English  |  正體中文  |  简体中文  |  Items with full text/Total items : 27221/39064
Visitors : 2403661      Online Users : 67
RC Version 4.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
Scope Adv. Search

Please use this identifier to cite or link to this item: http://ntour.ntou.edu.tw:8080/ir/handle/987654321/29944

Title: 以類神經網路預測龍洞海域的颱風波浪
Prediction of typhoon waves at Longdong using artificial neural networks
Authors: Fan-Yu Chiu
Contributors: NTOU:Department of Marine Environmental Informatics
Keywords: 颱風波浪;類神經網路
typhoon waves;neural network model
Date: 2011
Issue Date: 2011-11-25T03:30:19Z
Abstract: 本研究以易與取得的氣象資料為原則,建構一個計算簡單且快速推算的颱風波浪模式推算,模式中係以倒傳遞類神經網路(BPNN)為架構,在2002-2007年間本文找到可用的資料中分成12個學習颱風,3個測試颱風,本研究依風場參數不同分為3個模式,建立在台灣東北角龍洞海域颱風波浪推算模式。本文經驗證與比較得到最佳模式為採用中央氣象局所發佈的颱風規模與路徑,配合聯合颱風警報中心(JTWC)提供的颱風暴風半徑四個象限大小所計算的暴風半徑涵蓋面積做為風場輸入資料,與颱風中心相對測站偏移速度、颱風中心與觀測站距離、颱風方位角作為影響測站未來波高的輸入參數。模式中以模糊歸屬函數修正颱風受地形效應的影響,並取輸入參數6小時延時資料,推算不同路徑颱風對測站下兩小時颱風波浪的波高。推算結果顯示其推算值與實測值波高之相關係數CC 為0.839;均方根誤差RMSE 為76公分;3個測試颱風波高峰值誤差與最大峰值百分比分別為百分之 5、 9、 2.9;峰值時間誤差分別為0、 0、 -2小時,故由結果顯示使用本模式所建構的倒傳遞類神經網路能有效的推算颱風波浪的波高。
This study collected typhoon information, occurred between 2002 and 2007, and wave height measured at Longdon buoy station to construct 2-hour typhoon wave predictive models for the coastal sea near Longdon. The model was based on back propagation neural network method. Typhoon information was obtained from the Central Weather Bureau and the Joint Typhoon Warning Center (JTWC). After examining all the data, it was found that there were fifteen typhoons having sufficient typhoon parameters and wave height for model building. Among the 15 typhoons, 12 of them were used for training and 3 for validation. We tested three models, which consisted different parameters to represent the strength of typhoon. Tests of this study found that the model using typhoon area based on the radii of the four quadrants reported by JTWC, the speed of typhoon center relative to the Longdong wave station, the distance of the typhoon center and the bearing of the typhoon as input parameters has the best result. It is to note that the latter two parameters were represented by fuzzy membership functions. The validation runs showed that the correlation coefficient between the 2-hour predicted significant wave height and the measured wave height was 0.839, while the root mean square error was 76 cm. The percentage errors in the peak wave heights during the typhoon period between the model and measurement for the three typhoons were 5%, 9%, and 2.9 % respectively. The errors in the time of occurrence of the peak wave height were 0, 0, and -2 hours respectively. The minus sign means that the predicted peak wave height occurs earlier than the measured. The validation showed that the model constructed in this study can satisfactorily predict typhoon wave height.
URI: http://ethesys.lib.ntou.edu.tw/cdrfb3/record/#G0T984E0012
Appears in Collections:[海洋環境資訊系] 博碩士論文

Files in This Item:

File Description SizeFormat

All items in NTOUR are protected by copyright, with all rights reserved.


著作權政策宣告: 本網站之內容為國立臺灣海洋大學所收錄之機構典藏,無償提供學術研究與公眾教育等公益性使用,請合理使用本網站之內容,以尊重著作權人之權益。
網站維護: 海大圖資處 圖書系統組
DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - Feedback