|
FhSim
3.1.0
Marine systems simulation
|
| ID | 0442 |
| Class | KNOWN-LIMITATION |
| Severity | 2 |
| Status | blocked |
Models: —
Found: 2026-11-03, test audit of the speed work (feature me-tw-rulings), at e0a3e14
Decision needed: whether to work on the load-law angle form (MARE-0131), the E-modulus and the Freeze schedule to bring these rows inside the 20 % band, or to accept the bounds.
tests/NetValidation_LaderCage_Test.cpp, suite NetValidation_LaderCage_Fig14Drag_VALIDATION: the model drag of the three Lader (2009) cages against KF2012 Fig. 14 at k = 0.05 (18 s, Freeze wake, loads read at 18 s). The KF2012 total uncertainty band is +-20 %. Five of the 21 rows are above +20 % (every one above the measurement):
| Row | Model [N] | Measured [N] | Error |
|---|---|---|---|
| Sn 0.19, U 0.12 m/s | 10.661 | 8.1 | +31.6 % |
| Sn 0.35, U 0.26 m/s | 74.716 | 57 | +31.1 % |
| Sn 0.43, U 0.26 m/s | 88.031 | 70 | +25.8 % |
| Sn 0.43, U 0.76 m/s | 207.915 | 165.1 | +25.9 % |
| Sn 0.43, U 0.93 m/s | 237.731 | 173 | +37.4 % |
Close to the band: Sn 0.35 U 0.39 (+19.79 %) and Sn 0.43 U 0.12 (+19.67 %). No row is below -20 %.
Bound: owner ruling 2026-11-03 (2 rows expected, 5 found). The test asserts -20 % <= error <= bound with the error rounded up to the next 5 %: +35, +35, +30, +30 and +40 % in kKnownDragDeviations; all other rows within +-20 %.
The model over-predicts the drag of the denser cages at low speed and of Sn 0.43 at high speed by up to 37 %.
Known reasons (doc/user/validation/lader2009.md, section "S3 drag against KF2012 Fig. 14"): the default law's angle form (MARE-0131; KF2012 find the same over-prediction for Sn 0.35 and 0.43 and attribute it to the angle harmonics a3, b4: -28 % drag for Sn 0.43 at 1 m/s from a3 = 0 to 0.1); the provisional E-modulus (awaiting the author), which shapes the cage at low speed; the Freeze schedule, whose round has not converged at Sn 0.35 and 0.43 when the loads are read. Not tuned (R20).
Remove the entry of the row from kKnownDragDeviations; the row then has to be within 20 %.
A bound that stays after the cause is fixed hides a regression up to that bound. Lower or remove the bounds when MARE-0131 is resolved.