Microstructure from file

technical aspects of .dri file generation (e.g. debug mode ) etc...
Bernd
Posts: 1504
Joined: Mon Jun 23, 2008 9:29 pm

Re: Microstructure from file

Post by Bernd » Wed Oct 21, 2015 7:04 pm

Dear Raina,

I think the problem is that you can't append if the files are not existing yet. That means, if you chose a new output file name in case of restart, you must use "overwrite".

The output about compression is default output which is not related to your choice.

Bernd

Raina
Posts: 13
Joined: Thu May 21, 2015 2:03 am
anti_bot: 333

Re: Microstructure from file

Post by Raina » Wed Oct 21, 2015 10:35 pm

Hello Bernd,

Thanks for your professional advice, but still It doesn't work. :cry:
I don't think ''overwrite'' is a good choice, since it will ask me to type in Phase Data later, and I don't think I'm gonna use that as my initial structure. Therefore, I prefer using ''restart''. However, when I type in the same output name as my .rest file, still the simulation ended.
4.png
4.png (55.63 KiB) Viewed 34749 times
Apart from that, I also would like to view some 3D simulation results. The Micress manuals I downloaded from this website recommend using Paraview, but when I use this software, I found that it only enable me to view the microstructure on the surface of the box. Is there any possibility that I can view the inner microstructure ? Or this usually seldom happens in 3D simulations.

Also, I would like to make a small animation for 3D simulation results, but the information given by the manual is too short. Do you have any recommended resources about making animation in Paraview ? Or do you have some more convenient software to recommend ?

Thanks again for your patience.

Raina

Bernd
Posts: 1504
Joined: Mon Jun 23, 2008 9:29 pm

Re: Microstructure from file

Post by Bernd » Wed Oct 21, 2015 11:22 pm

Dear Raina,

whether you use "append" or "overwrite" has nothing to do with the input which is required for the simulation. The difference is just, in case there is already result data there with the same name, whether these files will be overwritten, or whether the new results will be appended to the existing files. But if you chose "append" and there are no results with the same name, MICRESS complains and stops because it cannot open these files for appending.
In all cases you have to type in phase data, why is this a problem? Phase data are not stored in the restart file, they are necessary for opening the .rest file. Reading from a restart file does not replace any input, although some information like positions of initial grains will be ignored.

With Paraview it is also possible to view the inside of a simulation domain. You can use for example volume representation with transparency, or create contours (iso-surfaces). Paraview is a quite powerful tool, therefore we currently do not use other software. Of course you can also create animations once you have build up a suitable view. There is even an interface to Python so that everything can be scripted. But Paraview is not our product, so please refer to their documentation.

Bernd

Raina
Posts: 13
Joined: Thu May 21, 2015 2:03 am
anti_bot: 333

Re: Microstructure from file

Post by Raina » Thu Oct 22, 2015 6:35 pm

Hello Bernd,

As you told me, I used ''restart reset_time" and "overwrite", later it asked me to type in Grain Input. If I continue typing in Grain Input as shown below, what is Micress going to use as initial structure? Read from .rest file or read from Grain Input I typed in?

#
# Grain input
# ===========
# Type of grain positioning?
# Options: deterministic random [deterministic_infile] from_file
random
# Integer for randomization?
125
# Number of different types of grains?
1
# Number of grains of type 1?
1800
# Input for grain type 1
# ----------------------
# Geometry of grain type 1
# Options: round rectangular elliptic
round
# Minimal value of x-coordinates? [micrometers]
0.00000
# Maximal value of x-coordinates? [micrometers]
800.000
# Minimal value of z-coordinates? [micrometers]
0.00000
# Maximal value of z-coordinates? [micrometers]
800.000
# Minimum grain radius? [micrometers]
20.0000
# Maximum grain radius? [micrometers]
20.0000
# Shall grain type 1 be stabilized or shall
# an analytical curvature description be applied?
# Options: stabilisation analytical_curvature
stabilisation
# Should the Voronoi criterion for grains of type 1 be applied?
# Options: voronoi no_voronoi
voronoi
# Phase number for grain type 1? (int)
1
# Determination of grain orientations?
# Options: random fix fix_direction
random
# Minimal distance between grains (real) [micrometers]?
15.000
#

Thanks,
Raina

Bernd
Posts: 1504
Joined: Mon Jun 23, 2008 9:29 pm

Re: Microstructure from file

Post by Bernd » Thu Oct 29, 2015 3:58 pm

Dear Raina,

sorry, I somehow didn't see your question :oops:

Restart overwrites all additional grain input, so user input will be ignored.

