Please use this identifier to cite or link to this item: http://nopr.niscair.res.in/handle/123456789/26430
Title: Real Time wave forecasting using artificial neural network with varying input parameter
Authors: Vimala, J.
Latha, G.
Venkatesan, R.
Keywords: ANN;Wave Forecasting;Correlation coefficient;Neural network;Computational elements
Issue Date: Jan-2014
Publisher: NISCAIR-CSIR, India
Abstract: Prediction of significant wave heights (Hs) is of immense importance in ocean and coastal engineering applications. The aim of this study is to predict significant wave height values at buoy locations with the lead time of 3,6,12 and 24 hours using past observations of wind and wave parameters applying Artificial Neural Network. Although there exists a number of wave height estimation models, they do not consider all causative factors without any approximation and consequently their results are more or less a general approximation of the overall dynamic behaviour. Since soft computing techniques are totally data driven, based on the duration of the data availability they can be used for prediction. In the National data buoy program of National institute of Ocean Technology, not all the buoys have wind sensors and wave sensors and so it is attempted to apply neural network algorithms for prediction of wave heights using wind speed only as the input and then using only wave height as the input. The measurement made by the data buoy at DS3 location in Bay of Bengal (12o11’21"N and 90o43’33"E) are considered, for the period 2003-2004. Out of this, the data of period Jan 2003-Dec 2003 was used for training and the data for the period July 2004- Nov 2004 is used for testing. Real time wave forecasting for 3,6,12 and 24 hours were carried out for a month at the location chosen and the results show that the ANN technique proves encouraging for wave forecasting. Performance of ANN for varying inputs have been analysed and the results are discussed.
Page(s): 82-87
URI: http://hdl.handle.net/123456789/26430
ISSN: 0975-1033 (Online); 0379-5136 (Print)
Appears in Collections:IJMS Vol.43(01) [January 2014]

Files in This Item:
File Description SizeFormat 
IJMS 43(1) 82-87.pdf441.5 kBAdobe PDFView/Open


Items in NOPR are protected by copyright, with all rights reserved, unless otherwise indicated.