Hi there,
I would like to model an RC retaining wall in Plaxis 2D (V21). The wall has a varying thickness, which is 1.3 m thick at the bottom and 0.4 m thick at the top (please see the Figure below). And below the wall, there's a 1.4 m x 1.7 m footing (which then connects to a bored pile).
It is common practice to model such a wall with a plate element. However, I wonder how to correctly demonstrate the area (A) and inertial moment (I) value of this varying thickness.:
1. Can I model it as a single plate with an "average" thickness of (1.3 + 0.4)/2 = 0.85 m ? However, with this model, the EA and EI of the plate at the bottom (where it intersects with the footing) will be lower than reality, thus resulting in a lower moment force.
2. If I model it as a plate with an "average" thickness of 0.85 m, the moment force around the top of the pile might also get a bit bigger than reality (due to the higher EA and EI).
3. Should I average the plate into a few "sections" (e.g. the average is made every 2 m in length)?
I believe these are very simple questions for experienced Plaxis users. I am looking forward to hearing from you. Thank you for your time and attention.
Kind regards,
M. Rayyan
Dear Muhammad,
I think you sum up the problem of using a single plate very well, nothing to add to that. If you follow the approach under 3) it would seem that it's more accurate as to some degree one takes into account the increase of stiffness with depth. However, note that there will be sudden changes in stiffness, which may cause peak bending moments where the stiffness changes, and the peak bending moments are not realistic either. In that respect using more sections and thus smaller sudden changes in stiffness would be better.
Or you don't use a plate at all, but model the wall exactly using volume elements with a linear elastic material set. You can obtain structural forces in Output using the option "Structural forces from volume plates" in the Tools menu. Note that for an accurate results you would have to seriously refine the mesh in your wall then.
With kind regards,
Dennis Waterman
Answer Verified By: Muhammad Rayyan
Hi Dennis,
Thank you very much for your advice! I see, seems like using the "Structural forces from volume plates" would be the most accurate option here, I'll try to use it. I really appreciate your help.