跳到主要內容

臺灣博碩士論文加值系統

(35.172.136.29) 您好!臺灣時間:2021/07/29 08:44
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

我願授權國圖
: 
twitterline
研究生:許峰源
研究生(外文):Feng-Yuan Hsu
論文名稱:支援向量機於降雨逕流預報之研究
論文名稱(外文):Support Vector Machines for rainfall-runoff forecasting
指導教授:林國峰林國峰引用關係
指導教授(外文):Gwo-Fong Lin
學位類別:碩士
校院名稱:國立臺灣大學
系所名稱:土木工程學研究所
學門:工程學門
學類:土木工程學類
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:66
中文關鍵詞:類神經網路支援向量機倒傳遞類神經網路強健性降雨逕流預報
外文關鍵詞:neural networkssupport vector machinesback-propagation networksrobustnessrainfall-runoff forecasting
相關次數:
  • 被引用被引用:6
  • 點閱點閱:133
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
降雨逕流的預報不論是在颱洪時期的防洪規劃上或是平時的水資源規劃上,都是不可或缺的部分。因此,如何得到一個準確且可靠的預報結果是相當重要的課題。在模式的準確度上已有許多人研究出許多改善的模式,而預報結果的可靠度則是較少人探討的問題。若想得到一個可靠的預報結果,就需要一個強健性高的預報模式。而在近幾年研究中,顯示有一種稱為「支援向量機」的類神經網路能取代常用的倒傳遞類神經網路,主要的原因除了在準確度上有所提昇之外,在強健性上也較佳。然而,研究中雖然有提出支援向量機具有較佳的強健性,但於強健性方面的討論卻很少,基於此,本研究將分別以支援向量機及倒傳遞類神經網路架構流量預報模式,並針對模式強健性的部份,分成訓練資料量、訓練場次挑選、訓練資料內噪音及初始權重這四個可能影響模式表現的因素進行討論。結果顯示,支援向量機的預報結果不但準確度優於倒傳遞類神經網路,且四個因素對支援向量機的影響更遠小於對倒傳遞類神經網路的影響。這表示支援向量機的準確度及強健性均高於倒傳遞類神經網路,亦即支援向量機的預報成果是更為有效且可靠的。因此,本研究建議在架構降雨逕流模式時,採用支援向量機取代傳統的倒傳遞類神經網路。
To yield reliable forecasts of stream flow, a robust flood forecasting model is required. For this purpose, effective flood forecasting models based on the support vector machine (SVM), which is a novel kind of neural networks (NNs), are proposed. Based on statistical learning, SVMs have better generalization ability than back-propagation netwoks (BPNs), which are the most frequently used convectional NNs. In addition, the robustness of SVMs is one of the major advantages over BPNs. However, the robustness of hydrological models has received little attention in literature. To make comparisons between SVMs and BPNs, two kinds of NN-based (SVM-based and BPN-based) forecasting models are constructed to yield one- to five-hour ahead forecasts. Then an application is conducted to clearly demonstrate the advantages of SVMs and representative results are discussed in depth. Firstly, the results show that SVM-based models perform better than BPN-based models for one- to five-hour ahead forecasts. Thus, SVM-based models forecast stream flow more accurately. In addition, the results indicate that the performance of BPN-based models highly depend on four factors: (a) number of training events, (b) selection of training events, (c) noise included in training data and (d) initial weights. On the contrary, the influence of the four factors on the performance of SVM-based models is much less than that of BPN-based models. Hence the performance of SVM-based models is more reliable than that of BPN-based models. In conclusion, the SVM-based models are much more accurate and robust than BPN-based models. The proposed SVM-based models are recommended as an alternative to the existing models because of their accuracy and robustness.
摘要 I
Abstract II
目錄 III
圖目錄 V
表目錄 VII
第一章 緒論 1
1-1 前言 1
1-2 文獻回顧 2
第二章 研究區域與水文資料 5
2-1 研究區域 5
2-2 水文資料概述 6
第三章 理論模式與架構 7
3-1 倒傳遞類神經網路 7
3-2 支援向量機 8
3-3 模式架構與參數設定 11
3-4 評鑑指標 12
第四章 結果與討論 14
4-1 訓練資料量的影響 14
4-2 訓練資料挑選的影響 15
4-3 資料噪音的影響 16
4-4 初始權重的影響 17
第五章 結論 18
參考文獻 19
Bae, D.H., Jeong D.M., Kim, G, 2007, “Monthly Dam Inflow Forecasts Using Weather Forecasting Information and Neuro-Fuzzy Technique,” Hydrological Sciences Journal 52 (1), 99-113.
Choy, K.Y., Chan, C.W., 2003, “Modeling of River Discharges and Rainfall Using Radial Basis Function Networks Based on Support Vector Regression,” International Journal of Systems Science, 34(14-15), 763-773.
Chang, F.J., Chang, Y.T., 2006, “Adaptive Neuro-Fuzzy Inference System for Prediction of Water Level In Reservoir,” Advances in Water Resources. 29(1), 1-10.
Chen S.T., Yu P.S., 2007, “Pruning of Support Vector Networks on Flood Forecasting,” Journal of Hydrology, 347(1-2), 67-78.
Lin, G.F., Chen, L.H., 2005, “Application of An Artificial Neural Network to Typhoon Rainfall Forecasting,” Hydrological Processes,” 19(9), 1825-1837.
Liong, S. Y., Sivapragasam, C., 2002, “Flood Stage Forecasting With Support Vector Machines,” Journal of the American Water Resources Association. 38(1), 173-186.
Liong, S. Y., Sivapragasam, C., 2005, “Flow Categorization Model for Improving Forecasting,” Nordic Hydrology. 36(1), 37-48.

