lp:~elmer-csc-ubuntu/elmercsc/+git/elmer-github-devel

Author: Mika Malinen
Author Date: 2024-07-08 10:39:14 UTC

Add a label to test complex-valued eigen solution

+ misc. typo corrections

Author: RupertGladstone
Author Date: 2024-07-05 12:26:11 UTC

Merge branch 'elmerice' into glads_gl

Author: fgillet
Author Date: 2024-07-04 07:07:16 UTC

Update documentation

Author: Thomas Zwinger
Author Date: 2024-07-03 13:17:25 UTC

Types for rocALUTION

Author: Matias Zwinger
Author Date: 2024-06-20 14:56:37 UTC

Add nix flake

Author: Your Name
Author Date: 2024-06-07 15:59:34 UTC

Expand the solver/preconditioner choises somewhat.

Author: Eelis Takala
Author Date: 2024-05-06 16:51:58 UTC

Fix Homogenization 3D

* Don't require homogenization for all coils
* Add missing nu component (homog)

  By convention, we are using the 2D approximation in 2D
  stranded homogenization parameters: in local coordinates they are
  xx and yy. But, the third one (zz) has been missing, causing some
  strange tensors as output. In this commit the third is added. Note that
  now the third has to be declared in the sif:

  ```sif
  Component: Nu 33
  Component: Nu 33 im
  ```

  Had to touch these:
  `fem/tests/circuits_harmonic_homogenization_coil_solver/sif/2241.sif`
  `fem/tests/circuits_harmonic_stranded_homogenization/sif/2241.sif`

TODO:
  * we need to hack coil solver such that we don't have the problem
  shown in: https://github.com/ElmerCSC/elmerfem/issues/460#issuecomment-2059231516
  This should be fixed soon! @raback

Author: RupertGladstone
Author Date: 2024-04-06 16:29:54 UTC

Merge branch 'elmerice' into elmerice_FISOC

Author: Eelis Takala
Author Date: 2024-01-26 15:00:34 UTC

Resolve y-axis and w based local coordinate system

The main application for the y-axis and w based local coordinate system
is the stranded homogenization model where the homogenization parameters
are oriented using the RotM transformation. The RotM transformation
needs a local coordinate system and the wish is to be able to define it
without direction solvers (in case of arbitrary wire cross-sections).

A test case derived from `circuits_harmonic_stranded_homogenization` is
added. The original case computes the skin and proximity losses for
a stranded coilcurrent via so called homogenization, where the strands
are not geometrically in the model but their frequency responce is
estimated using `sigma` and `nu` parameters that are gotten from the
elementary solutions of 2D computations (transversal fields and
perpendicular current). In 3D the cross-section needs to be oriented
using the RotM transformation that requires the definition of a local
coordinate system. For that we have previously used the direction
solvers (`Alpha` and `Beta`).

The main point now is to get rid of the direction solvers.
The other major change is to use `CoilSolver` solution to drive
the circuit (instead of `WSolver`). There are many benefits for
using the `CoilSolver`, for example, one can calculate closed coils.

The important new keywords (with respect to ) here are:
 1. `Body: Local Coordinate System Beta Reference and Gamma = Logical True`
   * CoordinateTransform computes alpha = beta x gamma:
     where beta is set to "Beta Reference" and
     Gamma is set to "coilcurrent e" computed by `CoilSolver`
 2. `Component: Coil Normal(3) = Real 0. 0. 1.`
   * `CoilSolver` needs to know how the coil is oriented
 3. `Component: Desired Current Density = Real 1.0`
   * `CoilSolver` will set a certain current density over the coil
 4. `Component: Coil Use W Vector = Logical True`
   * Coil uses a defined `W Vector` for which a name can be provided
   * The `W Vector` set here is used for driving the component with circuits
 5. `Component: W Vector Variable Name = String "CoilCurrent e"`
   * Here we choose the correct field for `W Vector`
 6. `Boundary: Coil Start = Logical True`
   * `CoilSolver` needs a boundary for knowing where coil starts (if not closed)
 7. `Boundary: Coil End = Logical True`
   * `CoilSolver` needs a boundary for knowing where coil ends (if not closed)

Issue #429

Author: Juhani Kataja
Author Date: 2024-02-27 12:05:43 UTC

matc OpenMP fix: main.c

Don't try to set number of threads without omp.h header.

Author: Juhani Kataja
Author Date: 2024-01-29 14:49:15 UTC

Add module: Probes.F90

Contains solver module that integrates variables
over bodies.

Author: Foad
Author Date: 2023-12-12 23:30:29 UTC

Resolved build issues in MSYS environment: Addressed various compilation and linking errors in ElmerFEM project. Modified CMakeLists to conditionally include POSIX dynamic linking functions and Windows equivalents, ensuring compatibility across different environments. Adjusted linker flags to exclude -ldl in MinGW, and set linker path explicitly for MSYS. Also implemented namespace changes to avoid std::min and std::max conflicts. These updates enhance cross-platform build stability, particularly for Windows and Unix-like systems within MSYS.

