Hi,

This is not directly related to Caesar II software; my question is with regards to the necessity of combining WIND and Wave impacts on the same load cases for offshore piping systems; B31.3 does not suggest combining 2 different occasional events in one load case however, this seems to be an unwritten rule that I see so many stress engineers consider W+T+P+WIND+WAVE case as the governing scenario to assess the loads/stresses instead of assuming W+T+P+WIND and W+T+P+WAVE separately. Can anyone give a reasonable explanation for this plaese?


many thanks in advance,

Hurricanes probably result in significant wind and wave events.

Underline added by me.

Wave loads aren't due to earthquakes, even though it's common for them to be analyzed in the exact same manner.

The problem with combining wind and wave is that it's not always clear which combination will result in the highest stresses.

1) It's generally accepted that maximum stress will happen when occasional loads are applied laterally to the pipe.
2) It's generally accepted that pipe will be oriented in the cardinal directions (N/S/E/W).
3) It's generally accepted that occasional loads would therefore be applied in 4 directions.
4)If you wanted every combination in the 4 directions for two independently acting occasional loads, then you will have 16 (OPE) load cases and 32 (OCC) resulting load cases. This is ignoring the previous 8 (OPE) load cases, and their 16 (OCC) load cases, where we consider wind acting alone and wave acting alone. Therefore, we have a minimum of 24 (OPE) cases and 48 (OCC) cases to cover wind and wave.
5) Also consider that in the offshore world, you also have transportation load cases that are run similarly to wind and wave, which could conceivably double the number of load cases to 48 (OPE) and 96 (OCC).

Note, this is considering only the 4 cardinal directions.

I will note that I've also considered in the past the 4 ordinal directions on top of the 4 cardinal directions, e.g. for just wind, and have found cases of failure for one of the ordinal directions that didn't trigger for the cardinal directions.

So now if we wanted to run that complete gamut, then we'd have:

8(OPE) for wind alone
8(OPE) for seismic alone
64(OPE) for wind + seismic combined
plus double the quantity above for OCC equivalents, resulting in a total of 80(OPE) cases 160 (OCC) cases, again, not counting transportation cases, which would double these values to 160 (OPE) and 320 (OCC).

Even if you used a spreadsheet to save yourself time to generate all the load cases and pasted them into CAESAR, chances for CAESAR non-convergence would come at an all-time high.

Thus, if you were to combine wind and wave, there's likely a simplified combination method that is assumed to be the most conservative, but not necessarily proven to be.

There's a generalized expectation from these floating monoliths to shut down during a hurricane, however, and then assess the damage and limp along until it's repaired.

Hi Pooria,

I suggest to have a look on DNV-RP-D101 (older ed. 2008), Appendix G. Unfortunately, the newer 2017 Ed. does not include any more the tabulated loading cases.

You'll find there that WIN and "U" Inertial Load (that assimilates Wave effect, of course) are initially summed separately using SRSS method, and then resultant-combined WIN and "U" loads are summed using Scalar method to determine the overall effect.

In general, SRSS summation might be used to avoid the development of those complicated loads cases' series...and the obtained results are on the conservative side...
I've seen similar approaches on some past FPSO projects with reputed engineering companies.

Best regards,