Dear Dave,
I have to model a 8" buried pipeline. In this preliminary phase I have to verify only the pipeline sections at starting and ending point and my battery limit is the first support above ground ( my model starts at this support and I will provide the axial displacement). But, my question Is:

I have a 8" pipeline, buried depth=1,5m and 100°C design temperature; for contain the thermal expansion i placed a buried expansion loop about at 90 meters from the vertical bend(where the pipeline become buried), then I modeled further 250 meters of pipeline up to virtual anchor point. the loop and vertical bends are 40 ND radius (this is a Client requirement, they don't want hot bend) .I have some doubts regard the output results:

1) I modeled the soil by Peng method and ALA Method:
-In the first case the axial displacement at first above support is 88mm
-In the second case the axial displacement at first above support is 70 mm.
As you already said above, maybe ALA method is better than Peng, then I can consider it.

2) I tried also to model a straight pipe section up to virtual anchor point (250 meters). In this case the displacement at first support is 90 mm (axial) and 50mm (vertical); Instead with an expansion loop I have a 70 mm axial displacement. Then, I think that, the expansion loop does not work enough, maybe because the bend radius are so large that the bend is not able to dissipate the thermal expansion (As opposed to have a hot tight bend that works better).
Finally, It's better to insert an expansion loop that absorbs only 20 mm of displacement or it is better to have a straight pipe?

3)Otherwise, What Can I do? I tried to reduce the distance between the vertical AG/UG bend and the loop, I tried to increase the buried depth, I tried to increase the loop width but the results is ever the same.
Please help me, thank you