Hello.
For other purposes, which cannot be done in ANSYS, I extract the mass and stiffness matrices. As a check I use the extracted mass and stiffness matrices to compute the natural frequencies. Doing so does not recreate the natural frequencies predicted by ANSYS.
I extract the matrices from ANSYS in the following way:
*SMAT,MatM,D,import,full,file.full,MASS
*print,MatM,MMAT_ANSYS.txt
I then load the matrices into matlab:
https://se.mathworks.com/matlabcentral/answers/69394-how-can-i-import-the-stiffness-matrix-generated-by-ansys-and-contained-in-a-file-txt
and make them symmetric.
How do I, from the extracted matrices read into Matlab, then compute the natural frequencies predicted by ANSYS, which I have confirmed by an analytical example?
I do not believe sorting of degrees of freedom to be an issue, since it is only a line of beam elements.
I have attached the matrices from ANSYS and Matlab, along with my complete ANSYS APDL script.
If any other information is needed, please let me know, and I will provide it.
Thank you in advance!
Kind Regards
Christoffer
Hi,
If you have the mass and the stiffness matrices, then you can simply use the
"[V, D]=eig(M, K)" command,
in Matlab.
In the response to this code, you have two matrices (V and D). The columns of V present eigenvectors of A (mode shapes). The diagonal matrix D contains eigenvalues (natural frequencies).
Hi Emadaldin,
I am aware of doing your suggested actions, but this does not produce the natural frequencies predicted by ANSYS.
https://www.researchgate.net/project/PERMAS4EDU/update/5f92bccb7600090001f050ae
- Badges
- Science topic
- Similar topics
- Engineering
- Computer Engineering
- Computer