由於目前地球暖化,致使氣候變遷加劇使得降雨強度增大,降雨時間延長等極端事件發生頻率上升,造成平均日暴雨量增加,排水系統無法負荷等問題,更增加了降雨預測技術的挑戰.總觀臺灣地區在降雨強度逐年增加且集中之情況下,帶來重大的災害影響,導致最大日降雨量,颱風時降雨量以及水庫入流量超乎警戒值等;另外,降雨量與流量之增加,超出公共排水設施的原有設計負荷而導致淹水情事.因此,建構一套高精度,即時的降雨預測模型,精準且及時的預測降雨量,為現階段工程防洪所需依據之重要課題.
有鑑於此,本研究擬用灰色系統的特性並依據2007年[林耀煌]提出的創新灰預測模式為基礎,提出創新灰預測修正模式.創新灰預測模式對時間序列在極端值預測及延遲預測方面較傳統灰色模式有明顯的成效,惟對時間序列中大於二倍標準差以上的極端值之預測,在預測精度及時間上仍有待改善.本研究擬提出二種修正方法,第一以動態指數法與創新灰預測模式建構動態灰預測修正模式,第二用創新灰預測模式結合模糊隸屬函數建構模糊灰預測修正模式,分別改進時間序列平滑度與改進波動過大類型之極端值預測,亦將與倒傳遞類神經網路,時間序列方法（ARIMA）進行比對與分析,並透過每年最大日降雨量,颱風的時降雨量與水庫入流量之預測,驗證本研究成果的合理性.

The current global warming has led to dramatic climate changes, such as increased intensity and prolonged time of rainfall, resulting in problems of increased average daily rainfall and overloading of the drainage system. This has increased difficulty in resulted in challenges to rainfall prediction technology. In overview, the rainfall intensity has been increasingly and becoming more concentrated in Taiwan in recent years, posing great impact of major disasters as the maximum daily rainfall, rainfall in typhoon and water reservoir inflow are all above the warning values. In addition, the increase in rainfall and flow beyond the original design loads of the public drainage facilities has caused floods. Hence, building a high precision, real- time rainfall prediction model to accurately and timely predict rainfall has become an important topic for flood control at the present stage.
This study attempts to use the grey system characteristics, along with the innovative grey prediction model proposed by Lin (2007), as the basis to propose the modification of novel grey prediction model. The novel grey prediction model is apparently more effective than traditional grey models in terms of prediction of the extreme values and delays of time series. However, it needs improvement in prediction precision and time in case of the prediction of the extreme values above two standard deviations in the time series. This study plans to propose two modification methods: the first is to use the dynamic index method and the innovative grey prediction model to build the modified dynamic grey prediction model; the second method is to combine the innovative grey prediction model with the fuzzy membership function to construct the fuzzy grey prediction modification model to improve the time sequence smoothness and prediction of extreme values with overly large fluctuations. In addition, the methods will be compared with the ANN and ARIMA for matching and analysis. The rationality of the research findings will be verified by the prediction of the annual maximum daily rainfall, the hourly rainfall in typhoon and the reservoir inflow amount.

[1] Labadie, J. W., Lazaro R. C., and Morrow, D. M., “Worth of Short-Term Rainfall Forecasting for Combined Sewer Overflow Control,” Water Resources Research, No. 5, pp. 1489-1497 (1981).
[2] J. Obeysekera, T. B., Tabios III, G. Q., and Salas, J. D., “On Parameter Estimation of Temporal Rainfall Models,” Water Resources Research, Vol. 23, No. 10, pp. 1837-1850 (1987).
[3] P.Burlando, R. Rosso, Cadavid, L. G. and Salas, J. D., “Forecasting of Short-TermRainfall Using ARMA Models,” Journal of Hydrology, Vol. 144, pp.193-211 (1993).
