進階搜尋


下載電子全文  
系統識別號 U0026-2308201815591300
論文名稱(中文) 應用資料探勘方法於瘋狗浪研究
論文名稱(英文) Application of Data Mining Methods on Coastal Freak Wave Study
校院名稱 成功大學
系所名稱(中) 水利及海洋工程學系
系所名稱(英) Department of Hydraulics & Ocean Engineering
學年度 106
學期 2
出版年 107
研究生(中文) 蘇瑋琳
研究生(英文) Wei-Lin Su
學號 N86051112
學位類別 碩士
語文別 中文
論文頁數 72頁
口試委員 指導教授-董東璟
口試委員-滕春慈
口試委員-蔡政翰
口試委員-蕭士俊
中文關鍵字 海岸瘋狗浪  資料探勘  機率預測  人工類神經網路  群集分析 
英文關鍵字 coastal freak wave  data mining  probability prediction  artificial neural networks  cluster analysis 
學科別分類
中文摘要 臺灣周遭海岸每年發生許多瘋狗浪事件,造成嚴重傷亡。海岸瘋狗浪發生原因相當複雜,至目前仍無一套理論能預測瘋狗浪何時何地會發生。本研究分別採用類神經網路及群集分析方法,從過往實際發生之落海事件建置瘋狗浪機率預警系統。採用過去瘋狗浪相關研究中得知會顯著影響瘋狗浪發生之海氣象參數,做為預警系統之輸入因子,兩種方法均使用相同的分析資料,經資料探勘方法充分訓練後,再進行驗證,結果顯示,不論類神經網路方法或群集分析方法,預警所得之正確率皆達七成以上,皆具有良好預警能力。整體而言,類神經網路系統具有較高的正確率,顯示有較好的預警能力,但群集分析系統之回應率較佳,表示預測結果可信度很高,在實際運作中誤報率較低,也具有一定的參考價值。經由敏感度測試證實,示性波高、波浪尖峰週期與班傑明非線性指數(Benjamin Feir Index , BFI)為影響瘋狗浪發生之顯著因子。本文從少數預警失敗案例發現,這些失敗案例主要集中於波高偏小的時刻,顯示預警系統仍有改善的空間;此外也發現本研究建置之群集分析系統與中央氣象局現行運作測試中的群集分析瘋狗浪預警系統比較,驗證正確率皆在同一個級距,此方法失敗的案例主要是發生在波向風向差因子主導之瘋狗浪事件,較無法準確預測。本文考慮未來預警系統作業化之穩定性,評估使用不同輸入資料建置預警系統之差異,結果顯示以實測海氣象觀測資料建置預警系統最佳。
英文摘要 There are many coastal freak wave events every year in Taiwan, which cause serious casualties. The mechanic of coastal freak wave is quite complicated. There is still no theory to predict when and where coastal freak wave will occur. In this study, data-driven warning systems based on an artificial neural network (ANN) and cluster analysis are proposed to predict the occurrence possibility of a CFW. It shows that, regardless of the neural network method or cluster analysis method, the accuracy rate of warnings is more than 70%, and all have good warning capabilities. The artificial neural network system has a higher accuracy rate and shows good early warning capability. However, the cluster analysis system has a better response rate, indicating that the prediction results are highly reliable and the false alarm rate is higher in actual operations, it also has a certain reference value. Through sensitivity analysis, it was confirmed that the significant wave height (Hm0), wave peak period (TP), and Benjamin Feir Index (BFI) were significant factors influencing the occurrence of coastal freak wave. This article further understands a few number of cases of warning failures, and finds that the warning system still has been improved when the two warning systems aim at wavelet-high coastal freak wave incidents. This study considers the stability of the operation of the warning system, evaluates the differences in the use of different input data to build an early warning system, and the results show that it is best to build an early warning system based on buoy observation data.

