Suppose the tank settlement is defined by a displacement in vector D1.

Should D1 be added for the sustained stress check? That is, create a sustained case like SUS(W+P1+D1)?

Although displacement is a secondary type loading, in this instance it remains in the system like a sustained loading and does not cycle with temperature.

Another Aspect -

For example if we have OPE((W+P1=T1=D1) minus SUS(W+P1), isn't the result for the EXP stresses going to be very conservative because the D1 component does not cycle?

Paragraph 319.2.1.c (of B31.3) considers settlement as a single cycle effect that does not influence the fatigue life significantly.

How should the load cases for the EXP stresses be setup?

I'll appreciate your answers and insights.

Somnath