Peter, thanks again for your time and efforts in helping me.

When I said "increased displacements in the inner part of the excavation" I meant that the right simulation would have given me a displacement contour where vectors have direction towards the free faces of the excavation, so it's like the excavation shrunk (the technical term is convergence). The rest of faces have to remain without displacement (outer part of the box). In consequence, the highest displacements occur near the excavation.

Also, the min and max principal stresses should always be concentrated around the excavation (induced stresses), like in this example (keep in mind that this is not the same project, just an example to illustrate the idea):

And well... that's it. As a remainder, this is a cube (box) with an excavation (mining stope) inside and is under the action of s1, s2 and s3 compressive stresses just like this image below. And, the boundaries of the box should be constrained to have zero displacements. 

The issue is that these stress cancel each other out if I model this situation as in other softwares. I hope now you can understand what I meant.


My project is attached, dear Peter. Thanks again for everything.