論文目次 摘要 I
ABSTRACT II
誌謝 IX
目錄 X
表目錄 XII
圖目錄 XIII
符號表 XV
第一章 緒論 1
1.1 研究動機及目的 1
1.2 文獻探討 2
1.3 論文架構 5
第二章 研究方法 6
2.1 類神經網路 9
2.1.1 倒傳遞類神經網路架構 10
2.1.2 倒傳遞類神經網路運作原理 12
2.1.3 倒傳遞類神經網路演算法 16
2.2 群集分析 20
2.2.1 群集分析運作原理 21
2.2.2 群集分析演算法 22
第三章 研究海域及使用資料 27
3.1 研究海域 27
3.2 使用資料 29
第四章 機率預警模式建置 34
4.1 機率預警模式建置流程 34
4.2 類神經網路模式建置 37
4.2.1 類神經網路模式參數 37
4.2.2 類神經網路模式建置與驗證結果 41
4.3 群集分析模式建置 45
4.3.1 群集分析模式參數 45
4.3.2 群集分析模式建置與驗證結果 48
4.4 兩模式比較與評估 55
第五章 結論與建議 63
參考文獻 66

參考文獻 [1] Agrawal, R., Imieliński, T., and Swami, A., “Mining association rules between sets of items in large databases,” in Proceedings of ACM SIGMOD, 207-216, 1993.
[2] Ankerst, M., Breunig, M.M., Kriegel, H-P. and Sander, J., “OPTICS: Ordering Points to Identify the Clustering Structure,” SIGMOD, 1999.
[3] Bayes, T., “An essay towards solving a problem in the doctrine of chances.”, Philos Trans Soc Lon, 53, 370-418, 1763.
[4] Benjamin, T.B. and Feir, J.E., The Disintegration of Wave Trains on Deep Water,J. Fluid Mechanics, Vol. 27, 417-430, 1967.
[5] Breiman, L., Friedman, J.H., Olshen, R.A., and Stone, C.I., Classification and regression trees. Belmont, Calif. : Wadsworth, 1984.
[6] Breiman, L., Cutler, A., Random Forests. Machine Learning, 45, 5-32, 2001.
[7] Böse, M., Wenzel, F., Erdik, M., PreSEIS: A Neural Network-Based Approach to Earthquake Early Warning for Finite Faults.Bulletin of the Seismological Society of America, 98 (1), 366 – 382, 2008.
[8] Cybenko, G., Approximation by Superpositions of a Sigmoidal Function.Mathematical Control Signal Systems,2,303-314, 1989.
[9] Chien, H., Kao, C.C., and Chuang, Z.H., On the characteristics of observation coastal freak waves, Coastal Engineering Journal, Vol. 44, No. 4, 301-319, 2002.
[10] Chang, F.J., Tseng, K.Y., Chaves, P., Shared near neighbours neural network model: a debris flow warning system. Hydrological Processes, 21, 1968 – 1976, 2007.
[11] Chi, Q., Shu, J., Zhang, X., Zhou, Q., The Application of Improved BP Neural Network for Power Load Forecasting in the Island Microgrid System. IEEE Electrical and Control Engineering, 6138 – 6141, 2011.
[12] Chen, W.B., Liu, W.C., Hsu, M.H., Predicting typhoon-induced storm surge tide with a two-dimensional hydrodynamic model and artificial neural network model. Natural Hazards and Earth System Sciences, 12, 3799 – 3809, 2012.
[13] Dempster, A., Laird, N., and Rubin, D., Maximum likelihood from incomplete data via the EM algorithm, Journal of the Royal Statistical Society, Series B, vol. 39, no.1, 1-38, 1977.
[14] Ester, M., Kriegel, H.P., Sander, H., Xu, X., A Density-Based Algorithm for Discovering Clusters in Large Spatial Datasets with Noise. Proc. 2nd Int. Conf. KDDD. Portland, Oregon, p. 1232–1239, 1996.
[15] Goda, Y., Random Seas and Design of Maritime Structures, 2nd edn., World Scientific, Singapore, p. 443, 2000.
