import from .dat file abaqus

hi all,

I try to import certain lines from my .dat file, generated from abaqus. However something strange happens. When I try to import the certain lines with use of Import Data button, select my lines and import it as a table and write it as a function 'importfile'. Then It's not importing the lines I want, but an arbitrary selection, even though the corresponding rows match. However if I copy and paste data from my abaqus generated .dat file, to an empty .dat file, open that one in matlab and select the lines I want to import, then it dóes import the correct lines.

Anyone a clue what is happening here? I tried almost all features in the Import Data toolbox, such as writing it as a different file (table, cell, etc.) but nothing works so far. The file is pretty large, can there be a problem?

I uploaded the file, I need the following data points, for every increment I need the values on the column below the PU2 and the frequency. Is someone keen to help me out? :)

       NODE FOOT-  U1             U2             PU1            PU2          
            NOTE
  
      3739      9.4520527E-02 -1.9431870E+00  9.5103755E-02  1.9543827E+00 
      3739 SSD  1.0516381E-02 -2.0889238E-01  6.3486346E+00 -1.7386427E+02 


                              INCREMENT NUMBER    31
                              FREQUENCY =   78.976194    

[Mario Malic]

Counter = 1;
FID = fopen('results.txt', 'rt'); 
tline = fgetl(FID);
File_Data{Counter} = tline;
while ischar(tline)
    Counter = Counter+1;
    tline = fgetl(FID);
    File_Data{Counter} = tline;
end
fclose(FID);
File_Data = File_Data';

This will give you your file cell variable File_Data.

You will probably have to find a cell that contains 'INCREMENT NUMBER', example, Increment number is line 8, then you take 8-3, 8-4, 8 and 8+1 line for a single increment number.

% 1       NODE FOOT-  U1             U2             PU1            PU2          
% 2            NOTE
% 3
% 4      3739      9.4520527E-02 -1.9431870E+00  9.5103755E-02  1.9543827E+00 
% 5      3739 SSD  1.0516381E-02 -2.0889238E-01  6.3486346E+00 -1.7386427E+02 
% 6 
% 7      
% 8                              INCREMENT NUMBER    31
% 9                              FREQUENCY =   78.976194   

I would not know how this code below would work, but you can use some ideas from it.

Wanted_Rows = 1:1:length(File_Data);
Delete_Lines = [1:13 114 115 216 217, length(File_Data)]; % Last line of File_Data is -1 which denotes end of file so we remove it here
File_Data(Delete_Lines) = [];
File_Data_Split = split(File_Data(1:length(File_Data))); % Splits the cell
Data_Ready = cellfun(@str2num,File_Data_Split);

[KSSV]

This might help: https://in.mathworks.com/matlabcentral/answers/307258-extract-nodes-and-elements-from-abaqus-input-file-to-matlab

[Frank Oosterveld]

@Mario & @KSSV, great posts, thanks, the first bit helps me a lot. Now it imports the .dat file in one cell element. Also, when I now call for the lines I want (in the first place for one increment) I get the following:

File_Data =

  3×1 cell array

    {'      3739      6.9484012E-02 -1.4426835E+00  6.9688737E-02  1.4466171E+00 '}
    {'      3739 SSD  5.3377998E-03 -1.0660863E-01  4.3928653E+00 -1.7577375E+02 '}
    {'                              FREQUENCY =   79.918425    '                  }

This is nice!!

However, when I implement the ideas of your second part of the script, it's not yet working. Did some small research, providing the same procedure of cellfun(@str2num,Data, 'uniformoutput', false). With minor edits, such as including 'uniformoutput', false, I do get a cell output, but only with double numbers of the first row.

    Data_Ready =

  3×1 cell array

    {1×5 double}
    {0×0 double}
    {0×0 double}

this can be converted with cell2mat to numeric values, but that's only the first row..

So my question now is how do I convert this cell structure into (at least) the numeric values I'm interested in (values below PU2 and the frequency)?

[Frank Oosterveld]

Found it! forgot the split function, with use of the following extra scripting, it works!! Thanks a lot for your time and attention :)

Wanted_row = [45994 45995 45999]
File_Data_1 = File_Data(Wanted_row)

row1 = split(File_Data_1{1})
row2 = split(File_Data_1{2})
frequency = split(File_Data_1{3})

cell2mat(cellfun(@str2num,row1,'uniformoutput',false))
cell2mat(cellfun(@str2num,row2,'uniformoutput',false))
cell2mat(cellfun(@str2num,frequency,'uniformoutput',false))

[Mario Malic]

Great to hear it works, KSSV's solution might be more appropriate when you would like to generalise the information extraction, without writing down the wanted rows like in that example.