# Import of M,D,K matrices after MOR

**URL:** <https://calculix.discourse.group/t/import-of-m-d-k-matrices-after-mor/522>\
**Category:** Uncategorized\
**Created:** [April 1, 2021, 1:06pm UTC](https://calculix.discourse.group/t/import-of-m-d-k-matrices-after-mor/522 "2021-04-01T13:06:34Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![FrancescoPassigato](https://avatars.discourse-cdn.com/v4/letter/f/958977/32.png) [@FrancescoPassigato](https://calculix.discourse.group/u/FrancescoPassigato)\
**Post date:** [April 1, 2021, 1:06pm UTC](https://calculix.discourse.group/t/import-of-m-d-k-matrices-after-mor/522/1 "2021-04-01T13:06:34Z")

</div>

Hi,

I would like to import the M,D,K matrices obtained by Model Order Reduction with an external tool. Is that possible?  
After that I would like to export the results of the analysis with Calculix to obtain the data to be insterted in a multibody simulation software (MBSim). The interface is already available, in fact I need following files: .frd, .dof, .mas, .sti, which are exported after a classical simulation with Calculix.

Thank you very much in advance  
Best regards  
Francesco Passigato

---

<div class="post-metadata">

**Author:** ![carlomontec](https://avatars.discourse-cdn.com/v4/letter/c/7c8e57/32.png) [@carlomontec](https://calculix.discourse.group/u/carlomontec)\
**Post date:** [April 1, 2021, 5:17pm UTC](https://calculix.discourse.group/t/import-of-m-d-k-matrices-after-mor/522/2 "2021-04-01T17:17:38Z")

</div>

Hi Francesco,

to my knowledge that is not possible in the current official implementation of CCX. Many of CCX data data structures are based in nodal information, so I could think its not trivial to use system matrices that have no nodal information.

Your objective is only to create output files from your current system matrices?

Because then you could create the DOF MAS and STI files in Python/Julia/MATLAB fairly straightforward.

The FRD file contains nodal data, and the node+element definition of the whole model, so I cannot see how can you elaborate an FRD if you don’t have the nodal information. If you can transform your ROM data into nodal data, you could create the FRD file manually, with the bare minimum to be readable by CGX.

Kind regards  
Carlo Monjaraz

---

<div class="post-metadata">

**Author:** ![FrancescoPassigato](https://avatars.discourse-cdn.com/v4/letter/f/958977/32.png) [@FrancescoPassigato](https://calculix.discourse.group/u/FrancescoPassigato)\
**Post date:** [April 8, 2021, 3:49pm UTC](https://calculix.discourse.group/t/import-of-m-d-k-matrices-after-mor/522/3 "2021-04-08T15:49:52Z")

</div>

Hi Carlo,

thank you very much for the reply.  
I think it would be no problem to export the nodal data from the current FEM Model. Therefore another question: which data/information are needed in Calculix, in order to carry out an analysis which produces the FRD file?

Thank you again for your support  
Best regards  
Francesco Passigato

---

<div class="post-metadata">

**Author:** ![carlomontec](https://avatars.discourse-cdn.com/v4/letter/c/7c8e57/32.png) [@carlomontec](https://calculix.discourse.group/u/carlomontec)\
**Post date:** [April 21, 2021, 8:08am UTC](https://calculix.discourse.group/t/import-of-m-d-k-matrices-after-mor/522/4 "2021-04-21T08:08:44Z")

</div>

Hi Francesco,

I am sorry for my late response.

I think any analysis will create an FRD as long as you request to create a results file. FRD files can be text or binary based. For your purposes the text one is definitely easier to use. The CGX (not CCX, attention! ) documentation has information of the FRD files.

Check the cards \*NODE FILE and \*EL FILE in the CCX documentation to output text based FRD files.