[16] Hertz, J.,Krough, A.,Palmer, R.G., Introduction to the theory of neural computation.New York:Addison-Wesley, 1991.
[17] Hagan, M.T, Menhaj, M.B., Training feedforward networks with the Marquardt algorithm, IEEE Transactions on Neural Networks, 5(6):989-993, 1994.
[18] Hinneburg, A., Keim, D.A., An efficient approach to clustering in large multimedia databases with noise. Proc. of the KDD, 98:58−65, 1998.
[19] Janssen, P. A., Nonlinear four-wave interactions and freak waves. Journal of Physical Oceanography, 33(4), 863-884, 2003.
[20] Komen, G.J., Cavaleri, L., Donelan, M., Hasselmann, K., Hasselmann, S., Janssen (Eds.), P.A.E.M., Dynamics and Modelling of Ocean Waves, Cambridge, 1994.
[21] Kharif, C., Pelinovsky, E., and Slunyaev, A.: Rogue Waves in the Ocean, Springer-Verlag, 2009.
[22] Kung, H.Y., Chen, C.H., Ku, H.H., Designing intelligent disaster prediction models and systems for debris-flow disasters in Taiwan. Experts Systems with Application, 39, 5838 –5856, 2012.
[23] Longuet-Higgins, M., and Drazen, D., On Steep Gravity Waves Meeting a Vertical Wall: a Triple Instability, Journal of Fluid Mechanics, Vol. 466, pp. 305-318, 2002.
[24] López, V.F., Medina, S.L., Paz, J.F.d., Taranis: Neural networks and intelligent agents in the early warning against floods. Experts Systems with Application, 39, 10031 – 10037, 2012.
[25] Morgan, J. N., Sonquist, J. A., Problems in the analysis of survey data and a proposal, Journal of the American Statistical Association, 58, 415-734, 1963.
[26] MacQueen, J.B., Some methods for classification and analysis of multivariate observations, Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, 1, 281-297, 1967.
[27] Mase, H., Yasuda, T., Mori, N., Real-Time Prediction of Tsunami Magnitudes in Osaka Bay, Japan, Using an Artificial Neural Network, Journal of Water, Port, 137 (5), 263 – 268, 2011.
[28] Quinlan, J. R., Induction of Decision Trees, Machine Learning, (1), 81-106, 1986.
[29] Rumelhart, D.E., Hinton, G.E., Williams, R.J., Learning internal representations by back-propagating errors. Nature, 323, 533 – 536, 1986.
[30] Romano, M., Liong, S.Y., Vu, M.T., Zemskyy, P., Doan, C.D., Dao, M. H., Tkalich, P., Artificial neural network for tsunami forecasting. Journal of Asian Earth Sciences, 36, 29 – 37, 2009.
[31] Sunkpho, J., Otamakorn, C., Real-time flood monitoring and warning system. Journal of Science and Technology, 33(2), 227 – 235, 2011.
[32] Thirumalaiah, K., Deo, M.C., Real-Time Flood Forecasting Using Neural Networks. Computer-Aided Civil and Infrastructure Engineering, 13, 101 – 111, 1998.
[33] Tsai, C.H., Su, M.Y., and Huang, S.J., Observations and Conditions for Occurrence of Dangerous Coastal Waves, Ocean Engineering, Vol. 31, 745-760, 2004.
[34] Tamura, H., Waseda, T., Miyazawa, Y., Freakish sea state and well-windsea coupling: Numerical study of the Suwa-Maru incident. Geographical Research Letter, Vol.36, L01607, 2009.
[35] Vapnik, V. N., The Nature of Statistical Learning Theory, Springer-Verlag, NY, USA, 1995.
[36] Ward, J.H., Hierarchical grouping to optimize an objective function, Journal of the American Statistical Association, 58, 236-244, 1963.
