Altair RISE

A program to recognize and reward our most engaged community members

Nominate Yourself Now!

.mat file in POSTFEKO

Running POSTFEKO v2017.1.1-303889 build 2017-05-17 on a 64-bit Windows machine.

Is there a way to import .mat data into Postfeko?

Release notes for 300391 state 'Improved options are available to read data from *.mat file and to manipulate the imported data.' but I don't see these options. I would like to have access to the complete Y-matrix (or Z-matrix) instead of just the single column/row that is available in the .out file.

Find more posts tagged with

English

Feko

Accepted answers

All comments

JIF

Hello EMCanuck1,

That reference to a *.mat file is about Matlab / Octave *.mat files. It does not sound like it is the *.mat file you are referring to. I think you want the *.mat file that the FEKO solver export (MoM impedance matrix), correct?

Or are you talking about something else?

Altair Forum User

Yes, I want the .mat file that the FEKO solver exports. I tried using the mat2ascii utility on my .mat files (.mat.0, .mat.1, .mat.2 and .mat.3) but the utility just returns an empty matrix (no data, and it says that the size of the matrix is 4 x -1). My simulation is just a center-fed dipole. I get a good .out file, but I am really having trouble accessing the .mat file.

Thanks

JIF

In that case, you won't be able to import the file in POSTFEKO - that is not the *.mat file that POSTFEKO support. You need to use the mat2ascii file to convert the file to a human readable format. The .mat.0, .mat.1, etc. indicates that you are running in parallel. Could you run sequential (single core) and run the mat2ascii utility on that file and see if you then get what you are looking for?

Altair Forum User

Perfect! Running FEKO in sequential mode produced a single .mat file that I was able to read using the mat2ascii utility. Your help fixed my problem in a jiffy.

If it helps anyone else, I managed to find a workaround that I used before this fix. I needed the complete NxN Y-parameter matrix of my antenna. But with one driving port (call it port p), the FEKO .out file only gives me one row/column of the Y-parameter matrix Y(:,p) = Y(p,:). So what I did was to create N ports on the structure (one on every segment of interest). I then ran the mesher once and adjusted its settings so that the segments were centered on my ports. (This worked except for the ports at the very tips of the antenna). I then ran N simulations, enabling only one port at a time. After each simulation, I extracted the single row/column of the Y-parameter matrix from the .out file, until I had the complete NxN matrix. Admittedly not very elegant, but it worked. Now that the mat2ascii utility is working, I will use that instead.

Thanks Jif