I’ve been trying to implement eigenvalue buckling simulation for solid elements

- October 12, 2024 at 6:49 am
  
  3343534645
  Subscriber
  
  I've been trying to implement eigenvalue buckling simulation for solid elements using MATLAB to perform linear buckling analysis. When calculating the geometric stiffness matrix for eigenvalue buckling, I referred to ANSYS documentation Help 3-61: https://www.mm.bme.hu/~gyebro/files/ans_help_v182/ans_thry/thy_geo3.html. However, after trying it out, I found that the geometric stiffness matrix calculated using the formula differs from the one output by ANSYS APDL, which makes me doubt the result. Does anyone know how to solve this? I’d appreciate any help or advice from experienced folks.
- October 14, 2024 at 2:45 am
  
  3343534645
  Subscriber
  
  I derived Cook's theory and also rewrote Borst's theory. When they introduced the calculation formula for the geometric stiffness matrix of solid elements, the geometric stiffness matrix they computed had non-zero elements. Why is the geometric stiffness matrix output by ANSYS APDL entirely composed of non-zero elements? Was the calculation incorrect?
- October 14, 2024 at 10:59 am
  
  3343534645
  Subscriber
  
  Thank you for your reply. Could you recommend any literature? I would also like to personally implement linear buckling analysis in MATLAB and benchmark it against ANSYS, which would be a very rewarding task! Thank you so much!
- October 15, 2024 at 8:53 am
  
  3343534645
  Subscriber
  
  Why haven't any ANSYS staff responded to me? Recently, I also tried Němec's theory (http://creativecommons.org/licenses/by/4.0/). However, the geometric stiffness matrix calculated from the theoretical formulas still doesn't match the one in ANSYS APDL. What could be the reason for this? I sincerely hope that someone knowledgeable can resolve this long-standing confusion for me.