Article Published In Vol.2 (Sept-Oct-2014)

An Accurate Heave Signal Prediction Using Artificial Neural Network

Author : Mohammed El-Diasty

An accurate heave modeling is required for several applications, including hydrographic surveying. This paper proposes an adaptive heave signal modeling, which uses a neural network-based modelling. A recurrent neural network and three-layer feed forward neural network trained using the Levenberg-Marquardt learning algorithm is used for this purpose. Computational results with five different datasets of real time heave are provided to validate the effectiveness of the artificial neural network-based model. It is shown that the new neural network-based model give a reliable heave model with excellent performance. Also, a comparison is made between the developed artificial neural network models and the autoregressive models. It is shown that the new artificial neural network-based model give the best performance results (i.e., the mean square error MSE).

Keywords: Heave Prediction, Neural Network, Autoregressive, Recurrent Network