How to creat a data file of abaqus which can be readed by Matlab

[lanfeng]

Hi,
I am doing optimization with abaqus and matlab. And I have force and displacement data, both experimental and numerical data which come from abaqus simulation, compare the experimental and numerical data to get optimal parameter with matlab optimization tool. But now I have a problem how to output the force-displacement of a node set in abaqus to a file which can be readed by matlab?
Can anyone give me some help?
Thank you for your attention!

[lanfeng]

Hi,all
Thank you for your attention. Under the help of biofriend, I used the odbreport command in abaqus to report the displacement-force data into a .rep file. As the following:
-----------------------------------------------------------
Step name 'Load'
description :
number : 1
domain : TIME
procedure : *STATIC
timePeriod : 1.0
totalTime : 0.0
previousStep : Initial
nlgeom : Yes

Number of history regions = 1

History Region 'Node CONICAL-1.1'
description : Output at instance CONICAL-1 node 1
History Point
position : NODAL
instance : CONICAL-1
node : 1

History Output 'RF2'
description : Reaction force
type : SCALAR
Frame value Data
0.0 0.0
0.001 -0.000589552
0.002 -0.00117862
0.0035 -0.0020613
0.00575 -0.00338319
0.009125 -0.00536105
0.0141875 -0.00831677
0.0217813 -0.0127225
0.0331719 -0.0192631
0.0502578 -0.0289014
0.0758867 -0.110891
0.11433 -0.271967
0.16433 -0.515706
0.21433 -0.928407
0.26433 -1.36514
0.31433 -1.99691
0.36433 -2.62109
0.41433 -3.45311
0.46433 -4.30737
0.51433 -5.3008
0.56433 -6.39789
0.61433 -7.51737
0.66433 -8.86923
0.71433 -10.2136
0.76433 -11.7104
0.81433 -13.3232
0.86433 -14.928
0.91433 -16.7865
0.96433 -18.6739
1.0 -20.015

History Output 'U2'
description : Spatial displacement
type : SCALAR
Frame value Data
0.0 0.0
0.001 -0.005
0.002 -0.01
0.0035 -0.0175
0.00575 -0.02875
0.009125 -0.045625
0.0141875 -0.0709375
0.0217813 -0.108906
0.0331719 -0.165859
0.0502578 -0.251289
0.0758867 -0.379434
0.11433 -0.57165
0.16433 -0.82165
0.21433 -1.07165
0.26433 -1.32165
0.31433 -1.57165
0.36433 -1.82165
0.41433 -2.07165
0.46433 -2.32165
0.51433 -2.57165
0.56433 -2.82165
0.61433 -3.07165
0.66433 -3.32165
0.71433 -3.57165
0.76433 -3.82165
0.81433 -4.07165
0.86433 -4.32165
0.91433 -4.57165
0.96433 -4.82165
1.0 -5.0

End of ODB Report.

Which I need to load into matlab is only the numerical data in red. I am a new user of Matlab.
Are there anyone can help me how can I write a matlab file to load the numerical data into matlab?

[creanea]

Yeap, its pretty straight forward. If you're new to matlab, its probably easiest to just use the import wizard.

Change the file extension from .rep to .txt. Then open matlab and go to File - Import wizard and open the file. You can then create arrays/matrices from your data.

Once you're more accustomed to matlab you can write an mfile that will do it automatically using a function like textread. There are several others either.

[lanfeng]

Yeap, its pretty straight forward. If you're new to matlab, its probably easiest to just use the import wizard.

Change the file extension from .rep to .txt. Then open matlab and go to File - Import wizard and open the file. You can then create arrays/matrices from your data.

Once you're more accustomed to matlab you can write an mfile that will do it automatically using a function like textread. There are several others either.

Hi,creanea,
thank you for the answer.
Can you tell me more details:
1. How can I change .rep to .txt
2. How can I creat arrays/matrices? Is the created arrays/matrices is the data I need?
3. Can you give me a example matlab file?
Thank you

[creanea]

1. Right click and rename it with .txt at the end. It mightn't even be necessary.

2. Open the import wirard, click on the file. Your data will automatically but into arrays. You can select any data in the file, similar to excel.

3. Loads of examples in the Matlab help.

[lanfeng]

1. Right click and rename it with .txt at the end. It mightn't even be necessary.

2. Open the import wirard, click on the file. Your data will automatically but into arrays. You can select any data in the file, similar to excel.

3. Loads of examples in the Matlab help.

Hi,creanea,
thank you for your quickly help. It really gives me inspiration.
But I still have problems:
1.I want to ask Matlab change .rep to .txt automatically, because I will do many times calculation, I can not do it by hand every time. So I look for a matlab command to change .rep to .txt
2. The same problem like 1. I want to write a matlab procedure to make matlab select the data(red ones) automatically, to fill this data into a new file as a 2*n matrics.
3. In my .rep there a lot other characters, and that I feel it is difficult to deal with. And the examples in matlab help I see only for simple number data or with a few headers. I don't know if there is example file similer to my problem.

[creanea]

You should see look for the functions textscan and str2double in the matlab help then, although there may be better ones I'm not sure.

I've attached an mfile I used to read from .odb files and write to .inp files. I can't remember exactly how it works but it shouldn't take long to understand and may give you ideas on where to start for on your own code.

You will need to search for the first and last line of the data you want to extract - this is done using the nested loops starting on line 45. Maybe ignore the first half of the mfile as it involves using a python script to take stress results from abaqus.

Forget about changing the file type, its not needed, textscan will read any type of text file.

[lanfeng]

Thank creanea's useful advise. Now I read the data to a 94*1 cell array.
But I still have a problem:
How can I write the red and blue numeric data into a txt file with two column respectively?
Now I can only write all of the data into one column.
'Report:'
'ODB'
':'
':'
':'
':'
''
'number'
''
'name'
':'
':'
':'
':'
':'
':'
':'
':'
'of'
'Region'
':'
'Point'
':'
':'
':'
'Output'
':'
':'
'value'
'0.0'
'-0.005'
'-0.01'
'-0.0175'
'-0.02875'
'-0.045625'
'-0.0709375'
'-0.108906'
'-0.165859'
'-0.251289'
'-0.379434'
'-0.57165'
'-0.82165'
'-1.07165'
'-1.32165'
'-1.57165'
'-1.82165'
'-2.07165'
'-2.32165'
'-2.57165'
'-2.82165'
'-3.07165'
'-3.32165'
'-3.57165'
'-3.82165'
'-4.07165'
'-4.32165'
'-4.57165'
'-4.82165'
'-5.0'
'Output'
':'
':'
'value'
'0.0'
'-0.000589552'
'-0.00117862'
'-0.0020613'
'-0.00338319'
'-0.00536105'
'-0.00831677'
'-0.0127225'
'-0.0192631'
'-0.0289014'
'-0.110891'
'-0.271967'
'-0.515706'
'-0.928407'
'-1.36514'
'-1.99691'
'-2.62109'
'-3.45311'
'-4.30737'
'-5.3008'
'-6.39789'
'-7.51737'
'-8.86923'
'-10.2136'
'-11.7104'
'-13.3232'
'-14.928'
'-16.7865'
'-18.6739'
'-20.015'
'of'

[creanea]

The unloaded file will read the first set of data you posted and put it in a 2 by n matrix. You'll need to adjust it for the next one. I think you can simply replace 'data' and 'history' with 'value' and 'output'.

There's definitely better and more efficient ways to do this but it will work.

Arthur

[lanfeng]

Hi,creanea,
thank you so much.
My problem solved with the help from you.
chen