Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor | Multi-disciplinary Studies | en_US |
| dc.contributor | Department of Civil and Structural Engineering | en_US |
| dc.creator | Tam, Chin-hung | - |
| dc.identifier.uri | https://theses.lib.polyu.edu.hk/handle/200/4068 | - |
| dc.language | English | en_US |
| dc.publisher | Hong Kong Polytechnic University | - |
| dc.rights | All rights reserved | en_US |
| dc.title | Prediction of ocean waves in Hong Kong western waters by using artificial neural networks | en_US |
| dcterms.abstract | The potential of using artificial neural networks (ANNs) with back-propagation algorithm and time delay technique for prediction of ocean waves in Kau Yi Chau (KYC) and West Lamma Channel (WLC), Hong Kong is explored. MATLB with Neural Network Toolbox 3.0 was employed to develop the ANNs. Hourly significant wave height and corresponding wind parameters of each location were placed in the ANN. Because of instruments failure or servicing, the data must have been missed. The interpolation method had to be employed in order to fill in the data gap. The interpolated data were filled in only within two-hour gap. Each location of data was divided into two categories. One of the categories used the whole set of data to develop a wind-generated waves network. Another category used the high-waves data (>= 1 m) to develop high-waves network. Each set of data was also divided into training, validation and testing sets. During the ANNs manipulation, three layers of ANN were employed. The error measurement of the ANNs performance and training rule was based on the default mean squared errors (MSE) and Marquardt-Levenberg respectively in MATLAB. Different transfer functions and number of neurons were also tested in order to improving output accuracy. The results were analyzed by statistical method, such as mean error (ME), average error, mean squared errors (MSE), residual standard error (RSE), coefficient of determination (R2) and F - test. After the statistic analysis, two individual Kau Yi Chau networks were developed for general condition (3 layers with 5-time delay, 10-20-1, tansig: logsig: purlin) and high-waves situation (3 layers with 3-time delay, 3-15-1, tansig: logsig: purlin). During the general condition in KYC network, the percentage of average error and the coefficient of determination was 28.19% and 0.3732 respectively. Moreover, the percentage of average error and the coefficient of determination was 9.63% and 0.2323 respectively during high-waves situation. One network was developed only for West Lamma Channel during general condition (3 layers with 15-time delay, 10-15-1, tansig: tansig: purlin). The percentage of average error and the coefficient of determination was 32.27% and 0.2475 respectively. The results of F -test were positive in all of the three networks. They inferred that they had a statistically significant relationship between wind and wave. | en_US |
| dcterms.extent | viii, 103 leaves : ill. ; 30 cm | en_US |
| dcterms.isPartOf | PolyU Electronic Theses | en_US |
| dcterms.issued | 2000 | en_US |
| dcterms.educationalLevel | All Master | en_US |
| dcterms.educationalLevel | M.Sc. | en_US |
| dcterms.LCSH | Ocean waves -- China -- Hong Kong | en_US |
| dcterms.LCSH | Neural networks (Computer science) -- China -- Hong Kong | en_US |
| dcterms.LCSH | Hong Kong Polytechnic University -- Dissertations | en_US |
| dcterms.accessRights | restricted access | en_US |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| b15179692.pdf | For All Users (off-campus access for PolyU Staff & Students only) | 3.61 MB | Adobe PDF | View/Open |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/4068