I have modeled a plane stress plate in Abaqus with the following properties:

• Height (H): 40
• Length (L): 80
• Thickness (d): 3
• Elastic Modulus (E): 2×10^9
• Poisson's Ratio (ν): 0.25

A large displacement of Delta = 10 was applied to the upper edge with *NLGEOM=ON to include geometric nonlinearity. The deformation is homogeneous (all points of the plate have the same deformation), and the computed stress S11=0 and S22=4.463×10^8 matches the 3D model results exactly, which suggests the stress computation is reliable.

However, the reaction force from the 2D plane stress model is 8.5687×10^10, while the expected value (based on S22×deformed area; deformed area = (80 - 4.341)(3 - 0.1628)) is 9.5801×10^10 (This is also verified by using the 3D model).

I also tested with a modified thickness of 1, which did not resolve the discrepancy.

Has anyone encountered this kind of inconsistency in Abaqus for plane stress elements under large deformation? Is there a known limitation or a setup detail I may be overlooking?

Note: The plane strain model with the same setup gives the correct reaction force (S22×deformed area), albeit different from 3D model.

PS: I have attached my input file for review.

Any insights or suggestions would be greatly appreciated!