I have a linear static analysis that I have set up in Abaqus 2022 and Hyperworks 2022.2 using the Optistruct solver profile. The analysis is for a rectangle flat plate(1m x 2m), clamped on all edges, with a uniform pressure applied to the surface. The model uses second order shell elements with elastic material properties. Both models have been set up in the same way. I have attached both input decks.
I am using Roark's formula for max deflection at the centre as the comparison (6.595mm).
If I apply a coarse mesh (0.25) to both models then Abaqus will give a closer answer than Optistruct.
If I use a fine mesh (0.01) then both solvers give approximately the same answer (Abaqus still slightly better).
I would expect the deflection to be different between the solvers but there is a large disparity. Comparison of mesh refinement below.
Mesh | Abaqus | Optistruct |
|---|
0.25 | 6.321 | 7.715 |
0.05 | 6.595 | 6.64475 |
0.01 | 6.595 | 6.597 |
Can anyone advise why they are so different? Is it to do with how load is applied in Optistruct?
Thanks,
Iain
