Cantilever Mesh Bone Remodelling

[laurenboag1]

Hello,

I am trying to model the benchmark cantilever problem from the waffenschmidt paper.

Currently I have fixed all dofs at one end and applied a downwards force on the surface at the opposite end. I believe I need to create a smaller surface to apply the downwards force rather than apply it over the end end of the cantilever but I am unsure how to do this on cubit.

Any advice would be appreciated. I have attached my cubit mesh and journal file for reference.

Thanks,

Lauren

[Karol Lewandowski]

You need to create a small cube, move it to the face you want to apply load on and the use the option imprint.

[laurenboag1]

How do I move the cube to the correct position?

[Karol Lewandowski]

Use the Move tool.

[laurenboag1]

I can't find a move tool. I also don't know how to select the smaller cube as it is inside the larger body.

[laurenboag1]

I have managed to move the block by typing a journal command.

[laurenboag1]

Hi,

Could someone please verify this model is correct before I begin the MoFEM analysis?

I am unsure if my units are correct/consistent. The paper also applies a traction of 1.6 N/mm2 where as I have applied a force 1.6 N.

Thank you.

[Karol Lewandowski]

I assume that you have in mind Waffenschmidt 2012 paper (Application of an anisotropic growth and remodelling formulation to computational structural design).

I would modify the mesh file like this:

reset

brick x 1600 y 1000 z 300

brick x 100. y 100. z 100.

vol 2 move -795 0 0

imprint volume 1 with volume 2

delete volume 2

#BOUNDARY CONDITIONS

create displacement on surface 6 dof all fix

create force on surface 14 force value 1.6 direction ny

#MESHING

volume all scheme Tetmesh

volume all size auto factor 5

mesh volume 1

And then the command line:

../tools/mofem_part -my_file cantilver_3D.cub -my_nparts 1 && mpirun -np 1  ./bone_adaptation -my_file out.h5m  -my_order 2 -my_output_prt 1 -ksp_type fgmres -pc_type lu -pc_factor_mat_solver_package mumps -ksp_atol 1e-12 -ksp_rtol 1e-12 -snes_monitor -snes_type newtonls -snes_linesearch_type basic -snes_max_it 1000 -snes_atol 1e-6 -snes_rtol 1e-8 -ts_type beuler -ts_max_snes_failures 1 -ts_monitor  -ts_dt 0.1 -ts_final_time 50 -my_load_history load_history_cant.in -mass_postproc -ts_monitor -ksp_monitor -young_modulus 3800 -poisson_ratio 0.3 -c 0.4 -rho_ref 1e-0 -psi_ref 0.01 -m 3 -n 2 | tee log

Although, I am not sure about the density unit, I think they used inconsistent one in the article. You need play with that. 

[laurenboag1]

Hi Karol,

Why is the young's modulus 3800 MPa?, I could not find a reference for the value of bone used in the paper.

Kind regards,

Lauren

[Karol Lewandowski]

In the paper they gave lambda and mu coefficients (Lame constants). So you need to use the following  table for conversion:

 

[laurenboag1]

When running the analysis for concrete parameters, should the loading be applied with the same stepwise function as before or just with a constant force, which I have selected as 100MN ?

Also is the poissons ratio determined from the above conversion table, as only Young's modulus is known for concrete?

Kind regards,

Lauren 

[Lukasz Kaczmraczyk]

Check literature, what is concrete Poisson ratio.