[4] Klatt, P., and Schultz, G. A., “Flood Forecasting on the Basis of Radar Rainfall Measurement and Rainfall Forecasting,” Proceedings of Hamburg Symposium, IAHS Publ. No. 145, pp. 307-315 (1983).
[5] Zevin, S. F., “A Probabilistic Approach to Flash Flood Forecasting,” Ph.D. Dissertation, Dept. of Hydrology and Water Resources Administration, Univ. of Arizona, U.S.A. (1986).
[6] Lin, C. B., Su, S.F., and Hsu, Y. T., “High-precision forecast using grey models”, International Journal of Systems Science, 32:609-619 (2001).
[7] Deng, J. L., “Control problems of gray system,” Systems & Control Letters, No. 1, pp. 288-294 (1982).
[8] Deng, J. L., “Introduction to grey system,” Journal of Grey System, No.1, pp.1-24 (1989).
[9] Deng, J. L., “Grey Differential Equation,” Journal of Grey System, No.1, pp.1-14 (1993).
[10] Kayacan, E., Ulutas, B., and Kaynak, O., “Grey system theory-based models in time series prediction,” Expert Systems with Applications, 37:1784–1789, (2010).
[11] Chang, F. J., and Hwang, Y. Y., ”A self-organization algorithm for real-time flood forecast”. Hydrological Processes, No. 13, pp.123-138 (1999).
[12] Chiang, Y. M., Chang, L. C., and Chang, F. J., “Comparison of static-feedforward and dynamic-feedback neural networks for rainfall–runoff modeling,” Journal of Hydrology, No. 290, pp. 297-311 (2004).
[13] Chang, L. C., Chang, F. J. and Chiang, Y. M., “A two-step-ahead recurrent neural network for stream-flow forecasting,” Hydrological Processes, No. 18, pp. 81-92 (2004).
[14] Lin, G. F., and Chen, G. R., “A systematic approach to the input determination for neural network rainfall–runoff models,” Hydrological Processes, No. 22, pp. 2524–2530 (2008).
[15] Chiang, Y. M., and Chang, F. J., “Integrating hydrometeorological information for rainfall-runoff modelling by artificial neural networks,” Hydrological Processes, No. 23, pp. 1650-1659 (2009).
[16] Chen, Y. H., and Chang, F. J., “Evolutionary artificial neural networks for hydrological systems forecasting,” Journal of Hydrology, No. 367, pp. 125-137 (2009).
[17] Lin, G. F., and Wu, M. C., “A hybrid neural network model for typhoon -rainfall forecasting,” Journal of Hydrology, No. 375, pp. 450-458 (2009).
[18] Yu, P. S., Chen, C. J., Chen, S. J., and Lin, S. C., “Application of Grey Model Toward Runoff Forecasting,” Journal of The American Water Resources Association, Vol. 37, No. 1, pp. 151-166 (2001).
[19] Kang, M. G., Park, S. W., and Cai, X. M. “Integration of Hydrologic Gray Model with Global Search Method for Real-Time Flood Forecasting,” Journal of Hydrologic Engineering, Vol. 14, No. 10, pp. 1126-1145 (2009).
[20] Yu, P. S., Chen, C. J., and Chen, S. J., “Application of grey and fuzzy methods for rainfall forecasting,” Journal of Hydrologic Engineering, No. 5, pp. 339-345 (2000).
[21] Bedient, P. B., Holder, A., Benavides, J. A., and B. E., “Vieux ,Radar-based flood warning system applied to tropical storm Allison,” Journal of Hydrologic Engineering, Vol. 8, No. 6, pp. 308–318 (2003).
[22] Kang, B. S., Kim, S. Y., and Ko, I. H., “Grid-based estimation of reservoir flood inflow using radar-observed precipitation,” Proc., Conf. Korea Water Resources Association, Korea Water Resources Association, Seoul, South Korea, pp. 183–188 (2005).
