Extraction of stiffness matrix

[limaomaolwh]

Hello. I am currently using your open-source program to calculate the stress distribution in a beam. While extracting the stiffness matrix from the 'system.k' , I noticed that the dimension of the stiffness matrix is much larger than I anticipated. I am a bit confused and would greatly appreciate it if you could provide some clarification on this matter.

Could you please explain why the dimension of the stiffness matrix is larger than expected? It would be helpful if you could provide some additional background information or insights on this issue.

Thank you in advance for your time and assistance. I look forward to hearing from you.

[limaomaolwh]

I have learned that the 18 columns of the stiffness matrix of each additional unit during the dynamics analysis are
u1 u2 u3 θ1 θ2 θ3 F1 F2 F3 M1 M2 M3 P1 P2 P3 H1 H2 H3.
But when the number of cells is minimum still reaches 30, I don't know what the extra 12 are respectively.

[Cardoza2]

@limaomaolwh Sorry for the late response. We were in between maintainers due to the fact that the author of the package graduated and left our lab. I'm still learning the in's and outs of the code package, but from what I can tell system.K is the Jacobian of the system, not the stiffness matrix. If you would like the stiffness matrix, you need the inverse of the compliance matrix found in each beam element. so for instance inv(assembly.elements[end].compliance) should return the stiffness of the final element. However, if you created the assembly, you should already have that. If you're looking for for stress-strain recovery, GXBeam has a FEA solver built in for cross-sectional analysis and has convenience functions for that. You can see more on that in the documentation. Let me know if there are any other questions, otherwise I'll mark this issue resolved.