installing JuliaFEM correctly?

New to Julia
1

Happy Monday,
I was looking for a less memory intensive alternative (compared to julia fenics) for doing finite element method, so I wanted to try JuliaFEM. As Navier Strokes is not a predefined problem I wanted to define it my self. When adding JuliaFEM it gave me some warning messages. I couldnt paste the testing output as it goes over the caracter limit.
I tried to follow the example of adding a problem but I could not, so I was guessing maybe JuliaFEM is not properly installed? The predefined example problems that come with it though work fine though.

type Truss <: FieldProblem
end
ERROR: syntax: extra token "Truss" after end of expression
function get_unknown_field_name(problem::Problem{Ginger})
    return "displacement"
end
UndefVarError: Ginger not defined

Stacktrace:
 [1] top-level scope at In[17]:1
2

Ping @ahojukka5

We have an active “support” in gitter: JuliaFEM/JuliaFEM.jl - Gitter consider asking in there.

3

Maybe this example could help you:
PDEAssembler.jl

4

I think there’s nothing wrong with the installation, ]add JuliaFEM should be enough.

With new Julia 1.0 syntax, type should be immutable struct and if you define a problem Truss, you should use that name also in that function (now you use Ginger why the error is thrown).

Br,
Jukka

5

Im unable to install juliaFEM, im getting the following error:

(@v1.10) pkg> add JuliaFEM
Resolving package versions…
ERROR: Unsatisfiable requirements detected for package Reexport [189a3867]:
Reexport [189a3867] log:
├─possible versions are: 0.2.0-1.2.2 or uninstalled
├─restricted by compatibility requirements with LinearSolve [7ed4a6bd] to versions: 1.0.0-1.2.2
│ └─LinearSolve [7ed4a6bd] log:
│ ├─possible versions are: 0.1.0-2.28.0 or uninstalled
│ └─restricted to versions * by an explicit requirement, leaving only versions: 0.1.0-2.28.0
└─restricted by compatibility requirements with JuliaFEM [f80590ac] to versions: 0.2.0 — no versions left
└─JuliaFEM [f80590ac] log:
├─possible versions are: 0.5.0-0.5.1 or uninstalled
└─restricted to versions * by an explicit requirement, leaving only versions: 0.5.0-0.5.1

6

I think JuliaFEM is abandoned.

7

Hi @PetrKryslUCSD, I want to solve 3D poissons eqn (electrostatic type problem) in julia, can you suggest good packages that are well documented as well as robust for solving large problems (~10 million DOFs).

8

Look at gridap, I’d say. It can even offer support for parallel execution.

1 Like
9

Ok I will look into it and respond. Thanks

10

If your scope is open source but not Julia, consider Salome-Meca or Code Aster.

1 Like