Improving water surface current estimation from radar data

Title

Improving water surface current estimation from radar data

 Author

Bristogiannis, Yperion (TU Delft Electrical Engineering, Mathematics and Computer Science; TU Delft Delft Institute of Applied Mathematics)

Contributor

van Gijzen, M.B. (mentor)

Degree granting institution

Delft University of Technology

 Programme

Applied Mathematics

 Date

2021-02-25

 Abstract

The offshore industry can increase safety and efficiency during offshore operation by predicting actual wave induced ship-motions minutes ahead of time. By only using on-board equipment, such as the navigational radar, it is possible to deterministically predict incoming waves since the phase resolved
description of the waves can be acquired. The accuracy of the speed with which the these waves travel can be improved by taking into account the water surface current velocity. The state of the art method on computing the water surface current velocity is based on the three-dimensional Fourier transform over a large number of radar images followed by a LLSQ fit. For real time wave prediction, collecting a large number of radar images can result in long start up times.
This MSc thesis looks into an alternative method based of the polar Fourier tranform over fewer radar images, the conjugate product of the resulted spectra and a LLSQ fit. In addition, this thesis improves the method studied based on further work on the state of the art method found in literature. In the scope of this project the alternative method based on the polar Fourier transform is implemented as well as the suggested improvements. For evaluation, a comparison takes place between the alternative method and different approaches on the final part of the algorithm around the LLSQ fit. However, it is also shown that the goal, estimating the water surface current velocity, is potentially unachievable and the estimation of the effective velocity is proposed instead for the purpose of wave prediction.

 Subject

sea current estimation
radar images
polar Fourier transform
LLSQ fit

 To reference this document use:

http://resolver.tudelft.nl/uuid:c6b9227d-835c-4886-99b0-df4807fbdfc7

 Part of collection

Student theses

 Document type

master thesis

 Rights

© 2021 Yperion Bristogiannis
Files
PDF thesis_report_MSc_AP_Yper ... iannis.pdf 14.64 MB
Close viewer