Do you mean your system is qualified by B31.3 with SE< =SA and still has troubles due to thermal strain?

If yes (and the answer is based- more or less- on the example you've given), this would be really surprising vs. the fact B31.3 includes now the range stress due to axial loads in the calculated displacement stress range, SE. In brackets, it includes also axial load stress intensification factors.

About your remarks: I really don't know a criterion to understand when a thermal expansion of a piping system has the potential of behaving like a sustained load for system components. Usually is not more than what is called "an especially weak location in the system" and here is quoted the vessel nozzle connection. That's why the piping loads are often considered as sustained or primary load for vessel nozzle connection and I think is also a very prudent approach (but I can understand that nobody wants to have a junction nozzle-shell in a state "ready to become" a plastic hinge!).

However I'm afraid that considering "especially weak locations" everywhere in the piping system would end in the failure of the philosophy that has been successfully applied for almost a century. That's why I cannot confirm the similarity you've mentioned.