For those who have not yet purchased the new Peng book, his argument here was first published in an ASME paper. That paper is available on the web at Pressure Elongation.
There's a picture in that document (and not in the book) that settles what I believe to be misunderstanding here. I am referring to the phrase "pressure elongation generates a displacement that is a self-limiting load". He is NOT saying that pressure is self-limiting. He is saying that it generates a DISPLACEMENT that is self-limiting. The illustration shows a small diameter branch coming off a long, large diameter run. The long run exends along its axis due to the force of pressure. This pressure-based deflection of the header is the self-limiting load on the branch piping. The branch must have sufficient flexibility to safely accept this pressure displacement.
Pressure is NOT self-limiting. If it was, expansion joints would not need tie rods.
I might suggest the following approach to address this load in a CAESAR II analysis. Again this is a LONG run with branch piping and we're concerned about branch stress.
1) turn on Bourdon (translation only) to see where the header moves due to pressure alone
2) enter this deflection as a length of header pipe in the proper side of and close to the branch and call it a CUT LONG
3) with Bourdon turned off, sustained stress will come from W+P
4) expansion stress range will come from (T+W+P+CS)-(W+P)
Of course there can be several other ways of doing it. (I didn't run this so let me know if it works.)