Author: Eelis Takala
Author Date: 2023-12-07 05:32:41 UTC

Implement post processing for homogenization

Here we implement post processing for homogenization model.
New field, called the "proximity loss", is created and it
contains the proximity losses caused by the homogenization
parameter `nu`. Skin losses are added to the existing
joule losses.

Author: Peter Raback
Author Date: 2023-12-01 15:34:37 UTC

baby steps

Author: Juha Ruokolainen
Author Date: 2023-11-21 13:03:13 UTC

small fix.

Author: Peter Raback
Author Date: 2023-11-02 15:08:06 UTC

First version of patch fitting that does something.

Author: Peter Raback
Author Date: 2023-09-29 21:28:55 UTC

Minor fix

Author: Thomas Zwinger
Author Date: 2023-09-04 11:42:31 UTC

merged devel and elminated conflicts

Author: Thomas Zwinger
Author Date: 2023-08-17 13:05:49 UTC

removed confusing 3rd body in ExtrudedMesh8-test

Author: RupertGladstone
Author Date: 2023-05-26 09:01:09 UTC

Merge branch 'elmerice' into glads_revisit

Author: Eelis Takala
Author Date: 2023-03-30 13:07:39 UTC

Add steady case for massive

Author: Peter Raback
Author Date: 2023-03-02 22:10:44 UTC

Update ListFindPrefix too

Author: Eelis Takala
Author Date: 2022-12-09 20:28:33 UTC

Implement Coil detection to SkinBC

Here we find a problem with the element body id.
Problem with test 'fem/tests/mgdyn_harmonic_wire_impedanceBC_circuit':
For some reason GetBody(Element) equals 0 each time
`LocalMatrixSkinBC` function is called. This should be called only when
`Body Id` is set to 3 because in
'fem/tests/mgdyn_harmonic_wire_impedanceBC_circuit':
```
  Boundary Condition 2
    !
    ! Impedance BC
    !
    Name = "WireSurface"
    Target Boundaries(1) = 2
    Layer Electric Conductivity = Real 58e6
    Layer Relative Permeability = Real 1
    Body Id = 3
  End
```

Try to associate component to a boundary

Here we try a method for associating a boundary with a
component. This could then be treated like the component bodies
in order to create the Impedance BC.

However, For some reason, CurrentModel % Boundaries is of zero length
at CircuitUtils.F90:205
```fortran
print *, "SIZE(CurrentModel % Boundaries)", SIZE(CurrentModel % Boundaries)
```

Why?

Checkout

Author: RupertGladstone
Author Date: 2022-10-11 15:40:10 UTC

file naming issue

Author: Peter Raback
Author Date: 2022-05-23 10:46:57 UTC

Merge branch 'devel' into TetraSplit

Author: Mika Malinen
Author Date: 2022-05-11 07:58:59 UTC

Fix matrix updates

Author: RupertGladstone
Author Date: 2022-03-09 13:24:45 UTC

typo re mesh

Author: Peter Raback
Author Date: 2022-01-03 14:01:05 UTC

baby steps

Author: Eelis Takala
Author Date: 2021-05-27 12:05:39 UTC

Fix conflicts

Author: Thomas Zwinger
Author Date: 2020-11-10 19:52:44 UTC

Merge branch 'elmerice' into devel

Author: J Todd
Author Date: 2020-09-22 13:53:47 UTC

Merge branch 'elmerice' into elmerice_meshadapt

Author: Peter Raback
Author Date: 2020-07-07 23:27:17 UTC

small fix

Author: Peter Raback
Author Date: 2020-06-26 14:08:55 UTC

Merge branch 'devel' into TurbulenceVec

Author: Peter Raback
Author Date: 2020-06-02 06:27:02 UTC

Squashed commit of parallel FCT steps. Still a little dubious...

Author: Foad
Author Date: 2020-04-13 21:38:17 UTC

using new eigen findmetis

Author: Foad
Author Date: 2020-04-08 16:53:32 UTC

compiles on macOS

Author: Peter Raback
Author Date: 2020-01-17 08:33:22 UTC

Preliminary solution for the n-t challenge in contact mechanics. Do not merge as such.

Author: Juhani Kataja
Author Date: 2017-12-14 13:59:43 UTC

ci: removed a typo

Author: Josefin
Author Date: 2017-04-10 12:13:51 UTC

cleaned iscal test-sif

Author: Juhani Kataja
Author Date: 2017-02-01 13:34:41 UTC

Initial work on including SuperLU in cmake system

Untested: SolveSuperLU assumes some other superlu version

Author: Sami Ilvonen
Author Date: 2015-02-09 10:10:23 UTC

Merge branch 'svnsync'