Hi, I am not sure on that, but I do recall that the Lagrange basis might be only implemented for quadrilateral in 2D and cuboid (?) in 3D elements.

Best regards,
Stanisław Gepner, Ph.D.

Faculty of Power and Aeronautical Engineering
Nowowiejska 21/25 Str.
00-665 Warsaw, Poland
phone +48 (22) 234 51 70

This email from eh.asgari@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 to disable email stamping for this address.

 

Dear All,

I am trying to compare the performance of different expansion types.

However, when I set TYPE="GLL_LAGRANGE_SEM", I get the following error:

=======================================================================
                EquationType: UnsteadyNavierStokes
                Session Name: MeshWith3PrismsLayers
                Spatial Dim.: 3
          Max SEM Exp. Order: 5
              Num. Processes: 128
              Expansion Dim.: 3
             Projection Type: Continuous Galerkin
         Advect. advancement: explicit
        Diffuse. advancement: implicit
                   Time Step: 0.0005
                No. of Steps: 1000000
         Checkpoints (steps): 1000
            Integration Type: IMEX
            Splitting Scheme: Velocity correction (strong press. form)
=======================================================================
Initial Conditions:
  - Field u: 1.0
  - Field v: 0.0
  - Field w: 0.0
  - Field p: 0.0
renaming "MeshWith3PrismsLayers_0.chk" -> "MeshWith3PrismsLayers_0.bak0.chk"
Writing: "MeshWith3PrismsLayers_0.chk"
 (0.186233s, HDF5)
[gra723:25064:0:25064] Caught signal 11 (Segmentation fault: address not mapped to object at address 0x3025)

Here is my setup:

<?xml version="1.0" encoding="utf-8" ?>
<NEKTAR>
  <EXPANSIONS>
    <E COMPOSITE="C[30011]" NUMMODES="5" TYPE="GLL_LAGRANGE_SEM" FIELDS="u,v,w,p" />
  </EXPANSIONS>
  <CONDITIONS>
        <FUNCTION NAME="InitialConditions">
                <E VAR="u" VALUE="1.0" />
                <E VAR="v,w,p" VALUE="0.0" />
        </FUNCTION>
    <SOLVERINFO>
      <I PROPERTY="SolverType"              VALUE="VelocityCorrectionScheme"/>
      <I PROPERTY="EqType"                  VALUE="UnsteadyNavierStokes"    />
      <I PROPERTY="Projection"              VALUE="Continuous"              />
      <I PROPERTY="EvolutionOperator"       VALUE="Nonlinear"               />
      <I PROPERTY="TimeIntegrationMethod"   VALUE="IMEXOrder2"              />
      <I PROPERTY="GlobalSysSoln"         VALUE="IterativeStaticCond" />
      <I PROPERTY="Driver"                  VALUE="Standard"                />
  </SOLVERINFO>
    <GLOBALSYSSOLNINFO>
          <V VAR="u,v,w">
            <I PROPERTY="GlobalSysSoln"       VALUE="IterativeStaticCond" />
            <I PROPERTY="Preconditioner"       VALUE="LowEnergyBlock"/>
            <I PROPERTY="IterativeSolverTolerance"    VALUE="1e-8"/>
          </V>
          <V VAR="p">
            <I PROPERTY="GlobalSysSoln"         VALUE="IterativeStaticCond" />
            <I PROPERTY="Preconditioner"        VALUE="FullLinearSpaceWithLowEnergyBlock"/>
            <I PROPERTY="IterativeSolverTolerance"    VALUE="1e-7"/>
          </V>
       </GLOBALSYSSOLNINFO>

Everything works fine with TYPE="MODIFIED". Noteworthy, I tried TYPE="GLL_LAGRANGE_SEM" with the periodic pipe tutorial and it worked there, which is adding to my confusion!

Could you kindly point out if something is wrong with my setup?

Kind regards

syavash

_______________________________________________
Nektar-users mailing list
Nektar-users@imperial.ac.uk
https://mailman.ic.ac.uk/mailman/listinfo/nektar-users