本研究針對區域雨水利用規劃之各項影響供水可靠度因子，以區域歷年日雨量分佈資料為基礎，經由區域降雨頻率、雨水逕流量、水量平衡與供水可靠度等分析，得到不同需水條件及供水可靠度要求下之雨水貯槽設計容量建議值，並以二種不同型式之類神經網路模型建立小區域雨水逕流供水系統可靠度模擬與預測模式，並測試模式穩定度與學習速率，未來可作為規劃系統設置後之可靠度預測或應用於智慧型控制裝置開發。
就預測成功率而言，BPNN模型約為83%、RBFNN 模型約為98.6%，此等結果與訓練或測試階段之結果相若，顯示模式本身有良好的穩定性。另檢視在神經網路訓練過程中之失誤或失敗的19%(BPNN)、6%(RBFNN)，顯示其高估一級者佔10.18%(BPNN)、0.93%(RBFNN)及低估一級者佔8.80%(BPNN)、0.46%(RBFNN)；換言之，即使本模式評估有誤，但影響差異亦僅一級，且高估一級者與低估一級者樣本點相若。表示依目前建立之模式，其預測結果將較趨向樂觀，且與預測結果趨於悲觀之差異僅約9%(BPNN)、0.5%(RBFNN)，顯示出模式本身的穩定性。
以本研究之BPNN 4-3-1架構、修正BPNN4-3-1-1架構以及RBFNN架構，應用於台北市郊相同區域實際設置雨水徑流利用設施對供水可靠度進行預測，並收集2000~2007年之系統實際供水數據進行可靠度計算驗證，比對的三種預測供水可靠度模式中，RBFNN供水可靠度之預測模型較為保守，而BPNN的二種優化模式則有較為樂觀的估計，這與原模式效能分析時所得結論是一致的。
依據模式模擬與實際案例數據之比對結果，RBFNN模式雖然在預測精度上也較一般BPNN佳，但在與實際案例資料比對後發現，誤差率反較修正BPNN網路架構4-3-1-1大，這也凸顯出RBFNN網路設計的學習速度雖然比BP網路快，但是透過調整隱藏層、節點數目以及訓練參數之BPNN優化模式對實際之預測仍有其優勢存在。未來如改變不同傳輸函數（例如二元雙彎曲函數等）、優化網絡結構（開發新的方法，以獲得最佳一些隱藏的節點）、選擇網絡類型和訓練算法進行模擬，或採用不同系列的類神經網路架構來進行模式建構分析比較，相信對進行相關研究或實務設計將可提供不同的貢獻。

Many factors have to be considered in the reliability analysis of planning the regional rainwater utilization tank capacity. Based on the historical daily rainfall data, the following four analyze procedures were conducted: the regional daily rainfall frequency, the amount of runoff, the water continuity, and the reliability. The suggested designed storage capacity could be obtained according to the conditions with the demand and supply reliability. By using the output data, two different types of artificial neural network models were used to build up small area rainfall–runoff supply systems for the simulation of reliability and the prediction model. This study should have significant benefit in the future application of the instantaneous prediction or the development of related intelligent instantaneous control equipment.
In terms of the success rate of prediction as a whole, the results are about 83% for BPNN and 98.6% for RBFNN. Such result is similar to the training or testing results, indicating the model itself has good stability. Inspecting the misses or failures of 19% for BPNN and 6% for RBFNN during the artificial neural network training process of this research, it shows that the percentage of overestimation by one grade is 10.18% for BPNN and 0.93% for RBFNN, while for underestimation by one grade; the percentage is 8.80% for BPNN and 0.46% for RBFNN. In other words, even if there are mistakes in the estimation of this model, the variance in the effect is just one grade and the sample points of overestimation by one grade is similar to the underestimation by one grade. This also indicates that, based on the current model, the variance of prediction result tends to be “optimistic” but the prediction result tends to be “pessimistic”, only around 9% for BPNN and 0.5% for RBFNN, further indicating the stability of the model itself.。
Compared with three models, RBFNN was more conservative and BPNN had more optimistic estimates. Besides, we observed the predicted behavior of three models with the sensitivity analysis of the parameter between runoff coefficient and water supply and shows that the inference of water supply of RBFNN model was more conservative than the BPNN4-3-1-1 model.
Despite the fact that RBFNN was more reliable than BPNN, it still made a conservative estimate for the actual monitoring data. The error rate of RBFNN was still higher than the correction of BPNN 4-3-1-1. Although the learning speed of RBFNN was faster than BPNN, it could keep the advantage at the actual prediction by adjusting the number of hidden layers and nodes. This should have significant benefit in the future application of the instantaneous prediction or the development of related intelligent instantaneous control equipment. It is believed that changing different transfer functions (e.g., binary, logistic, sigmoid, etc.) for the simulation or adopting different series of ANN structure for model construction analysis is necessary in the future.

