臺灣博碩士論文加值系統

颱風襲臺常造成淹水的災害，過去利用二維淹水模式產生淹水資料結果相當費時，無法在颱風期間及時演算，本文首先利用SOBEK產生大量淹水資料當作訓練資料，結合k-means聚類法和新型類神經網路-支援向量機(support vector machines, SVM)發展一套颱風淹水預警系統。
主要架構分成三部分：分類、預報以及空間推估，首先先區分出淹水區，在將淹水區的資料利用k-means聚類法以不同的淹水歷線型態進行分類，根據地理空間特性找尋鄰近的淹水監測站作為控制點。接著，在每個控制點建構預報模式，利用降雨量和水位兩個因子作為SVM預報模式的輸入項，預報控制點未來1至3小時的水位，將水位轉換成水深後，接著將各控制點預報的水深、二度分帶座標(X, Y)、雨量、9個淹水影響因子，分別是高程、坡度、坡向、總曲率、平面曲率、剖面曲率、與河川的距離、地形濕度指數(Topographic wetness index, TWI)、逕流強度指數(Stream power index, SPI)當作輸入項，利用SVM空間推估模式，即可推估未來1至3小時淹水區網格點的淹水深度。
本研究以宜蘭縣的宜蘭河流域與美福大排來驗證所提出的方法，結果顯示此方法能夠準確的預報未來1至3小時的淹水深度，以地理資訊系統(geographic information system, GIS)繪製各網格點的預報淹水深度，預報結果的淹水深度圖能夠反應出所收集到的淹水潛勢圖資料。

口試委員會審定書 #
誌謝 i
中文摘要 ii
ABSTRACT iii
目錄 iv
圖目錄 vii
表目錄 xii
第1章 緒論 1
1.1 研究動機 1
1.2 文獻回顧 3
1.3 研究目的 6
1.4 論文架構 7
第2章 理論與方法 9
2.1 聚類演算法 9
2.1.1 k-means聚類法 9
2.2 SOBEK模式 11
2.3 SVM-支援向量機 12
第3章 研究區域與資料蒐集 16
3.1 研究區域 16
3.2 淹水資料蒐集 17
3.3 淹水影響因子資料 23
第4章 模式建立與應用 31
4.1 颱風淹水預警系統 31
4.1.1 SOBEK模式建立 32
4.1.2 k-means分類 34
4.1.3 控制點預報模式 36
4.1.4 空間推估模式 37
4.2 交替驗證與評鑑指標 39
4.2.1 交替驗證 39
4.2.2 評鑑指標 39
第5章 結果與討論 40
5.1 SOBEK模式檢定驗證結果 40
5.2 k-means分類結果 50
5.3 控制點預報結果 57
5.4 空間推估結果 67
第6章 結論與建議 93
6.1 結論 93
6.2 建議 94
參考文獻 95
附錄 99

