The prediction capability of recurrent-type neural networks is investigated for realtime short-term prediction (nowcasting) of ship motions in high sea state. Specifically, the performance of recurrent neural networks, long-short term memory, and gated recurrent units models are assessed and compared using a data set coming from computational fluid dynamics simulations of a self-propelled destroyer-type vessel in stern-quartering sea state 7. Time seriesof incident wave, ship motions, rudder angle, as well as immersion probes, are used as variables for a nowcasting problem. The objective is to obtain about 20 s ahead prediction. Overall, the three methods provide promising and comparable results.
RECURRENT-TYPE NEURAL NETWORKS FOR REAL-TIME SHORT-TERM PREDICTION OF SHIP MOTIONS IN HIGH SEA STATE
A Serani;M Diez
2021
Abstract
The prediction capability of recurrent-type neural networks is investigated for realtime short-term prediction (nowcasting) of ship motions in high sea state. Specifically, the performance of recurrent neural networks, long-short term memory, and gated recurrent units models are assessed and compared using a data set coming from computational fluid dynamics simulations of a self-propelled destroyer-type vessel in stern-quartering sea state 7. Time seriesof incident wave, ship motions, rudder angle, as well as immersion probes, are used as variables for a nowcasting problem. The objective is to obtain about 20 s ahead prediction. Overall, the three methods provide promising and comparable results.File | Dimensione | Formato | |
---|---|---|---|
prod_464619-doc_182253.pdf
solo utenti autorizzati
Descrizione: RECURRENT-TYPE NEURAL NETWORKS FOR REAL-TIME SHORT-TERM PREDICTION OF SHIP MOTIONS IN HIGH SEA STATE
Tipologia:
Versione Editoriale (PDF)
Licenza:
NON PUBBLICO - Accesso privato/ristretto
Dimensione
1.3 MB
Formato
Adobe PDF
|
1.3 MB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.