跳到主要內容

臺灣博碩士論文加值系統

(3.231.230.177) 您好!臺灣時間:2021/08/02 10:23
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:黃大任
研究生(外文):Ta-Jen Huang
論文名稱:自組非線性系統應用於波高預測
論文名稱(外文):Study on Wave Height Forecasting by Using Self-Organization Algorithm Model
指導教授:顏沛華顏沛華引用關係
指導教授(外文):Pei-Hwa Yen
學位類別:碩士
校院名稱:國立成功大學
系所名稱:水利及海洋工程學系專班
學門:工程學門
學類:河海工程學類
論文種類:學術論文
論文出版年:2008
畢業學年度:96
語文別:中文
論文頁數:79
中文關鍵詞:試誤法迴歸平方和誤差平方和波高
外文關鍵詞:trial and errorsum of squares due to regressionsum of squares errorwave height
相關次數:
  • 被引用被引用:5
  • 點閱點閱:100
  • 評分評分:
  • 下載下載:14
  • 收藏至我的研究室書目清單書目收藏:0
波浪為海面上最主要天然作用力之一,台灣四周環海,波浪對人們的海上活動影響甚巨,不論是在航行運輸、海岸工程、休閒活動、防災救難及海上軍事操演等,都會受到波浪作用所左右,若能對海面的波高作出合理的預測,則對人們從事海上活動將有莫大助益。
本研究分別以大鵬灣及小琉球氣象浮標水文資料來建立波高預測模式,輸入變數是以浮標本身觀測波高為主體,再配合可能影響波高之其他相關水文資料為輸入變數,如週期、風速及風向等,利用GMDH(Group Method of Data Handling)理論所特有之自組多層演算方法,建立輸入∼輸出關係之「波高預測模式」進行波高預測。首先係以試誤法(Trial and Error)決定出最佳的建模筆數及輸入變數,並以逐步迴歸之GMDH(SGMDH)來改善GMDH演算法會衍生高階非線性項而降低其實用性的缺點,再經由模式之先期模擬檢定作稽延時間之修正,即能架構最佳之「波高預測模式」。另本研究嘗試將即時觀測的新資料再納入模式中重新建模,利用遞迴結構對模式參數作即時修正,使模式具時變性而能自我調整,以達長期觀測、精確預測的目標。
應用本文模式推算結果的波高變化趨勢與浮標實測值一致,由各季觀測值與預測值之誤差比較,得知大鵬灣預測結果最大誤差21.66 ~3.89 cm,最小誤差0.19 ~ 0.01 cm,平均誤差僅4.96 ~ 1.22cm;小琉球預測結果最大誤差12.46 ~ 1.23cm,最小誤差0.10 ~ 0.01cm,平均誤差亦僅3.13 ~ 0.74cm。故本文所建立的「波高預測模式」符合要求,可實際應用於海面波浪波高之預測推算。
Wave is one of the natural forces and will affect human’s activities in the ocean such as sailing, leisurely movements, offshore engineering, disaster rescuing and military training. Reasonable accuracy wave forecasting data should do great help for all kind of marine activities.
The hydrological data of wave height as well as period, wind speed and direction collected by the data buoy in Dapeng Bay and Little-Liouciou are used to be input variables to set up the GMDH wave height forecasting model. Trail and error procedure provided first to determine input variables and sums for the best model fitting. The SGMDH process then introduced to improve the ill condition of high rank nonlinear terms which is derived through the self-organizing multilayer algorithm of GMDH structure. The optimum wave height forecasting model could be constructed after the calibrate step to revised the time lag between simulated and measured results. In addition, model parameters could be modified by the stepwise regression algorithm using the update data and threshold to achieve the purpose of self-regulating, long term and time variable wave height accurate forecasting.
The trend of forecasting wave height series were in agreement with measuring materials of the data buoy in this research. The maximum and minimum error between forecasting and measuring data of Dapeng Bay data buoy are 21.66~3.89cm and 0.19~0.01cm respectively and have an average error of 4.96~1.22cm. As to Little-Liouciou data buoy, its maximum and minimum error are 12.46~1.23cm and 0.10~0.01cm respectively and get an average error of 3.13~0.74cm. So, the GMDH forecasting model proposed by this paper is in conformity with the requirement of forecasting and could be applied in practical usage for wave height prediction at the specified location.
