Please use this identifier to cite or link to this item:
|Title: ||Prediction of tides using hydrodynamic and neural network approaches|
|Authors: ||Vivekanandan, N.|
|Keywords: ||Artificial neural network;back propagation;conservation of laws;hydrodynamic;tide|
|Issue Date: ||Mar-2003|
subcontinent with a long coastline distributed among nine coastal states and
the islands group of Andaman-Nicobar and Lakshdeep Islands,
requires prediction of tides at desired location, based on data from one place
to another place, for planning and design needs, especially at interior of bays
or at the vicinity of civil structures. There is no exact analytical model that
can predict tides at desired locations, since the phenomenon involved is
uncertain and random in nature. Artificial Neural Network (ANN) model provides
a non-hydrodynamic mapping between given sets of input and output values.
Built-in-dynamism in network tracing, data error tolerance and lack of
requirements of any exogenous input, etc, make neural network modelling
attractive. This paper reports a study to predict tides at Navalakhi station
based on reference station at Okha of Gulf of Kutch. The paper also shows show
that the ANN results of prediction of tides at Navalakhi could be encapsulated
in hydrodynamic model so as to save enormous efforts involved in CPU time as
well as long duration of boundary conditions at Okha to be prescribed.
|Appears in Collections:||IJMS Vol.32(1) [March 2003]|
Items in NOPR are protected by copyright, with all rights reserved, unless otherwise indicated.