Ioannis N. D., Paulin C., Ioannis K.T., 2005, “Groundwater Level Forecasting Using Artificial Neural Networks,” Journal of Hydrology, 309, 229-240.
Mahesh Pal, Arun Goel,2006, “Estimation of Discharge and End Depth in Trapezoidal Channel by Support Vector Machines,” Water Resource Management, 21, pp.1762-1780。
Shivam Tripathi, V.V. Srinivas, Ravi S. Nanjundiah, Downscaling of precipitation for climate change scenarios:A support vector machine approach, Journal of Hydrology, 330, pp.621-640.
Vapnik, V., 1995, “The Nature of Statistical Learning Theory,” Sprinter, New York.
Vapnik, V., 1998, “Statistical Learing Theory.” John Wiley, New York.
Xu, Z.X., Li, J.Y., 2002, “Short-Term Inflow Forecasting Using An Artificial Neural Network Model,” Hydrological Processes, 16, 2423-2439.
Yu, P.S., Chen, S.T., 2005, “Updating Real-Time Flood Forecasting Using a Fuzzy Rule-Based Model,” Hydrological Sciences Journal, 50(2), 265-278.
Yu, X.Y., Liong, S. Y., 2007, “Forecasting of Hydrologic Time Series with Ridge Regression in Feature Space,” Journal of Hydrology. 332, 290-302.
Yu, X.Y., Liong, S.Y., Babovic, V., 2004, “EC-SVM Approach for Real-Time Hydrologic 487 forecasting,” Journal of Hydroinformatics 6(3), 209-223.

陳憲忠,2006,「支援向量機及模糊推理模式應用於洪水水位之即時預報」,國立成功大學水利及海洋工程學系博士論文。
陳信中,2006,「蘭陽溪洪水預報模式之研究」,國立台灣大學生物環境系統工程學研究所碩士論文。
陳憲忠、游保杉,2007,「洪水位之即時機率預報-結合支援向量機與模糊理論」,農業工程學報,第五十三卷,第四期,第1-20頁。
張逸凡,2005,「支援向量機在即時河川水位預報之應用」,國立成功大學水利及海洋工程學系碩士論文。
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