There is only one exception: "restart structure_only". This option allows your to overwrite only a specific grain (or several specific grains) with the microstructure which is stored in the .rest file.

Bernd

Raina
Posts: 13
Joined: Thu May 21, 2015 2:03 am
anti_bot: 333

Re: Microstructure from file

Post by Raina » Sat Nov 07, 2015 7:07 pm

Thanks Bernd! Your last reply really answered my question.
Since I'm not very familiar with "restart", I would like to ask more about this.

As you said, when using "restart", all additional grain input will be ignored. However, I'm wondering if I set new phase interaction data, will that be ignored too?

I've tried using "restart" to change the grain boundary mobility in the later simulation with the same restart file. However, later I found even though the threshold angle and mobility ratio are changed, all the out put files looks exact the same. I'm wondering whether I use "restart" in a correct way?

Thanks,
Raina

Bernd
Posts: 1504
Joined: Mon Jun 23, 2008 9:29 pm

Re: Microstructure from file

Post by Bernd » Mon Nov 09, 2015 8:58 pm

Dear Raina,

It is a typical application of the restart functionality to change the interface mobility after restart. This is possible and allowed, because the interface mobility is even not stored in the restart file, i.e. it cannot be overwritten. As a rule of thumb, all input parameters which cannot change during simulation or which change in a predefined way are not stored in the restart file, because they are not needed. They typically can be easily change when using restart. These are typically interface properties, diffusion coefficient values, output times and choices, relinearisation parameters, and even the grid spacing (at your own risk...).
But parameters which define the data structure of other variables which change during simulation, cannot be changed (e.g. the number and type of diffusion term used or the number of phases and elements). They are needed to correctly interpret the data stored in the .rest file.

If you say that changing the interface mobility with restart does not have any effect, it means that the interface mobility under these circumstances has not any effect. If this is true even on a numerical level (bit-identical results), then I can imagine only few reasons:
One is that the mobility is cut down anyway, e.g. if you specify a minimum value for the automatic time-step which is so big that for all interface cells the mobility is reduced to a value which is independent of the nominal mobility value.
Another reason could be that the driving-force is exactly 0. This could be caused by dGMax=0 (specified by the user in the dG options of the phase interaction data) or in case dS of the phase diagram is 0.
I would advise you to check the .mueS output to see whether how used mobiliy values depend on changes of the nominal mobilty.

Bernd

Raina
Posts: 13
Joined: Thu May 21, 2015 2:03 am
anti_bot: 333

Re: Microstructure from file

Post by Raina » Fri Nov 20, 2015 9:09 pm

Hello Bernd,

Could you tell me the format of the input file that can be read by Micress?
I would like to make some modification and then use "from_file", but I don't know which type of file can be read by Mrcress.

Thanks,
Raina

Bernd
Posts: 1504
Joined: Mon Jun 23, 2008 9:29 pm

Re: Microstructure from file

Post by Bernd » Sun Nov 22, 2015 11:54 pm

Hi Raina,

In general, MICRESS reads ASCII or vtk files. In which context do you want to read input data: Is it about microstructure information, concentration fields, or time or temperature dependent parameters like mobility or diffusion coefficients?

Bernd

Raina
Posts: 13
Joined: Thu May 21, 2015 2:03 am
anti_bot: 333

Re: Microstructure from file

Post by Raina » Thu Dec 10, 2015 1:30 am

Hello Bernd,

I think it's about Microstructure file. I tried using .phas file, but it failed.

Thanks,
Raina

Post Reply