[23] 向躍霖，「灰色擺動序列建模方式研究」，貴州環保科技，第十卷，第一期，第5-8頁 （2004）。
[24] 向躍霖，「灰色擺動序列的GM（1,1）擬合建模法及其應用」，湖南省化工環保，第十八捲，第299-302頁 （1998）。
[25] Li, B., “The Grey Modeling for Central Symmetry Sequences,” Journal of Grey system, No. 2, pp. 95-103, (1990).
[26]阮忠信，「水文灰色系統模式之研析及其與參數優選法之應用」，碩士論文，國立臺灣大學，台北，(1992)。
[27]王如意、莊文南，「微分水文灰色模式之研究與應用」，農業工程學報，第三十九卷，第三期，第67-86頁 （1993）。
[28]徐世忠，「中短期月需水量預測模式之建立」，碩士論文，國立台灣海洋大學，基隆（1996）。
[29]陳憲宗，「灰色預測在逕流預報之應用」，碩士論文，國立成功大學，台南（1998）。
[30] Lee, R. H., and Wang, R. Y., “Parameter-Estimation with Colosed Noise Effect for Differential Hydrological Grey Model,” Journal of Hydrology, Vol 208, Iss 1-2, pp. 1-15 （1998）.
[31]黃顯琇，「灰色系統與類神經網路在水文過程之預測」，碩士論文，國立成功大學，台南 (1999)。
[32] Wen, J. C., Huang, K. H., and Wen, K. L., “The Study of α in GM(1,1) Model,” Journal of the Chinese Institute of Engineers, Vol. 23, No. 5, pp. 583-589 （2000）.
[33] Zadeh, L. A., “Fuzzy sets,” Information and Control, No. 8, pp. 338-353 (1965)
[34]鄧聚龍，灰色系統理論教程，華中理工大學出版社，1990。
[35]劉思峰、史本廣，「灰色系統理論在科學發展中的作用和地位」，農業系統科學與綜合研究，第168-170頁 (2000)。
[36] Liu, S. F., “Emergence and development of grey systems theory and its frontier trends,” Chinese Journal of Management Science, No. 11, pp. 29-32（2003）.
[37]溫坤禮、張簡士琨、葉鎮愷、王建文、林慧珊，MATLAB在灰色系統理論的應用，全華科技圖書股份有限公司，台北，(2006)。
[38] Lin, Y. F., “A novel method of residual recognition to increase forecast precision of grey model,” National Conference of Grey Theory And Applications, pp. A65-A71 (2001).
[39] Cheng, B. Q., Han, F. Q., and Gui, Z. H., “Application of wavelet transform based grey theory to fault forecasting of hydroelectric generating sets,” Power System Technology, No. 29, pp. 40-44 (2005).
[40] Su, S. F., Lin, C. B., and Hsu, Y. T., “A high precision global prediction approach based on local prediction approaches,” IEEE Transactions on Systems, Man, and Cybernetic-Part C: Applications and Review, No. 32, pp. 416-425 (2002).
[41] Guo, L. C., Wu, W., Guo, J. Q., Wang, P., and Zhou, B. S., ”Applying Grey Swing Modelto Predict the Incidence Trend of Hemorrhagic Fever with Renal Syndrome in Shenyang,” Journal of China Medical University, Vol.37, No. 6, pp. 839-842 (2008).
[42] Hsu, Y. T., Liu, M. C., Yeh, J., and Hung, H. F., “Forecasting the turning time of stock market based on Markov–Fourier grey model,” Expert Systems with Applications, No. 36, pp. 8597–8603 (2009).
[43] Lin, Y. H., and Lee, P. C., “Novel high-precision grey forecasting model,” Automation in Construction, No. 16, pp. 771–777 (2007).
[44]史玉峰，寧津生，「指數平滑法改進灰色模式及其在形變數據分析中的應用」，煤炭學報，第三十卷，第二期，第206-209頁 （2005）。
[45] Lin, Y. H., Lee, P. C., and Chang, T. P., “Adaptive and high-precision grey forecasting model,” Expert Systems with Applications, No. 36 pp. 9658–9662 (2009).
