. So that I use access element[i]'s coefficients as gcoeff[I*numOfCoeffPerEle]. Ashok.

2 2

Parallel transposition not implemented yet for 3D-Homo-2D approach
by ceeao 14 Feb '16

14 Feb '16

Dear all, I just installed the library, and need to simulate DNS of a channel flow with oscillating pressure gradient. As I have two homogeneous directions I applied standard Fourier discretization in these directions. It seems like this case is not parallelized yet, and I got the error in the subject. I was wondering if I'm overlooking something. If not, are there maybe any plans in the future to include parallelization of 2D FFT's? Thank you in advance. Best, Asim Onder Research Fellow National University of Singapore ________________________________ Important: This email is confidential and may be privileged. If you are not the intended recipient, please delete it and notify us immediately; you should not copy or use it for any purpose, nor disclose its contents to any other person. Thank you.

2 5

EulerCFD in 1D with variable coefficients in mass and stiffness matrix
by Kühnelt Helmut 08 Feb '16

08 Feb '16

Dear Spencer, How can I implement or modify the EulerCFE code (only 1D is needed) to obtain a variant with mass and stiffness matrices with variable coefficients (i.e. M_A = \int_{-1}^{1} L_q L_p A(\xi,t) d\xi), as well as setting up similar matrices with variable coefficients for setting up source terms (e.g. M_s S_p). Best regards, Helmut

2 1

Changelog file
by David Moxey 07 Feb '16

07 Feb '16

Dear all, A quick note for you all; the latest master incorporates a CHANGELOG.md file, which catalogs changes as we develop the code. In order to keep things update, can I ask that, once you submit a merge request, you just add a short description of your change to the file and put a reference to the merge request number. Hopefully the layout should be self-explanatory, but if not let me know! Cheers, Dave -- David Moxey (Research Associate) d.moxey(a)imperial.ac.uk | www.imperial.ac.uk/people/d.moxey Room 363, Department of Aeronautics, Imperial College London, London, SW7 2AZ, UK.

1 0

Compressed format for XML files
by Sherwin, Spencer J 07 Feb '16

07 Feb '16

Hi, We have just merged into master a modification where the input mesh files produced by MeshConvert will be stored in a compressed format to make large input files much smaller and to also help with the start up and processing of input files since previously they were read in a very verbose format. This does mean however that input files sections of Vertices, Edges, Faces, Elements and Curves are no longer directly readable. If you wish to see an ascii format file you need to add :xml:uncompress to the end of the .xml file, i.e. MeshConvert file.msh out.xml:xml:uncompress There is also documentation in the User-guide of the master version. Please let us know if you have any problems with this option. Best regards, Spencer. Spencer Sherwin McLaren Racing/Royal Academy of Engineering Research Chair, Professor of Computational Fluid Mechanics, Department of Aeronautics, Imperial College London South Kensington Campus London SW7 2AZ s.sherwin(a)imperial.ac.uk<mailto:s.sherwin@imperial.ac.uk> +44 (0) 20 759 45052

1 0

Logfile for Nektar
by Amitvikram Dutta 02 Feb '16

02 Feb '16

Hi, Is there some form of a logfile implementation for Nektar++ which one can query as a simulation is proceeding in order to monitor convergence? Sincerely,

3 4

Nektar++ 4.2.0 and Nektar++ 3.2.0 compatibility issues
by Amitvikram Dutta 02 Feb '16

02 Feb '16

Hi, I've been using Nektar++ for DNS studies of Laminar Separation Bubbles over airfoils. So far I've been running Nektar++4.2.0 on my personal workstation. However, now that I am dealing with meshes containing more than 200k elements, I'll start submitting jobs to a cluster. The cluster however has Nektar++3.2.0 installed. Might there be any compatibility issues for the cases that I have been running so far? I have been using the IncNavierStokesSolver with Galerkin projection and a time-integration method of IMEXOrder2. Sincerely, *Amitvikram Dutta* *MASc Candidate* *Graduate Research Assistant * *Okanagan CFD Laboratory* *University of British Columbia | Okanagan Campus*

