6
$\begingroup$

I am trying to compute the eigenfunctions of an oblate spheroid (a=75 cm and b=60 cm) using Mathematica's FEM package and Chris' answer from here. Specifically, I am looking for eigenfrequencies around 433, 893, 913 and 2400 MGHz. Is there any way I could narrow my search besides getting all eigenfrequencies initially and then looking for the desired outcome which is impractical?

Here is my code for the first 4 eigenmodes:

Needs["NDSolve`FEM`"];

helmholzSolve3D[g_, numEigenToCompute_Integer, 
opts : OptionsPattern[]] := 
Module[{u, x, y, z, t, pde, dirichletCondition, mesh, boundaryMesh, 
nr, state, femdata, initBCs, methodData, initCoeffs, vd, sd, 
discretePDE, discreteBCs, load, stiffness, damping, pos, nDiri, 
numEigen, res, eigenValues, eigenVectors, 
evIF},

(*Discretize the region*)

If[Head[g] === ImplicitRegion || Head[g] === ParametricRegion, 
mesh = ToElementMesh[DiscretizeRegion[g, opts], opts], 
mesh = ToElementMesh[DiscretizeGraphics[g, opts], opts]];
boundaryMesh = ToBoundaryMesh[mesh];

(*Set up the PDE and boundary condition*)

pde = D[u[t, x, y, z], t] - Laplacian[u[t, x, y, z], {x, y, z}] + 
u[t, x, y, z] == 0;
dirichletCondition = DirichletCondition[u[t, x, y, z] == 0, True];
(*Pre-process the equations to obtain the FiniteElementData in \
StateData*)nr = ToNumericalRegion[mesh];
{state} = 
NDSolve`ProcessEquations[{pde, dirichletCondition, 
u[0, x, y, z] == 0}, u, {t, 0, 1}, Element[{x, y, z}, nr]];
femdata = state["FiniteElementData"];
initBCs = femdata["BoundaryConditionData"];
methodData = femdata["FEMMethodData"];
initCoeffs = femdata["PDECoefficientData"];

(*Set up the solution*)vd = methodData["VariableData"];

sd = NDSolve`SolutionData[{"Space" -> nr, "Time" -> 0.}];

(*Discretize the PDE and boundary conditions*)

discretePDE = DiscretizePDE[initCoeffs, methodData, sd];
discreteBCs = DiscretizeBoundaryConditions[initBCs, methodData, sd];

(*Extract the relevant matrices and deploy the boundary conditions*)

load = discretePDE["LoadVector"];
stiffness = discretePDE["StiffnessMatrix"];
damping = discretePDE["DampingMatrix"];
DeployBoundaryConditions[{load, stiffness, damping}, discreteBCs];

(*Set the number of eigenvalues ignoring the Dirichlet positions*)

pos = discreteBCs["DirichletMatrix"]["NonzeroPositions"][[All, 2]];
nDiri = Length[pos];
numEigen = numEigenToCompute + nDiri;

(*Solve the eigensystem*)

res = Eigensystem[{stiffness, damping}, -numEigen];
res = Reverse /@ res;
eigenValues = res[[1, nDiri + 1 ;; Abs[numEigen]]];
eigenVectors = res[[2, nDiri + 1 ;; Abs[numEigen]]];
evIF = ElementMeshInterpolation[{mesh}, #] & /@ eigenVectors;

(*Return the relevant information*)

{eigenValues, evIF, mesh}]

{ev, if, mesh} = 
helmholzSolve3D[Ellipsoid[{0, 0, 0}, {0.75, 0.6, 0.6}], 4, 
MaxCellMeasure -> 0.025]

Table[
DensityPlot[
if[[i]][x, y, 0.1], {x, -1, 1}, {y, -1, 1},
RegionFunction -> Function[{x, y}, x^2/0.75^2 + y^2/0.6^2 < 1],
PlotLabel -> ev[i] ,
ColorFunction -> Hue,
PlotLegends -> Automatic
],
{i, 1, 4}
]

Any suggestions?

|improve this question
$\endgroup$
9
$\begingroup$

You could use something like this:

{vals, funs} = 
 NDEigensystem[{-Laplacian[u[x, y, z], {x, y, z}] + u[x, y, z], 
   DirichletCondition[u[x, y, z] == 0, True]}, u, 
  Element[{x, y, z}, Ellipsoid[{0, 0, 0}, {0.75, 0.6, 0.6}]], 4, 
  Method -> {"Eigensystem" -> {"FEAST", "Interval" -> {425, 500}}}]

{{427.961, 428.783, 430.026, 430.156},...}

And here are the density plots:

Table[DensityPlot[funs[[i]][x, y, 0.1], {x, -1, 1}, {y, -1, 1}, 
  RegionFunction -> Function[{x, y}, x^2/0.75^2 + y^2/0.6^2 < 1], 
  PlotLabel -> vals[[i]], ColorFunction -> Hue, 
  PlotLegends -> Automatic, PlotRange -> All], {i, 1, 4}]

enter image description here

Slice density plots:

Table[SliceDensityPlot3D[funs[[i]][x, y, z], 
  Element[ {x, y, z}, Ellipsoid[{0, 0, 0}, {0.75, 0.6, 0.6}]], 
  PlotRange -> All, PlotLabel -> vals[[i]], 
  PlotTheme -> "Minimal"], {i, Length[vals]}]

enter image description here

And density plots:

Table[DensityPlot3D[funs[[i]][x, y, z], 
  Element[ {x, y, z}, Ellipsoid[{0, 0, 0}, {0.75, 0.6, 0.6}]], 
  PlotRange -> All, PlotLabel -> vals[[i]], 
  PlotTheme -> "Minimal"], {i, Length[vals]}]

enter image description here

|improve this answer
$\endgroup$
  • $\begingroup$ Thank you for your answer but I need to clarify something technical here. Does NDEigensystems compute eigenmodes from start, ie 0 and then narrows its search to the desired interval (425, 500 HZ here) or does it start from 425 Hz and then stops at 500 Hz? $\endgroup$ – anon Mar 28 '19 at 19:26
  • $\begingroup$ @GeorgeGiannoulis, I think the latter, but you could have a look at the FEAST algorithm.Thought that version is not the same as the one linked in Mathematica but that shlould not matter. NDEigensystem makes use if Eigensystem (like in your code) which then uses FEAST from a library. $\endgroup$ – user21 Mar 29 '19 at 5:36
  • $\begingroup$ OK one last thing here. I can't seem to understand what the boubdary is in your code. Is it a cube,a sphere, an ellispoid? Something else? $\endgroup$ – anon Mar 29 '19 at 10:16
  • $\begingroup$ @GeorgeGiannoulis, it's the ellipsoidI have updated the code. $\endgroup$ – user21 Mar 29 '19 at 10:23
  • $\begingroup$ Great! I d like to add some density plots though for the eigenvalues. My code looks something like this: $\endgroup$ – anon Mar 29 '19 at 10:58
6
$\begingroup$

You may try Eigensystem with 

Method -> {"FEAST", "Interval" -> {a, b}}

to search eigenvalue pairs within an interval. See the documentation of Eigensystem, Section "Methods", Subsection "FEAST" for more details.

|improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy