Determination of Hydrodynamic Characteristics of Ducted Propeller Using an Artificial Neural Network (CROSBI ID 614979)
Prilog sa skupa u zborniku | izvorni znanstveni rad | domaća recenzija
Podaci o odgovornosti
Valčić, Marko ; Dejhalla, Roko
engleski
Determination of Hydrodynamic Characteristics of Ducted Propeller Using an Artificial Neural Network
Dynamically positioned marine vessels are mostly equipped with azimuth thrusters with ducted propellers. The main reason for this is the possibility of azimuth thrusters to produce the demanded thrust in any horizontal direction, while the ducted propeller significantly increases the efficiency of heavily loaded screw propellers due to the negligible or very small propeller advance speeds. The test results of a systematic series developed at the institute MARIN are commonly used for the ducted propellers. An artificial neural network was trained using the regression polynomials of the propellers Ka 4-70 and Ka 4-55 in combination with the accelerating nozzle 19A which gives the possibility to determine the hydrodynamic characteristics of ducted propeller in open water conditions. The obtained results confirm the exceptional capabilities of neural networks, and the developed neural network provides the possibility of further applications for more accurate modeling of thruster in an otherwise very complex hydrodynamic model of propulsion of the dynamically positioned marine vessel.
ducted propeller; artificial neural network; dynamic positioning of a marine vessel
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
Podaci o prilogu
357-368.
2014.
objavljeno
Podaci o matičnoj publikaciji
Zbornik radova Sorta 2014
Dejhalla, R. i sur.
Rijeka: Tehnički fakultet Sveučilišta u Rijeci ; Brodogradilište Viktor Lenac
978-953-6326-90-7
Podaci o skupu
21. simpozij Teorija i praksa brodogradnje, in memoriam prof. Leopold Sorta, Sorta 2014
predavanje
02.10.2014-04.10.2014
Baška, otok Krk, Hrvatska