Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/15648
標題: 以類神經網路預測淡水河口暴潮偏差之研究
A Study on Storm-Surge Prediction at Tanshui Estuary by Artificial Neural Network
作者: 游智宇
You, Chih-Yu
關鍵字: Storm-Surge
暴潮偏差
Multi Layer Perception Network
Radial Basis Function Network
多層感知器類神經網路
徑向基底函數類神經網路
出版社: 土木工程學系所
引用: 1. Azmathullah H. Md., Deo M.C. and Deolalikar P.B. (2005) “Neural network for estimation of scour downstream of a ski-jump bucket,” Journal of Hydraulic Engineering., ASCE, Vol. 131, No. 10, pp.898-908. 2. Blain, C.A., Westerink, J.J. and Luettich, R.A., Jr. (1994) “The influence of domain size on the response characteristics of a hurricane storm surge model,” Journal of Geophys Research, Vol. 99, pp. 18467-18479. 3. Conner, W.C. et al. (1957) “Empirical methods for forecasting the maximum storm tide due to hurricanes and other tropical storms,” Monthly Weather Rev, Vol. 85, pp. 113-116. 4. Chang, F.J., Chen, Y.C. (2003) “Estuary water-stage forecasting by using radial basis function neural network,” Journal of Hydrology, Vol. 207, pp. 158–166. 5. Deo, M.C. and Sridhar, N.C. (1999) “Real time wave forecasting using neural networks,” Ocean Engineering, Vol.26, pp. 191-203. 6. Grubert, J.P. (1995) “Prediction of estuarine instabilities with artificial neural ,” J.Comput,Civ.Eng., Vol. 9, No. 4, pp. 266-274. 7. Horikawa (1978) Coastal Engineering, Tokyo University Press, pp. 153-167. 8. Hopfield, J. (1982) “Neural networks and physical systems with emergent collective computational properties”, Proceedings of the National Academy of Sciences of the USA, Vol.79, pp.2554 - 2588. 9. Harper, B.A. and Sobey, R.J. (1983) “Open-boundary conditions for open-coast hurricane storm surge,” Coastal Engineering in Japan, Vol. 7, pp.41-60. 10. Hopfield, J.J. and Tank, D.W. (1985) “Neural computation of decisions in optimization problems,” Biological Cybernetics, Vol.52, pp.141 - 152. 11. Hwang, R.R. and Yao, C.C. (1987) “A semi-implicit numerical model for storm surges,” Journal of the Chinese Institute of Engineers, Vol. 10, No. 5, pp. 463-472. 12. Haykin, S. (1994) “Neural Network : a comprehensive foundation,” Mac Millan, New York. 13. Hsu, T.W., Liao, J.M. and Lee, Z.S. (1999) “Prediction of storm surge at northestern coast of Taiwan by finite element method,” Chinese Institute of Civil and Hydraulic Engineering, Vol. 11, No. 4, pp. 849-857. 14. Jelesnianski, C. P., Chen, J. and Shaffer, W. A. (1992), SLOSH (Sea, Lake, and Overland Surges from Hurricanes.), NOAA Technical Report NWS 48. 15. Kawahara, M., Nakazawa, S., Ohmori, S. and Tagaki, T., (1980) “Two-step explicit finite element method for storm surge propagation analysis,” International Journal for Numerical Methods in Engineering, Vol. 15, pp. 1129-1148. 16. Lee, T.L. (2004) “Back-propagation neural network for long-term tidal predictions,” Ocean Engineering, Vol. 31, pp. 225-238. 17. Murty, T.S. (1984) “Storm surges: Meteorological ocean tides,” Can. Bull. of Fisheries and Aquatic Sci., No. 212, Dept. Fisheries and Oc., Canada. 18. Marzenna, S. (2003) “Forecast of storm surge by means of artificial neural network,” Journal of Sea Research, Vol.49, pp. 317-322. 19. Rumelhart D.E., Hinton G.E. and R.J. (1986) “Williams learning representations by back-propagating errors,” Nature 323, pp.533-536. 20. Serre S. (2003) “Tidal-level forecasting and filtering by neural network model,” Costal Engineering Journal, Vol. 45, pp.119-137. 21. Tsai, C.P., Lee, T.L. and Chu, L.H. (1999) “Forecasting of wave time series using back propagation neural network,” Journal of the Chinese Institute of Civil and Hydraulic Engineering, Vol.11, pp.589–596. 22. Tsai, C. P., Lin, C. and Shen, J. N. (2002) “Neural network for wave forecasting among multi-stations,” Ocean Engineering, Vol. 29, pp.1683-1695. 