[1]Adnan, R., Ruslan, F. A., Samad, A. M. and Zain, Z. M. (2013). New Artificial Neural Network and Extended Kalman Filter hybrid model of flood prediction system. In Signal Processing and its Applications (CSPA), pp. 252-257.
[2]Chang, L. C., Chang, F. J. and Wang, Y. P. (2009). Auto‐configuring radial basis function networks for chaotic time series and flood forecasting. Hydrological processes, 23(17), pp.2450-2459.
[3]Chang, L. C., Shen, H. Y. and Chang, F. J. (2014). Regional flood inundation nowcast using hybrid SOM and dynamic neural networks. Journal of hydrology, 519, pp.476-489.
[4]Chang, L. C., Shen, H. Y., Wang, Y. F., Huang, J. Y. and Lin, Y. T. (2010). Clustering-based hybrid inundation model for forecasting flood inundation depths. Journal of hydrology, 385(1-4), pp.257-268.
[5]Chen, C. S., Chen, B. P. T., Chou, F. N. F. and Yang, C. C. (2010). Development and application of a decision group Back-Propagation Neural Network for flood forecasting. Journal of hydrology, 385(1-4), pp.173-182.
[6]Chidthong, Y., Tanaka, H. and Supharatid, S. (2009). Developing a hybrid multi‐model for peak flood forecasting. Hydrological processes, 23(12), pp.1725-1738.
[7]Chua, L. H. and Wong, T. S. (2011). Runoff forecasting for an asphalt plane by Artificial Neural Networks and comparisons with kinematic wave and autoregressive moving average models. Journal of Hydrology, 397(3-4), pp.191-201.
[8]Cristianini, N. and Shaw-Taylor, J. (2000). An introduction to support vector machines and other kernel-based learning methods. Cambridge University Press, New York
[9]Cunge, J.A.(1980). Two-dimension ,modeling of flood plain. chap.17 of unateady flow in open channel, Water Resources Limited, London
[10]Doong, D. J., Lo, W., Vojinovic, Z., Lee, W. L. and Lee, S. P. (2016). Development of a New Generation of Flood Inundation Maps— A Case Study of the Coastal City of Tainan, Taiwan. Water, 521
[11]Hong, H., Tsangaratos, P., Ilia, I., Liu, J., Zhu, A. X. and Chen, W. (2018). Application of fuzzy weight of evidence and data mining techniques in construction of flood susceptibility map of Poyang County, China. Science of The Total Environment, 625, pp.575-588.
[12]Hsu, M.H., Chen, S.H. and Chang, T.J. (2002). Dynamic inundation simulation of storm water interaction between sewer system and overland flows. Journal of the Chinese Institute of Engineers, 25(2), pp.171-177
[13]Jardine N. and Sibson R. (1968). The construction of hierarchic and non-hierarchic classifications. Computer Journal, 11, pp. 177-184
[14]Jha, A., Lamond, J., Bloch, R., Bhattacharya, N., Lopez, A., Papachristodoulou, N., Bird, A., Proverbs, D., Davies, J. and Barker, R. (2011). Five Feet High and Rising - Cities and Flooding in the 21st Century. Policy Research Working Paper 5648. Washington: The World Bank
[15]Kuntiyawichai, K., Schultz, B., Uhlenbrook, S., Suryadi, F. X. and Van Griensven, A. (2011). Comparison of flood management options for the Yang River Basin, Thailand. Irrigation and drainage, 60(4), pp.526-543.
[16]Lai, C., Shao, Q., Chen, X., Wang, Z., Zhou, X., Yang, B. and Zhang, L. (2016). Flood risk zoning using a rule mining based on ant colony algorithm. Journal of hydrology, 542, pp.268-280.
[17]Lin, G. F. and Chen, L. H. (2004). A non-linear rainfall-runoff model using radial basis function network. Journal of Hydrology, 289(1-4), pp.1-8.
[18]Lin, G. F., Chen, G. R., Huang, P. Y. and Chou, Y. C. (2009). Support vector machine-based models for hourly reservoir inflow forecasting during typhoon-warning periods. Journal of Hydrology, 372(1-4), pp.17-29.
[19]Lin, G. F., Lin, H. Y., and Chou, Y. C. (2013). Development of a real-time regional-inundation forecasting model for the inundation warning system. Journal of Hydroinformatics, 15(4), pp.1391-1407.
[20]Liu, C. L., Hsaio, W. H. and Chang, T. H. (2018). Locality Sensitive K-means Clustering. Journal of Information Science and Engineering, 34(1), pp.289-305.
[21]MacQueen J.B. (1967). Some Methods for classification and Analysis of Multivariate Observations. Proceedings of 5-th Berkeley Symposium on Mathematical Statistics and Probability, Berkeley, University of California Press, pp. 281-297
[22]Nandi, A., Mandal, A., Wilson, M., & Smith, D. (2016). Flood hazard mapping in Jamaica using principal component analysis and logistic regression. Environmental Earth Sciences, 75(6), 465.
[23]Neshatpour, K., Malik, M., Sasan, A., Rafatirad, S., Mohsenin, T., Ghasemzadeh, H. and Homayoun, H. (2018). Energy-efficient acceleration of MapReduce applications using FPGAs. Journal of Parallel and Distributed Computing, 119, pp.1-17.
[24]Pan, T. Y., Lai, J. S., Chang, T. J., Chang, H. K., Chang, K. C. and Tan, Y. C. (2011). Hybrid neural networks in rainfall-inundation forecasting based on a synthetic potential inundation database. Natural Hazards and Earth System Sciences, 11(3), 771.
[25]Popescu, I., Jonoski, A., Van Andel, S. J., Onyari, E. and Moya Quiroga, V. G. (2010). Integrated modelling for flood risk mitigation in Romania: case study of the Timis–Bega river basin. International Journal of River Basin Management, 8(3-4), pp.269-280
[26]Prinsen, G.F. and Becker, B.P.J. (2011). Application of SOBEK hydraulic surface water models in the Netherlands Hydrological Modelling Instrument. Irrigation and Drainage, 60, pp. 35-41
[27]Rigby, E and Van Drie, R. (2008). ANUGA: A New Free and Open Source Hydrodynamic Model. Proceedings of Water Down Under, pp. 629-638.
[28]Risi, R.D., Paola, F.D., Turpie, J. and Kroeger, T. (2018). Life Cycle Cost and Return on Investment as complementary decision variables for urban flood risk management in developing countries. International Journal of Disaster Risk Reduction, 28, pp. 88-106.
[29]Tehrany, M. S., Pradhan, B., Mansor, S., & Ahmad, N. (2015). Flood susceptibility assessment using GIS-based support vector machine model with different kernel types. Catena, 125, pp. 91-101
[30]Tehrany, M.S., Shabani, F., Neamah Jebur, M., Hong, H., Chen, W., & Xie, X. (2017). GIS-based spatial prediction of flood prone areas using standalone frequency ratio, logistic regression, weight of evidence and their ensemble techniques. Geomatics, Natural Hazards and Risk, 8(2), pp.1538-1561.
[31]Vapnik, V. (2013). The nature of statistical learning theory. Springer science and business media.
[32]Verwey, A., Muttil, N., Liong, S. Y. and He, S. (2008). Implementing an Urban Rainfall-runoff Concept in SOBEK for a Catchment in Singapore. Proceedings of Water Down Under 2008, 36
[33]Zhang T., Ramakrishnan R., and Livny M. (1996). BIRCH: An efficient data clustering method for very large databases. Proceedings of ACM SIGMOD Conference, Montreal, Canada, pp. 103–114.
[34]台灣颱風洪水研究中心，「典寶溪及宜蘭河防災測試基地監測及加值應用研究」， 經濟部水利署水利規劃試驗所，2015。
[35]張斐章，張麗秋，2015，類神經網路導論-原理與應用第二版，滄海圖書
[36]許銘熙，鄧慰先，黃成甲，1996，「八掌溪流域洪水及淹水預報模式之研究(二)」，行政院國家科學委員會報告
[37]傅金城, 張駿暉, 葉森海, 黃成甲, 謝龍生, 游保杉, 葉克家和許銘熙，2010， 「淹水災害預警技術」，國研科技(25), pp.15-27.
[38]賴進松，張向寬，2001，「防洪示範區淹水境況模擬與決策支援系統之研究(一)-子計畫八：基隆河流域颱洪發生潰堤災害之境況模擬」，行政院國家科學委員會專題研究計畫