- Nektar-users - imperial.ac.uk

Compressible flow simulation initiation with Blasius flat-plat boundary layer
by devabratasahoo＠campus.technion.ac.il 31 Jan '17

31 Jan '17

Hi All, Can anyone please suggest me how to initiate the case of compressible flow over an open cavity with Blasius flat-plate boundary layer. Further, my simulation getting killed even before getting initiated when I am using a mesh with above 0.1million elements. Is it because of the low-performance computing resources used? Regards, Dev

1 0

linearised Navier-Stokes element matrix
by Martin Hess 26 Jan '17

26 Jan '17

Dear All, I would like to use the incompressible Navier-Stokes solver for parametric model order reduction. As a first step, I thought I have a look at the assembled system matrix in the steady stokes case, since I need access to these matrices in some point. What I found in the code are several layers of static condensation, but no access to the complete linearised Navier-Stokes element matrix. Could I somehow form that matrix from the available submatrices formed in CoupledLinearNS.cpp? Best, Martin Hess

2 1

Using high oder data as Initial Condition to a low order simulation
by Mehak 20 Jan '17

20 Jan '17

Dear all, I have generated a high order data for Quasi-3D: Homogeneous in z-direction (Number of homogeneous mode =8). Now i want to use the result as a initial condition for another simulation where i have (Number of homogeneous mode =7 or less). My solution is very smooth. The simulation is crashing with the segmentation fault error. EquationType: UnsteadyNavierStokes Spatial Dim.: 3 Max SEM Exp. Order: 6 Quasi-3D: Homogeneous in z-direction Expansion Dim.: 3 Num. Hom. Modes (z): 6 Projection Type: Continuous Galerkin Advection: explicit Diffusion: implicit Integration Type: IMEXOrder2 Any input from you will be very helpful. *Thanks * *Mehak naina oberoi * *mehak.n.oberoi(a)gmail.com <mehak.n.oberoi(a)gmail.com>*

2 1

checkpoint fields
by Asım Önder 12 Jan '17

12 Jan '17

Dear all, I've recently updated my local Nektar++ version to the latest one in the repository. In this new version, it looks like there are some changes in checkpoint filter. In particular, once in a while the code writes out chk files with a "bak0" extension. What would this extension supposed to mean? I did some research but couldn't find. I also noticed that the chk file for the same time instance is slightly larger compared to its "bak0" counterpart and to the rest of the fields, for instance: 62G config_22.chk 62G config_23.chk 62G config_24.bak0.chk 64G config_24.chk 62G config_25.chk 62G config_26.chk 62G config_27.chk 62G config_28.chk 62G config_29.chk 62G config_30.chk 62G config_31.chk 62G config_32.chk 62G config_33.chk 62G config_34.chk 62G config_35.chk 62G config_36.chk 62G config_37.chk 62G config_38.bak0.chk 64G config_38.chk It is now a little confusing, which one to keep, bak0 version or the other version, i.e. should I delete for example config_38.bak0.chk or config_38.chk? I'd appreciate any feedback on this. Best, Asim

4 5

Using PhysDeriv() for gradient-based source terms
by Kilian Lackhove 12 Jan '17

12 Jan '17

Dear List, i am trying to implement a source term based on the gradient of an auxiliary field in the APESolver. My problem is quite general, so i in terms of the simpler ADR solver, a minimal example would look like this: du / dt + v * grad(u) = c * grad(v) with u being a transported variable, v the advection velocity and c a constant. The problem is the gradient on the right hand side. I tried computing it by using ExpList::PhysDeriv(), but since the projection is not C1-continuous, this fails at the element edges. I attached an example plot to illustrate the problem. p0/rho0 uses the left axis, the gradient the right hand side one. Do you have any suggestions on how to handle the gradient term? Is there some other method i should use? 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 2

Compressible flow solver problem
by devabratasahoo＠campus.technion.ac.il 02 Jan '17

02 Jan '17

________________________________ Hi Users, I am solving my cavity flow case at a Re= 0.3x10^7 per mtr, M=0.15, and NumModes=4. I am not using the CFL tool, instead I am taking a constant time step = 0.0001. The simulation is running fine now but the problems now remains as follows: 1) The residuals are not good at all. (pics attached below) 2) I am still not able to capture the boundary layers in this mesh though the flow in cavity is captured but it can't be accurate. (Present mesh attached below) 3) My final aim is to run the simulation at a larger RE (1X10^7 per mtr at M-0.5). (presently used solver info, parameters and expansions used are attached below) 4) I also need to run the CFL tool to get the mean flow which I can further use as base flow for the stability analysis. I am trying it out but the solution is terminating when I use CFL tool. It is working only when I use CFL=0.1. 1) Residual [cid:a9b4a5b0-2a7d-45a9-841f-f78538ca0598] 2)Present Mesh [cid:3afb70e7-e590-47be-ab01-03fe8b38b5fd] 3) Solverinfo and Parameters used [cid:5f9937c8-6fed-49f2-aa25-153f71a94bea] 4) Expansions used [cid:f4dd55b8-2b6a-4185-becc-c037ff3a5fe2] Result(Mach Contour at 1Sec) [cid:797fa64e-55db-4abc-aa5b-4e6c83a1a06f] I tried the same case at a very low Re = 7500 using Incompressible flow solver and I could capture the boundary layer but the residuals still not good. When I try running it for longer time the solution is blowing up. Can anyone please help me out with where I am going wrong? How to reduce the residual and get a converged solution? Regards, Dev