23. Tsai, C.P., Lee, T.L., Yang, T.J., Hsu, Y.J. (2005) “Back-propagation neural networks for Prediction of Storm Surge,” Civil-comp Press. Vol.11, 589–596. 24. Tsai, C.P., Chen, H.B. (2006) “Finite volume method for storm-surge simulations,” Proc. The 6th Japan-Taiwan Joint Seminar on Natural Hazard Mitigation. 25. Unoki, S. (1966) 「高潮じわける氣壓と風の效果比較」,第13回日本海岸講演集。 26. Villiers, J. and Barnard, E. (1992) “Back-propagation neural nets with one and two hidden layers,” IEEE Transaction On Neural Network, Vol.4, pp.136-141. 27. Walton, R. and Christensen, B.A. (1980) “Friction factors in storm surges over inland areas,” Journal of the Waterway Port Coastal and Ocean division, ASCE, Vol. 106, No. WW1, pp. 261-271. 28. Westerink, J.J., Luettich, R.A., Baptista, A.M.,Scheffner, N.W. and Farrar, P. (1992) “Tide and storm surge predictions using a finite element model,” Journal of Hrdraulic Engineering, ASCE, Vol. 118, pp. 1373-1390. 29. Yen, G.T. and Chou, F.K. (1979) “Moving boundary numerical surge model, Jour. of Waterway,” Port, Coast. and Ocean Division, ASCE, Vol.105, No. WW3, pp. 247-263. 30. 徐月娟、曾淑芬、陳進益、蔡恆雄(1998),「台灣地區八十三年至八十六年颱風暴潮特性之初步研究」,交通部氣象局氣象學報第42卷第3期,第230-247頁。 31. 葉怡成(1998),「類神經網路模式應用與實作」儒林圖書有限公司。 32. 蔡瀚陞(2000),「淡水河口颱風暴潮水位之研究」,國立台灣大學土木工程學研究所碩士論文。 33. 蔡憶雯(2001),「淡水河口暴潮位及河川水位機率預報研究」,國立台灣大學土木工程學研究所。 34. 李宗霖、S. Rajasekaran、徐月娟、楊宗儒(2003),「序列學習神經網路在颱風期間之潮位預測」,第25屆海洋工程研討會論文集, 第275-279頁。 35. 郭金棟(2004),「海岸保護」,科技圖書公司,第24頁。
摘要: 目前台灣淡水河流域具有一些不利於防範洪水與暴潮之條件,如盆地低窪、過去遭遇累積地層下陷等之影響,加上颱風每年侵襲台灣北部頻繁常有洪患,由於颱風暴潮偏差其影響因素較為複雜,不容易以數值方法或經驗公式精確地預測,本文利用具有監督網路收斂最佳化優點的多層感知器類神經網路及即時演算的徑向基底函數類神經網路,建立颱風暴潮偏差之預測模式,包括最大暴潮偏差之預測及暴潮偏差時序列之預測。 最大暴潮偏差預測模式中首先以前人所提出最大暴潮偏差之經驗公式進行回歸得出相關係數僅為0.565。本文利用兩種不同類神經模式以22組颱風資料進行模擬並探討其各項因子對暴潮偏差預測之影響,最後結果以輸入測站最大氣壓差、風速即可獲得極佳的淡水河口最大暴潮偏差值預測且相關係數皆達0.9以上。 暴潮偏差時序列之預測模式在輸入即時測站氣壓差、風速及風向及前一時刻之暴潮偏差之情況下,驗證組及測試組之相關係數皆為0.85以上,結果顯示可證實時序列模式對於颱風暴潮偏差時序列的預測可達到良好的預測成果。
Taiwan northern area is always attacked by typhoon frequently every year and induces the flood disasters. At present, Tamsui river territory has some unfavorable conditions including the basin low-lying and land subsidence to control the flood with storm-surge. Thus the accurate prediction of the storm-surge is an important issue for the area. However, it is quite complex for the prediction of storm-surge and use the numerical method or empirical formula to predict the phenomenon is not easily. Alternatively This paper applies the artificial networks including the supervised multilayer perception neural network and the radial basis function neural network, for the prediction of the storm-surge . Based on the previous empirical formula of the maximum of storm-surge, it is only 0.565 to draw the correlation coefficient. This study chooses the stand atmosphere pressure variation, wind speed and wind direction parameters as the input neurons for the networks of typhoon about 22 groups and discuss the effect of each parameter on storm-surge forecast. The results agree well with the measured data of storm-surge, which all the correlation coefficient are more than 0.9. The results of the predicted and test model show that the correlation coefficient values are larger than 0.85 in the situation of predicted model inputted the atmosphere pressure variation, wind speed, wind direction and storm-surge of last moment parameters into the time series of storm-surge. This result illustrates that time series model forecast well for the storm-surge of the time during the typhoon.
URI: http://hdl.handle.net/11455/15648
其他識別: U0005-2708200722053300
文章連結: http://www.airitilibrary.com/Publication/alDetailedMesh1?DocID=U0005-2708200722053300
Appears in Collections:土木工程學系所

文件中的檔案:

取得全文請前往華藝線上圖書館



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.