Naslov
Artificial neural network boat seakeeping model based on full scale measurements
(Artificial network boat seakeeping model based on full scale measurements)

Autori
Matić, Petar ; Katalinić, Marko

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), stručni

Izvornik
ICTS 2020 Maritime, transport and logistics science conference proceedings / Marina, Zanne ; Patricija, Bajec ; Elen Twrdy ; - Portorož : Fakulteta za pomorstvo in promet Univerza v Ljubljani, 2020, 226-230

ISBN
978-961-7041-08-8

Skup
19th International Conference on Transport Science (ICTS 2020)

Mjesto i datum
Portorož, Slovenija, 17.09.2020. - 18.09.2020

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
seakeeping response ; small boat ; full scale measurements ; artificial neural network

Sažetak
Heave response of a boat is evaluated based on full scale seakeeping measurements. Vessel motions, position, speed and heading are recorded during sea trials in small-to-medium waves relative to the boat size. Motion data is collected by a navigation grade inertial motion sensor unit and the sea state is noted from a numerical wave model available for the test region. Small vessels are subject to non-linear response and, with sensor recordings delivering large quantities of motion data, artificial neural networks (ANN) are a proven tool to map such behavior. The collected data is analyzed and a heave response prediction model is developed and optimized. The work presents preliminary communication of efforts to combine the disciplines of experimental seakeeping and artificial intelligence data analysis. The evaluation of ANN’s capability and accuracy in predicting seakeeping response of a small vessel in moderate waves can be used to set directions for further investigation.

Izvorni jezik
Engleski

Znanstvena područja
Brodogradnja, Elektrotehnika

POVEZANOST RADA