A methodology is assessed for uncertainty quantification (UQ) of resistance, motions and slamming loads in variable regular wave representing a given sea state, and compared to irregular wave (benchmark) and deterministic regular wave studies. UQ is conducted over joint distribution of wave period and height; irregular wave inlet boundary condition is based on wave energy spectrum; deterministic study is conducted at most probable condition. Application to the high-speed Delft Catamaran at Fr=0.5 in sea state 6 is presented and discussed. Deterministic regular wave study shows average error for design optimization-related quantities (expected values of resistance, motions amplitude and slamming loads) equal to 25%. Variable regular wave UQ shows average error close to 6%, providing in addition empirical distribution functions. Extension to uncertain design through Karhunen-Loève expansion is presented and discussed; variable geometry studies show a potential reduction of 6.5% for calm water resistance and 3.9% for resistance in wave, with small variations in motions amplitudes; an increase of 6.4% of maximum slamming load is experienced by reducedresistance geometry, revealing a trade-off between performances and loads. UQ with metamodels reveals Kriging and polyharmonic spline as the most effective metamodels overall.
URANS Study of Delft Catamaran Total/Added Resistance, Motions and Slamming Loads in Head Sea Including Irregular Wave and Uncertainty Quantification for Variable Regular Wave and Geometry
Matteo Diez;Daniele Peri;
2012
Abstract
A methodology is assessed for uncertainty quantification (UQ) of resistance, motions and slamming loads in variable regular wave representing a given sea state, and compared to irregular wave (benchmark) and deterministic regular wave studies. UQ is conducted over joint distribution of wave period and height; irregular wave inlet boundary condition is based on wave energy spectrum; deterministic study is conducted at most probable condition. Application to the high-speed Delft Catamaran at Fr=0.5 in sea state 6 is presented and discussed. Deterministic regular wave study shows average error for design optimization-related quantities (expected values of resistance, motions amplitude and slamming loads) equal to 25%. Variable regular wave UQ shows average error close to 6%, providing in addition empirical distribution functions. Extension to uncertain design through Karhunen-Loève expansion is presented and discussed; variable geometry studies show a potential reduction of 6.5% for calm water resistance and 3.9% for resistance in wave, with small variations in motions amplitudes; an increase of 6.4% of maximum slamming load is experienced by reducedresistance geometry, revealing a trade-off between performances and loads. UQ with metamodels reveals Kriging and polyharmonic spline as the most effective metamodels overall.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


