碩士 / 國立臺灣海洋大學 / 海洋環境資訊系 / 105 / This study constructed a predictive model for the wave period for typhoon waves occurred at the Longdong coast on the northeastern coast of Taiwan. The model was based on the back propagation neural networks. For the model, the typhoon information obtained from the Central Weather Bureau and typhoon wave data from the Longdong wave buoy were used. Twenty-one typhoons occurred between 2000 and 2015 and tracked past near the wave station were selected. Among them, 18 were used for training and 3 for validation. Our study found that the model using the following input parameters has the best results: storm radius of the wind force 7 on the Beaufort scale, the speed of typhoon center relative to the wave station, the distance of the typhoon center to the wave station, and the bearing of the typhoon relative to the wave station. The typhoon distance and the bearing were represented by fuzzy membership functions. All the input parameters values obtained up to 6 hours prior to the current calculation time step were used, and the model made a 2-hour prediction for the mean wave period. From the three validation typhoons, we found that the mean correlation coefficient between the measured and the predicted wave periods was 0.77 with the root mean square error of 0.87 sec. The relative errors between the peak typhoon wave period in each typhoon episode predicted by the model and that of the measured were,-6.8%, 1.4%, and -6.5%, respectively. The negative value means that the predicted peak wave period is smaller than the measured value. Our results also showed that the time differences between the predicted occurrence time of the peak mean wave period and the measured occurrence time were 0, -2, and 0 hours, respectively. The negative value represents the predicted time is ahead of the measured time.
| Identifer | oai:union.ndltd.org:TW/105NTOU5276024 |
| Date | January 2017 |
| Creators | Lee, Wen-Hua, 李文華 |
| Contributors | Tsai, Chen-Han, 蔡政翰 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 46 |
Page generated in 0.0082 seconds