2 1

Compressible flow solver (Nektar++)
by devabratasahoo＠campus.technion.ac.il 27 Dec '16

27 Dec '16

Hi Fabian, I wanted to know could you come up with any solution of how to use the compressible flow solver in Nektar? I could come up with a better mesh for my case and I am able run the compressible flow solver with P-4, but residuals are still good at all. Below is the mach contour obtained at time 1sec simulated at a time step of 0.0001. [cid:797fa64e-55db-4abc-aa5b-4e6c83a1a06f] Furthermore, if I try to solve it using CFL tool, it terminates for all CFL values more than 0.1(which is not good). What about your status? Can anyone help me out of this situation? Regards, Dev

2 1

Need book exercise solutions
by Prashant Tarey 27 Dec '16

27 Dec '16

Hello All Nice to see so many Nektar++ users. Actually I am trying to work on the exercise problems given in the book "Spectral/hp Element Methods for Computational Fluid Dynamics: Second Edition- by George Karniadakis and Spencer J. Sherwin". And the solutions were given on Prof Sherwin's website, which is not opening since last 4-5 days. If someone has the copy of solutions for Chapter 3 and 4, please mail me at tarey.prashant(a)gmai.com Thanks!! Prashant

1 0

Nonphysical terminated simulation - CompressibleFlowSolver
by Selbach, Fabian 16 Dec '16

16 Dec '16

Dear users, I hope you can help one more time. I got nonphysical oscillations during my airfoil-simulation. At that time my simulations terminate with the message: "NaN found during time integration". For more information of the error and parameters see pictures below. I tried to solve this by a higher polynomial degree up to p=7 (this generates 1.5 Mio grid-points here for a 2D simulation), but the simulation terminates still. The time step itself is computed by a fixed CFL=0.9 condition. Maybe this is the wrong way, because this condition is not sufficient? So please can somebody give me a hint about that? Do I have to use another kind of grid, because the resolution of the boundary at the peak of the airfoil is not high enough or the grid lines have no be orthogonal to the surface? I am already using the spherigons-tool and tried to get a smooth boundary of my airfoil. I am using the CompressibleFlowSolver on a parallel run up to 96 CPUs by solving the Euler-Equations (EulerCFE) and the Navier-Stokes-Equations (NavierStokesCFE). The simulation terminates for both. By the way what is the difference between EulerCFE and EulerADCFE (there are no information about it anywhere)? I tried to run the example NACA0012 of the paper "Nektar++: An open-source spectral/hp element framework". The conditions use the EulerADCFE, but by running this example the simulation does not terminate, but also does not change (values of every variable in the whole domain are constant over the time - no changes?). Thanks for any help! Best regards Fabian Visualization of the grid at a polynomial degree of p=2 just before terminating: [cid:71608e01-1455-4c20-abbb-9508cb48f805] With auto-scale: [cid:bba140ce-246a-4689-a539-79ff287727b0] Same solution with scaling by user: [cid:a6bda960-d10e-4f91-a833-2e5537f8758c] Please take into account the scaling above. Error and conditions for example p=2: [cid:0ebd995c-1bae-4c43-9d08-91fe685022bf] [cid:300abb94-0f24-4dde-8cda-f67f326ea4f8] [cid:a43f5fef-3537-4385-a2d9-2f49afbcf006] Visualization of the grid at a polynomial degree of p=7 just before terminating: [cid:4d0f11c2-718a-4f85-860f-e5c90684699e] (a few time steps after that the simulation terminates and the values go against inf...)

5 11

High order outflow boundary conditions
by Sherwin, Spencer J 16 Dec '16

16 Dec '16

Hi, Just a quick note to highlight that we have now merged into Master a branch for high order outflow boundary condition on the incompressible Navier Stokes problems. This merge now supports the “convective like” outflow conditions proposed by Steve Dong (JCP 2015). It also allows for a slightly modified version of the velocity correction scheme where we formulate the pressure as a weak problem rather than starting from the strong form of the pressure Poisson problem. This is also consistent with the work of Guermond & Shen on these scheme. I have tried to motivate both of these modifications by restructuring the description of the method in the user guide. 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

2 2