中文摘要····················································Ⅰ
英文摘要····················································Ⅱ
謝 誌·······················································Ⅲ
目 錄·······················································Ⅳ
表目錄······················································Ⅶ
圖目錄······················································Ⅷ
照片目錄····················································Ⅹ
第一章 緒 論·················································1
1-1 研究動機···············································1
1-2 研究方法···············································3
1-3 相關文獻回顧···········································4
1-4 本文組織···············································7
第二章 GMDH理論 ·············································8
2-1 基本架構···············································8
2-2 基本理論···············································8
2-2-1 GMDH演算法···········································9
2-2-2 逐步迴歸之GMDH演算(SGMDH)···························14
2-2-3 遞迴結構之GMDH演算··································17
第三章 波浪預測模式之建立與驗證 ····························21
3-1 波浪預測在海上航行安全之重要性 ·······················21
3-2 波浪預測模式之建立 ···································26
3-3 波浪預測效能評鑑指標 ·································28
3-4 波浪預測模式之驗證(實例驗證) ·························29
3-4-1 大鵬灣及小琉球附近海域地形及氣海象概述 ············29
3-4-1-1 沿海海底地形 ···································29
3-4-1-2 氣海象狀況 ·····································30
3-4-2 資料概述············································31
3-4-3 資料浮標系統········································32
3-4-4 波浪預測模式參數之選定······························35
3-4-5 波浪預測模式之建立與預測····························35
3-4-6 資料更新重新建模之預測······························39
第四章 分析結果與討論 ······································40
4-1 GMDH及SGMDH之模擬結果與比較 ··························40
4-2 最佳輸入變數組合 ·····································41
4-3 波浪預測結果分析 ·····································42
4-4 以新數據重新建模之預測結果比較 ·······················56
4-5 波高預測應用探討 ·····································59
4-5-1 波浪變形原理 ·······································59
4-5-2 折射圖繪製方法 ·····································63
4-5-3 計算結果 ···········································68
第五章 結論與建議 ··········································74
5-1 結 論 ················································74
5-2 建 議 ················································75
參考文獻 ···················································77
1.顏月珠,「現在統計學」,三民書局。
2.王如意、易任,「應用水文學」,新編上下冊,國立編譯館出版。
3.海岸工程規劃,農復會特刊,1971。
4.陳陽益,「波浪通用模式與最大波之解析」,國立成功大學台南水工試驗所研究論文,1983。
5.湯麟武、歐善惠、莊甲子,「合成波浪及其波高分佈預報法之研究」,國立成功大學台南水工試驗所研究論文,1985。
6.林朝福,「波浪方向譜變形研究及在多柱結構分析之應用」,國立成功大學台南水工試驗所研究論文,1986。
7.湯麟武、黃煌煇、林西川,「非線性重力波之數值解析」,國立成功大學台南水工試驗所研究論文,1988。
8.高家俊、林志明,「近岸地區波浪特性之研究」,國立成功大學水利暨海洋工程研究所研究論文,1991。
9.李梅芳,「自組非線性系統應用於水位相關之研究」,碩士論文,國立成功大學水利暨海洋工程研究所,1993。
10.楊永祺,「自組非線性系統應用於水位即時預測」,碩士論文,國立成功大學水利暨海洋工程研究所,1997。
11.周益達,「自組非線性理論建立水位預測模式之探討」,碩士論文,國立成功大學水利暨海洋工程研究所,1998。
12.李文揚,「自組非線性系統應用於水庫優養化之預測」,碩士論文,國立成功大學水利暨海洋工程研究所,2000。
13.吳俊逸,「自組非線性系統應用於地層下陷之預測」,碩士論文,國立成功大學水利暨海洋工程研究所,2001。
14.郭一羽,「海岸工程學」,文山書局,2003。
15.海難災害防救業務計畫,交通部,2004。
16.陳亞嵐,「近岸風浪推算資料同化之研究」,碩士論文,國立成功大學水利暨海洋工程研究所,2004。
17.王英銘,「自組非線性系統應用於濁度預測」,碩士論文,國立成功大學水利暨海洋工程研究所,2005。
18.郭金棟,「海岸工程」,中國土木水利工程學會,2005。
19.江俊儒,「海洋波譜特性及其於分析波浪之應用」,碩士論文,國立成功大學水利暨海洋工程研究所,2007。
20.http://w3.oc.ntu.edu.tw/index.html,海洋學教材。
21.Bowditch (1958) "American Practical Navigator", H.O. Pub. No.9, pp774-775。
22.Stanley J. Farlow,「Self-organizing Methods In Model-GMDH TYPE Algorithms」,MARCEL DEKKER.INC。
23.Duffy,J.J. and M.A.Franklin,「A Learning Identification Algorithm and Its Application to an Environmental Systems」,IEEE Trans.System,Man,Cybernetic,Vol.SMC-5,NO.2,pp.226-239,1975。
連結至畢業學校之論文網頁點我開啟連結
註: 此連結為研究生畢業學校所提供,不一定有電子全文可供下載,若連結有誤,請點選上方之〝勘誤回報〞功能,我們會盡快修正,謝謝!
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