2 2

basis functions for Discontinuous Galerkin
by Manh Nguyen 01 Feb '16

01 Feb '16

Dear all, I am a new Nektar++ user. I want to use the Advection-Diffusion-Reaction Solver where Discontinuous Galerkin method is chosen (property "Projection" has the value "Discontinuous") to solve 2D Unsteady Advection equations over a mesh comprised of both triangles and quads. I have two questions. Q1) When TYPE = "MODIFIED" (in the "EXPANSIONS" node in a xml file), what are the basis functions over a triangle, over a quad used? The user guide v4.2.0, Section 3.2 addresses this part only generally. Q2) The approximate solution is a linear combination of basis functions. How can one export the coefficients of the linear combination to an output file? I opened a file *.fld, those coefficients seem to be written inside the node <"ELEMENTS FIELDS="u"...> but the output coefficients are in some kind of encrypted format. Are there any documents that I can find answers to the two questions? Of course, I appreciate very much if someone can answer those questions via this mailing list. Thank you very much in advance, Manh Nguyen Department of Computer & Information Science & Engineering, University of Florida

2 2

ExpList::SmoothField produces artifacts
by Kilian Lackhove 01 Feb '16

01 Feb '16

Dear all, I want to create a C0-continuous field from interpolated input data. The input field is obtained from a FVM code and was filtered with a gaussian in space and interpolated to the quadrature points. I store the field in a ContField2D m_bf = Array<OneD, MultiRegions::ExpListSharedPtr>(m_spacedim + 2); m_bf[i] = MemoryManager<MultiRegions::ContField2D>::AllocateSharedPtr(m_session, m_graph); copy the interpolated data into its m_phys array Vmath::Vcopy(nq, filteredField[i], 1, m_bf[i]->UpdatePhys(), 1); and then try to smooth the field. Until now, i just did: m_bf[i]->IProductWRTBase(m_bf[i]->GetPhys(), m_bf[i]->UpdateCoeffs()); m_bf[i]->MultiplyByElmtInvMass(m_bf[i]->GetCoeffs(), m_bf[i]->UpdateCoeffs()); m_bf[i]->BwdTrans(m_bf[i]->GetCoeffs(), m_bf[i]->UpdatePhys()); which yields the field depicted in DG.png. However, this does not ensure C0- continuity, so i tried to call ExpList::SmoothField afterward: m_bf[i]->SmoothField(m_bf[i]->UpdatePhys()); However, this produces a distorted field as shown in smoothed.png. I am using a 6th order modified expansion basis: <E COMPOSITE="C[10]" NUMMODES="6" FIELDS="p,u,v,w" TYPE="MODIFIED"/> and the mesh is depicted in smoothed_mesh.png. What is the cause of this problem and what can i do do fix it? Cheers, Kilian -- Kilian Lackhove, M.Sc. Fachgebiet für Energie- und Kraftwerkstechnik L1|08 114 Technische Universität Darmstadt Jovanka-Bontschits-Straße 2 D-64287 Darmstadt Germany Tel.: +49 6151 16 - 28915 Fax: +49 6151 16 - 6555 e-mail: lackhove(a)ekt.tu-darmstadt.de

2 4

Linear stability analysis of a three-dimensional baseflow
by Siddarth Chintamani 20 Jan '16

20 Jan '16

Hello, I am trying to perform stability analysis of linearized incompressible Navier Stokes equation of a three-dimensional baseflow. I understand that Nektar++ capabilities for stability analysis are limited to 2D or Quasi3D baseflow representation. Unfortunately, my problem has to be solved as 3D. Is there any way to relax the 2D or quasi3D representation? Any help will be greatly appreciated. Thank you, Siddarth Chintamani

3 9