- Nektar-users - imperial.ac.uk

Robin bcs with value dependent on other fields
by Vidhi Zala 16 Jun '17

16 Jun '17

Hello all, I have questions about a very specific boundary condition which I am trying to implement. For UnsteadyAdvectionDiffusion case, consider 3 fields - u1,u2 and u3. For u1, I have Robin bc with "Value" dependent on values of u2 and u3. I tried the following: I made the robin bc time dependent in order to force it to evaluate at each timestep. Then to enforce the "Value" part, I have to write solutions for u2 and u3 in a file ".bc" and load it up at each evaluate call for u1's robin bc. I was looking into examples to do this but could only find information about how to make the bcs load from file. The *.bc files in test folder of ADRSolver are compressed versions. It would be helpful if you could point me out to: 1. How to expand the *.bc file from given test examples 2. How to write out solutions for u2 and u3 at each timestep to a .bc file which u1 will utilize in next timestep 3. Is there a better way to do this? Thanks ! -- ,-~~-.___. ._. / | ' \ | |--------. Vidhi Zala ( ) 0 | | | School of Computing \_/-, ,----' | | | Scientific Computing and Imaging Institute ==== !_!-v---v--. University of Utah / \-'~; .--------. / __/~| ._-""|| | vidhi.zala(a)utah.edu =( _____|_|____||________| http://sci.utah.edu/people/vidhi.html

1 0

No convergence in a Compressible Flow Simulation
by Ismael El Houas Ghouddana 15 Jun '17

15 Jun '17

Dear nektar developers, I am working on a supersonic intake for my bachelor degree project. The problem is that the simulation is not able to converge. I want to ask you some advice in order to fix this problem. First of all, your teamate Daniele De Grazia has helped me before, in order to improve the session file. However, the simulation continues being not able to converge. Some data of the session file: -I am studying the problem through the Euler equations. -As boundary conditions, I use far field at the inlet, upper, lower and outlet regions. -I use a non collocated base with a number of points which are the double of the number of modes. Finally, I attach the session file and the last .chk before the simulation blows up,so you can understand better the problem and the conditions that I am using. Kind Regard, Ismael

1 0

GitLab maintenance
by David Moxey 13 Jun '17

13 Jun '17

Hi all, Apologies for the short notice but there are a number of performance and security updates I need to apply to GitLab before tomorrow's workshop. I'm taking it offline briefly to perform an upgrade -- it will be back up in around 30 minutes. Many thanks, Dave

1 0

Workshop programme
by Sherwin, Spencer J 05 Jun '17

05 Jun '17

Hi, A draft programme for the Nektar++ workshop is now available under: http://www.nektar.info/community/workshops/nektar-workshop-2017/ Looking forward to seeing many of you next week. Cheers, 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

Floquet analaysis
by F Mellibovsky 04 Jun '17

04 Jun '17

Dear all, I have computed a stable *2D periodic solution* with nektar++ time evolving the *incompressible Navier-Stokes solver*. Now I would like to analyse it's *stability to spanwise-dependent perturbations*. I have started doing this with time evolution, by using a homogeneous spanwise direction with just one mode and monitoring its growth rate. This is however extremely slow. I have noticed that in the user guide there's an example that does Floquet stability analysis on the periodic 2D vortex shedding solution of the flow around a cylinder, but unfortunately I haven't manage to find the conditions xml file used for the computations. Does anyone have it, so that I can use it to set up my *Floquet analysis*? I have doubts as to how a sequence of snapshots along a period are loaded and also about the differences between using Arpack or modified Arnoldi, among other issues. I would be very greatful if someone can send me an conditions file exemplifying the implementation of Floquet analysis. Many thanks in advance. Cheers, Fernando Mellibovsky

