Record Details

Neural networks to derive wave spectra

DSpace at IIT Bombay

View Archive Info
 
 
Field Value
 
Title Neural networks to derive wave spectra
 
Creator NAMEKAR, S
DEO, MC
 
Subject neural networks
wave spectra
ndbc data
network training
multi-peaked spectra
 
Description Knowledge of a design wave spectrum is needed in works like structural analysis, and laboratory wave simulations. Use of theoretical equations due to Pierson-Muskowitz (PM) and JONSWAP is traditionally made for this purpose. This paper presents an alternative approach based on neural networks. Networks were trained using a variety of learning schemes in order to estimate shapes of the wave spectra from given values of the representative wave height and period. The validation of the network for unseen inputs showed that the neural network could be a viable option in order to estimate the shape of the wave spectrum from the specified design wave parameters. The network-predicted spectral shapes were more satisfactory than those yielded by the common theoretical spectra. While use of available wave time history could be much beneficial for training, the network can also reasonably learn from the theoretical spectra, albeit with some loss of accuracy.
 
Publisher INTERNATIONAL SOCIETY OFFSHORE& POLAR ENGINEERS
 
Date 2011-10-27T21:45:38Z
2011-12-15T09:12:49Z
2011-10-27T21:45:38Z
2011-12-15T09:12:49Z
2004
 
Type Proceedings Paper
 
Identifier PROCEEDINGS OF THE FOURTEENTH (2004) INTERNATIONAL OFFSHORE AND POLAR ENGINEERING CONFERENCE, VOL 3,133-137
1-880653-62-1
1098-6189
http://dspace.library.iitb.ac.in/xmlui/handle/10054/16425
http://hdl.handle.net/100/2916
 
Source 14th International Offshore and Polar Engineering Conference (ISOPE 2004),Toulon, FRANCE,MAY 23-28, 2004
 
Language English