Toggle Main Menu Toggle Search

ePrints

Wave Modelling and Forecasting with Artificial Neural Networks

Lookup NU author(s): Dr Maryam Haroutunian, Dr David Trodden

Downloads


Licence

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0).


Abstract

There is a requirement for some marine vessels to know the exact condition of the seaway in which they are operating. Currently this is accomplished with the usage of wave radars, which are expensive and sometimes not viable for smaller vessels that can be more greatly affected by waves. This research utilizes image processing produced in the Newcastle University’s Hydrodynamics Laboratory with artificial neural networks to analyse current and future wave behaviour. The image processing is completed using two inexpensive digital cameras to reproduce waveforms over a certain time period. The artificial neural networks are tested over computer generated wave forms and then integrated with the wave forms captured from the digital cameras with analysis of both past, current, and future wave characteristics being analysed. The success of the image processing and neural networks in the laboratory setting provides encouragement for the future success of the project to be completed further with testing on an actual vessel and with an increasing scope of imaging.


Publication metadata

Author(s): Doyle C, Lee Y, Haroutunian M, Trodden DG

Publication type: Conference Proceedings (inc. Abstract)

Publication status: Published

Conference Name: 5th International Conference on Advanced Model Measurement Technology for The Maritime Industry (AMT’17)

Year of Conference: 2017

Pages: 299-310

Online publication date: 13/10/2017

Acceptance date: 18/08/2017

Publisher: University of Strathclyde

URL: https://www.dropbox.com/s/w2xqqhnsma289d0/AMT17%20Proceedings.pdf?dl=0


Share