IAHR Document Library

« Back to Library Homepage « Proceedings of the 18th IAHR APD Congress (Jeju, 2012)

Peak Wave Height Predictions Using the Soft Computing Techniques

Author(s): S. P. Nitsure; S. N. Londhe; K. C. Khare

Linked Author(s):

Keywords: Significant Wave Height; Soft Computing Techniques; Genetic Programming; Artificial Neural Networks; Kurtosis

Abstract: Ocean waves generated by extreme events such as hurricanes have devastating effects on the ships and marine structures. Safety of maritime activities relies heavily on the knowledge of wave parameters. Therefore researchers have been using various techniques to predict Significant Wave Heights (Hs) as accurately as possible. This paper describes the work of Hs forecasting with focus on the peak wave height predictions at two stations (which are frequently hit by hurricanes) near the US coastline. Hs are predicted with lead times of 12 to 48 h using soft computing techniques of Genetic Programming and Artificial Neural Networks. The innovation in this work is the data selection. Data with high kurtosis for Hs (indicating more peakedness) are selected. Models using current and previous components of shear wind velocity and the wind directions as inputs yielded better predictions. The qualitative and quantitative analyses indicate that the peak waves are predicted satisfactorily. GP performed nearly always better than the ANN. It seems that GP has understood the complex phenomena of formation and growth of peak waves.


Year: 2012

Copyright © 2024 International Association for Hydro-Environment Engineering and Research. All rights reserved. | Terms and Conditions