4 6

Re: [Nektar-users] NEKTAR ++
by Sherwin, Spencer J 08 May '17

08 May '17

Dear Hamed, Thanks for your interest in our software. I am not a user of visual studio but am cc’ing one of the researchers, Julian, who does use Nektar++ in this environment. Perhaps you could clarify what problems you are having and we can see if Julian can help. Regards, Spencer. On 1 May 2017, at 20:16, Hamed Faghanpour Ganji <h.f.ganji(a)gmail.com<mailto:h.f.ganji@gmail.com>> wrote: Dear Dr. Sherwin, Greeting! My name is Hamed Ganji and I am a young researcher in Acoustic fluid solid interaction, I see your invaluable publications and also your developed software (Nektar++) is found very interesting for my major goal in current project, really I would like to use it in my future works, but my familiar with this software and Visual studio and other related package according to NEKTAR ++ help in pages 9-16, so I could not use this one till at present, May you please help me with your guidance that how can use this useful open source codes. With Best Regards, Hamed Ganji 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

2 1

Re: [Nektar-users] probing variables over 2D mesh
by Sherwin, Spencer J 07 May '17

07 May '17

Dear Ronald, Very sorry for a slow response. I have not been keeping onto of emails this past week. I am forwarding your request to the nektar-users list to see if anyone else has an answer. We do not have a specific function for this currently unless someone else has implemented it. We do have some features in FieldConvert to extract a box or plane of points. Whilst these methods are now callable from the solvers this would be very costly since it would search out for the points in the mesh at every call and to make this more efficient you should ideally only be searching for the points to extract once and then only interpolate every time you wish to extract this information. This is currently how the history points filter works. Cheers, Spencer. On 29 Apr 2017, at 10:06, ronald russ <rnld_russ(a)yahoo.com<mailto:rnld_russ@yahoo.com>> wrote: Hello, I am currently doing 3D ADR simulations on a 3D domain. What I really need is to probe the variables, but not on specific point(s), on a 2D mesh of a unit sphere which is contained in the 3D domain. This means probing on nodes of the triangular grid of the sphere which is in 2D. Please could you let me know if there is an straightforward way to probe fields on the nodes on this 2D mesh. Best, Ronald 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

Polyhedral Meshes
by Kurt Sansom 26 Apr '17

26 Apr '17

I haven't tried using a polyhedral mesh with Nektar++ and the documentation only mentions hexahedral and tetrahedral meshes. Is anyone using polyhedral meshes with nektar++. What would that take? Regards, ~Kurt -- Kurt Sansom

1 0

2 bugs and 1 question
by Ankang Gao 24 Apr '17

24 Apr '17

Hellow, I have found two bugs in the codes, and I listed them in the attachment. I have a question about why the backward differentiation formula starts after m_pressureCalls>2, in function void Nektar::Extrapolate::AccelerationBDF(Array< OneD, Array< OneD, NekDouble > > & array). Since when m_pressureCalls=2, we can already get a first order accuracy result. thanks Ankang

5 12

Error in Compressible Flow Solver
by Ismael El Houas Ghouddana 22 Apr '17

22 Apr '17

Dear developers of Nektar++, My name is Ismael El Houas and I am an aeronautical engineer student from the UPC (Universitat Politècnica de Catalunya). Nowadays, I am working on my bachelor's degree final project. For my project, I am working with Nektar++ in order to make a CFD study of a supersonic diamond airfoil, so I am using compressible flow solver. I am considering a symmetrical solution, so I decided to simplify the problem and just study half of the domain. To perform this study, half of the domain, I impose symmetrical boundary conditions on the lower boundary. However, when I study the whole problem the simulation converges to the solution, but when I study just the half of the domain it does not converge. The reported error says: "vacuum is generated". After different tests, I could not find a solution to fix the problem. The boundary conditions I am imposing on the lower boundary for symmetrical solution are: Dirichlet = 0 condition in rhov in order to avoid entrance of fluid from this side, and Neuman = 0 for the other variables (rho,rhou and E). After different tests, I could not find a solution to fix the problem. For this reason, I ask you for some tips in order to be able to find a solution to this problem. Thank you for your attention and congratulations for your work. Kind Regards, Ismael El Houas

2 1