The “Variable: Forces” setting means that the FORC output vector (external - applied and reaction forces named RF in CalculiX) is used to visualize the (scaled) deformation:
This isn’t really a standard setting (it’s meant for very specific cases) and doesn’t mean that your model is wrong. Keep the default “Displacement” setting and switch to “True scale” to see the real deformation. If that’s ok then the model should be correct.
What surface ? The one with rigid body constraint or fixed BC applied ? Or are you still referring to the aforementioned postprocessing setting.
Btw. your model has a rather dense mesh, but consisting of C3D4 (linear tetrahedron) elements that should be avoided in most structural analyses. You should use C3D10 (quadratic tetrahedron) elements instead.
I have answered you in .pdf and 3D.stp&.mesh and the file attached in the link. Anyway, I have tried this method in (Cross-checked) Abaqus, “No extent surface damage was observed.”
Edit: Figure resource:
M. Zanchini, D. Longhi, S. Mantovani, F. Puglisi, and M. Giacalone, “Fatigue and failure analysis of aluminium and composite automotive wheel rims: Experimental and numerical investigation,” Engineering Failure Analysis, vol. 146, p. 107064, 2023, doi: 10.1016/j.engfailanal.2023.107064.
Ok, I see. They also used C3D4 elements in the paper, but they added SC8R and C3D8R elements too.
If fatigue calculations are your goal, you could try this open-source software: Fatlab download | SourceForge.net (it can use PrePoMax/CalculiX results too).
What the paper mentions about BCs and loads is:
The inner flange is fixed along its perimeter, and a multi-point constraint (RBE2) connects the centre hub housing to the loading point, which is 750 mm apart. A concentrated force is applied to the loading point so that the resulting bending moment is equal to that applied in the experimental fatigue test and according to Equation (3) and Table 2.
So you should make sure that you get the expected bending moment in your analysis.