[46] Grudnitski, B., and Osburn, L., “Forecasting S&P and Gold Futures Prices :An Application of Neural Networks,” September pp. 631-643 （1993）。
[47]葉怡成，類神經網路模式應用與實作，儒林，台北(1993)。
[48]Diane, E., “Machine Learning,” Training &Development Journal, Vol. 44, No. 12, pp. 24-29 (1990).
[49] Henson, T., Huxhold, W., and Bowman, D., “Anenhanced neural network learning algorithm withsimulated annealing,” Third Workshop on NeuralNetworks, pp. 87-94 (1992).
[50]余尚武，賴佩君，「灰色系統、模糊理論與約略集理論於權變投資組合保險策略之應用」，中國管理評論國際學報，第十卷，第四期，第1-31頁 （2007）。
[51]陳守煜、邱林，「水資源系統管理的廣義保證率與多目標模糊優化」，水科學進展，第四卷，第三期，第215-220頁 （1993）。
[52]陳守煜，「從研究汛期描述論水文系統模糊集分析的方法論」，水科學進展，第六卷，第二期，第133-138 頁 （1995）。
[53]李成林、榮國核，「察爾森水庫汛期模糊集分析與應用」，水利管理技術，第6 期，第1-5頁 （1996）。
[54]李新亭、田立暄，「1995 年汛期白山水庫與豐滿水庫防洪調度及主要經驗」，水利管理技術，第四期 （1996）。
[55]張斐章、徐國麟，「利用模糊集理論推估河川流量之研究」，農工學報，第三十六卷，第四期 （1990）。
[56] Wang, A. P., and Cheng, B. W., “Fuzzy Decision Analysis in Reservoir Operation during Flood Period,” Journal of Chinese Fuzzy Systems Association, Vol. 5, No.2, pp. 35-44 （1999）.
[57] Tay, F. E., and Cao, L. J., ”Application of support vector machines in financial time series forecasting,” Omega, No. 29, pp. 309-317. (2001).
[58] Oh, K. J., and Kim, K. J.,’’Analyzing stock market tick data using piecewise nonlinear model,’’ Expert Systems with Applications, Vol. 22, No. 3, pp. 249-255 (2002).
[59] Donald, S. R., and Jorgens, C., “Ethylene Economics and Production Forecasting in a Changing Environment,” Engineering & Process Economics, Vol. 3, No. 4, pp. 267-278 (1978).
[60] Box, G. E. P., and Jenkins, G., Time Series Analysis: Forecasting and Control. San Francisco: Holden-Day (1970).
[61]田自力、陳朝光，「灰色預測模型累加生成運算應以一次為原則」，灰色系統理論與應用研討會，第351-355頁 (2006)。
[62]陳志良、溫志宏、溫志超，「灰色建模級比性質之研究」，灰色系統理論與應用研討會，第111-115頁 （1998）。
[63]俏新平、李峰、宋中平鄧聚龍、張廷政，「關於GM（1,1）建模技術理論的最新發展」，灰色系統理論與應用研討會，第317-323頁 （2000）。
[64]陳宏旭、張偉哲，「灰色系統在台灣營造業就業人口之預測研究」，灰色系統理論與應用研討會，第226-229頁 （1997）。
[65] Chu, Y. W., and Chung, H. Y., “A fuzzy controller with grey prediction model for the motion of a robot,” National Symposium on Automatic Control, pp. 267-271 (2000).
[66] Chen, J. H., Hung, L. C., and Chung, H. Y., “Study on Fuzzy controller and grey prediction model for the motion of a robot,” National Symposium on Automatic Control, pp. 379-383 (2001).