[37] 尹相志,SQL Server 2008 Data Mining,悅知文化,2010。
[38] 王進德、蕭大全,類神經網路與模糊控制理論入門,全華科技圖書有限公司,2005。
[39] 朱執均,類神經網路之應用-南中國海海域潮汐預報及補遺,國立中山大學碩士論文,2000。
[40] 李賢文,「瘋狗浪」之初步研究,中央氣象局天氣分析與預報研討會,台北,第471-475頁,1998。
[41] 李宗霖、宋憶萍、黃奇鍊、黃清哲,類神經網路預測2020年台灣海域海平面變遷之研究,第34屆海洋工程研討會論文集,國立成功大學,第529-534頁,2012。
[42] 陳正昌、程炳林、陳新豐、劉子鍵,「多變量分析方法」,五南圖書出版股份有限公司,1994。
[43] 陳順宇,「多變量分析」,華泰書局,2000。
[44] 陳冠宇,瘋狗浪的可能機制,海洋工程學刊,第二卷,第一期,第93-106頁,2002。
[45] 晉鳳山,「改良式倒傳遞類神經網路之研究」,測驗統計簡訊,2-27頁,2002。
[46] 許明光、曾俊超、高家俊,臺灣地區「瘋狗浪」之調查及成因初探,第十五屆海洋工程研討會論文集,第513-524頁,1993。
[47] 張憲國、錢維安、何良勝,「應用類神經網路在台灣東岸海域颱風波浪推算之研究」,海洋工程學刊,第3卷,第1期,第73-95頁,2003。
[48] 張振洋、蔡政翰,堤前異常波浪與三連波的觀測,第27屆海洋工程研討會,pp. 147-153,2005。
[49] 張憲國、林立青、何良勝、李明璋,「花蓮港船隻預警神經網路模式之研究」,第28屆海洋工程研討會論文集,第923-928頁,2006。
[50] 張斐章,江衍銘,「以類神經網路建構定量降雨及多階段洪水預報模式」,台灣水利,第五十五卷,第二期,第25-33頁,2007。
[51] 董東璟,異常海象機率預警研究與作業試用(1/4)。交通部中央氣象局委託研究計畫(期末)成果報告,2016。
[52] 詹錢登、曾志民、王志賢、王啟明,類神經網路應用於颱風暴潮之預測,海洋工程學刊,6(1),1-24,2006。
[53] 葉怡成,應用類神經網路,儒林圖書公司,2001。
[54] 蔡政翰、林盈成、曾相茂,群波與瘋狗浪,海洋工程學刊,第一卷,第一期,第71-82頁,2001。
[55] 蔡仁智、蔡政翰,資料探勘應用於台灣海岸危險波浪發生之研究,第29 屆海洋工程研討會論文集,第541-546頁,2007。
[56] 蔡仁智,資料探勘應用於波浪測量與近岸危險波浪預警模式,國立臺灣海洋大學海洋環境資訊系博士論文,2010。
[57] 劉俊志,近岸突浪的分析探討,國立中山大學碩士論文,2006。
[58] 賴瑩嬛,應用群集分析探討侵臺颱風之分類特性,國立交通大學碩士論文,2002。
[59] 謝邦昌、邱志洲,類神經網路分析(Neural Network Analysis),曉園出版社,2000。
[60] 謝帆廸、蔡仁智、蕭松山、方惠民、蔡政翰,以類神經網路探勘技術建立海灘侵淤量變化模式,第37屆海洋工程研討會論文集,國立中興大學,第331-336頁,2015。
[61] 謝邦昌,SQL Server資料採礦與商業智慧(第二版)--適用SQL Server 2016/2014,2017。
論文全文使用權限
  • 同意授權校內瀏覽/列印電子全文服務,於2020-08-23起公開。
  • 同意授權校外瀏覽/列印電子全文服務,於2020-08-23起公開。


  • 如您有疑問,請聯絡圖書館
    聯絡電話:(06)2757575#65773
    聯絡E-mail:etds@email.ncku.edu.tw