查詢結果分析
來源資料
頁籤選單縮合
題 名 | Prediction of Typhoon Swells Using Neural Networks=應用類神經網路於颱風湧浪之預報 |
---|---|
作 者 | 蕭松山; 丁肇隆; 林銘崇; 蘇昭安; | 書刊名 | 海洋工程學刊 |
卷 期 | 7:2 2007.12[民96.12] |
頁 次 | 頁25-45 |
分類號 | 328.888 |
關鍵詞 | 預報; 類神經網路; 颱風湧浪; Predict; Neural network; Typhoon swell; |
語 文 | 英文(English) |
中文摘要 | 本文應用倒傳遞類神經網路建立一颱風湧浪預報模式,文中選用了中央氣象局花蓮測站之15場颱風波浪實測資料,作為預報模式的建立、測試與驗證。所使用之類神經網路架構輸入層參數,係採用颱風最大風速W(下標 max)、颱風中心移動速度V、颱風中心與測站距離d、颱風七級風暴風半徑R7、颱風方位角θ1、颱風中心移動方向與颱風中心到測站連線的夾角θ2及觀測站波高H(下標 b)等七個參數之即時與歷時資料,而以未來觀測站波高H(下標 p)為輸出層參數。本文模式對於各場颱風中之觀測站示性波高預報值與實測值比較之相對誤差均小於13.5%。 |
英文摘要 | A back-propagation neural network was employed to predict swells generated by typhoons, which damages of property and take human life in nearshore areas. Fifteen sets of records of wave-heights during typhoons around the east coast of Taiwan were used to develop and validate the network. The current and previous values of the maximum speed of the wind storms, W(subscript max); the speed of the typhoon, V; the distance from the center of typhoon to the observing station, d; the radius of force 7 wind, R7; the azimuthal angle, θ1 measured clockwise from north to the direction of motion, and the angle between the direction of motion of the typhoon and the line that connects the center of the typhoon to the observing station, θ2; and the significant wave heights recorded at the stations, H(subscript b), were used as input parameters. The predicted significant wave heights, H(subscript p), were selected as the output parameters. The network predicts the maximum significant wave heights quite effectively and the differences between the predicted and measured maximum wave heights are less than 13.5 % of the maximum measured wave heights. All the evidence shows that the developed Back-Propagation Neural Network is an effective method for predicting a swell. |
本系統中英文摘要資訊取自各篇刊載內容。