OK, but to make the problem easier let's imagine a system with one operation temperature and the whole system at the same temperature. If one +Y support lifts 100 mm during operation, is it "correct" the sustained stess calculated with CAESAR? (let's forget the influence of that support for expansion range for example). Is it reasonable to consider that "if the pipe became overstressed, it would yield and sag back to the support"?. I see that the yield and sag back is clear for small lifts but in case of doubt what I usually do is to make a copy of my model and run a cold sustained case with the supports that lift in operation including a Y gap of the same length than the lift during operation in order to check the sustained stresses.
Which would be the stress in the pipe to sag back 100 mm?

Regards and thank you for any reply