[67] Hung, L. C., Liu, Y. C., and Chung, H. Y., “An associated design of Fuzzy logic with grey-neural prediction in PID controller,” National Symposium on Automatic Control, pp. 628-631 (2001).
[68] Jiang, Y., and Li, S., “Rey System Model for Estimating the Pavement International Roughness Index,” Journal of Performance of Constructed Facilities, No. 19, pp. 62-68 (2005).
[69]陳濤捷，「灰色預測模型的一種拓廣」，系統工程，第八卷，第50-52頁(1990)。
[70]李群，「灰色預測模型的進一種拓廣」，系統工程理論與實踐，第十三卷，第64-66頁（1993）。
[71]何斌、蒙清，「灰色預測模型拓廣方法研究」，系統工程理論與實踐，第九卷，第138-141頁（2002）。
[72]李翠鳳、戴文戰，基於函數cotx變換的灰色建模方法，系統工程，第二十三卷，第110-114頁 （2005）。
[73]吳漢雄、溫坤禮、游美利、張廷政，「GM（1,1）模型中 值的棒棒分析」，第四屆灰色系統理論與應用研討會，第243-246頁 （1999）。
[74]陳宏旭、張偉哲，「灰色建模最佳背景值之研究」，灰色系統理論與應用研討會，第71-76頁 （1996）。
[75]陳宏旭、張偉哲，「最佳化灰色模型GM（1,1）之研究」，灰色系統理論與應用研討會，第77-80頁 （1998）。
[76]羅黨、劉思峰、黨耀國，「灰色模型GM（1,1）優化」，中國工程科學，第五卷，第50-53頁（2003）。
[77]陳光武、賴家瑞，「灰預測中固定參數 值之最佳化」，灰色系統理論與應用研討會，第A26-A31頁（2001）。
[78]潘淨一、蘇宏明，「灰色預測模型、幕函數與線性回歸整合之應用」，灰色系統理論與應用研討會，第A95-A97頁 （2001）。
[79]鄧聚龍、郭洪、溫坤禮、張廷政、張偉哲，灰預測模型方法與應用，高立圖書有限公司，台北，（1999）。
[80] 劉思峰、黨耀國、方志耕，灰色系統理論及其應用，科學出版社（第三版），（2004）。
[81] 劉思峰、鄧聚龍，「GM（1,1）模型的適用範圍」，灰色系統理論與應用研討會，第81-84頁 （1998）。
[82]黨耀國、劉思峰、霍振杰，「具有灰律數據序列的建模方式研究指數」，中國管理科學，第十二卷，第72-74頁 （2004）。
[83] 李秉展，「創新灰色模型在營建工程應用之研究」，博士論文，臺灣科技大學，台北 (2008)。
[84] Brigham, E. O., The Fast Fouries Transform and Its Applications, Prentice-Hall, Inc., 1998.
[85]張保隆，現代管理數學，華泰文化有限公司，台北，(2005)。
[86]陳憲宗、游保杉，「灰色系統理論在降雨預測之應用」，1997 年灰色系統理論與應用研討會論文集，第176-183頁（1997）。
[87] Chen, J. Y., Huang, H. S., and Liao, Y. Y., “Typhoon Rainfall and Short Term Rainfall Prediction of Cho-shui River Basin during the Typhoon Periods in 2005,” Journal of Chinese Soil and Water Conservation,Vol. 39, No. 1, pp. 11-26（2008）.
[88]張奕郎，「侵台颱風伴隨雨帶和降雨分析」，碩士論文，臺灣師範大學，台北 (1998)。
[89]鄭博文，「洪水時期水庫操作之模糊決策分析─以石門水庫為例」，碩士論文，私立中原大學，桃園（1997）。
[90]王安培，廖研毅，「灰色系統旬入流量預測之研究」，中原學報，第三十二卷，第一期，第9-20頁 （2004）。
[91]經濟部水利署水文水資源資料管理供應系統，http://gweb.wra.gov.tw/wrweb/。