參考文獻

1. 工業技術研究院能源與資源研究所，1997，〈雨水貯留供水系統技術成果彙編〉，經濟部技術處。
2. 工業技術研究院能源與資源研究所，1999，〈節約用水措施推動：初期雨水截留設施評估研究與效能測試〉，經濟部水資源局。
3. 工業技術研究院能源與資源研究所，2004，〈綠色廳舍暨學校水資源改善計畫〉，第21-22頁，內政部建築研究所。
4. 內政部營建署，2004，〈建築物雨水貯留利用設計技術規範〉，內政部營建署。
5. 李士畦，1999，〈雨水/中水利用規劃設計訓練班教材〉，第C-1~C-16頁，經濟部水資源局。
6.李士畦，2001，”替代水資源”，土木技術，第4卷7期，第81-87頁。
7. 李士畦、盧文俊、林志鴻等，2002，〈用水計畫書審查技術規範之研擬〉，經濟部水利署。
8.李士畦，林憲德，2006，“初期降雨操作策略與其在綠建築設計中的應用”，第十五屆水利工程研討會論文，中央大學，桃園。
9.林國峰等，2003，“優選法及回饋演算於降雨–逕流預報模式之研究(1)”，農業水利科技研究發展91年度成果發表討論會，屏東。
10.林怡菁，2008，〈臺灣地區屋頂雨水收集水質特性與使用評估之研究〉，國立台北科技大學環境工程與管理研究所碩士論文。
11. 陳仁宗、李士畦，1997，〈雨水貯留供水系統最適化設計技術準則〉，工業技術研究院能源與資源研究所。
12.陳弘凷、王國樑、黃泳塘，2007，”臺灣地區都市污水廠現況暨放流水利用潛勢”，節約用水季刊47期，第9-21頁。
13.葉怡成，1997，〈類神經網路模式應用與實作〉，儒林出版社。
14.經濟部水利署，2005，〈機關學校節約用水技術手冊〉，經濟部水利署。
15.Balram Panigrahi, Sudhindra N Panda, Bimal Chandra Mal, 2006, “Rainwater conservation and recycling by optimal size on-farm reservoir”, Resources Conservation and Recycling, Vol. 50(4), pp.459-474. (SCI)
16.Chen S., 1991,“Orthogonal least squares learning algorithm for radial basis function networks [J].＂, IEEE Trans on NN,Vol.2(2) ,pp. 302- 309.
17. Dawson, C.W., Wilby, R.L., 2001, “Hydrological modeling using artificial neural networks”, Progress in Physical Geography, Vol. 25(1), pp. 80-108. (SCI)
18. Edgar L. Villarreal, Andrew Dixon, 2005, “Analysis of a rainwater collection system for domestic water supply in Ringdansen, Norrko¨ ping, Sweden”, Building and Environment, Vol., 40, pp. 1174–1184. (SCI)
19. Elshorbagy, A., Simonovic, S.P., Panu, U.S., 2000, “Performance evaluation of artificial neural networks for runoff prediction”, Journal of Hydrologic Engineering, October, pp. 424-427. (SCI)
20. Enedir Ghisi & Sulayre Mengotti de Oliveira, 2007, “Potential for potable water savings by combining the use of rainwater and greywater in houses in southern Brazil”, Building and Environment, Vol. 42, pp. 1731–1742. (SCI)
21. E. Sazakli, A. Alexopoulos, M. Leotsinidis, 2007,“Rainwater harvesting quality assessment and utilization in Kefalonia Island, Greece”, Water Research, Vol., 41, pp. 2039–2047. (SCI)
22. E.W. Steel and Terence J. McGhee, 1979,“Water Supply and Sewerage”, 5th Edition, McGraw. Hill published, USA.
23. Ham F. M. and Kostanic I., 2001, Principles of neurocomputing for science engineering, McGraw-Hill: New York, NY.
24. Hecht-Nielsen, R., 1987, “Counterpropagation network”, Applied Optics, Vol. 26, pp. 4979–4984. (SCI)
25. Jeong, D.I., Kim, Y.O., 2005, “Rainfall-runoff models using artificial neural networks for ensemble streamflow predicting”, Hydrological Process, Vol. 19, pp. 3819-3835. (SCI)
26. Jeroen Mentens, Dirk Raes, Martin Hermy, 2006. “Green roofs as a tool for solving the rainwater runoff problem in the urbanized 21st century?”, Landscape and Urban Planning, Vol. 77, pp.217-226. (SCI)
27. Ju, Q., Yu, Z., Hao, Z., Zhu, C., Liu, D., 2007, “Hydrologic Simulations with Artificial Neural Networks”, Third International Conference on Natural Computation, August, China.(IEEE).
28. Kun Zhu, Linus Zhang, William Hart, Mancang Liu, Hui Chen, 2004, “Quality issues in harvested rainwater in arid and semi-arid Loess Plateau of northern China” , Journal of Arid Environments, Vol., 57, pp. 487–505. (SCI)
29. Liaw Chao-Hsien and Yao-Lung Tsai, 2004,“Optimum Storage Volume of Rooftop Rain Water Harvesting Systems for Domestic Use”, Journal of the American Water Resources Association (JAWRA), Vol. 40(4), pp.901-912. (SCI)
30. Lippmann, R. P., 1987, “An introduction to computing with neural nets”, IEEE ASSP Magazine, Vol. 4(2), pp. 4–22.
31. Lubing Handia, James Madalitso Tembo, Caroline Mwiindwa, 2003, “Potential of rainwater harvesting in urban Zambia”, Physics and Chemistry of the Earth, Vol.28, pp. 893–896. (SCI)
32. Martinson, D.B. and Thomas, T., 2003, “Improving water quality by design”, The 11th International Rainwater Catchment Systems Conference, Mexico City, Mexico.
33. Martinson, D.B. and Thomas, T., 2005, “Quantifying the first flush phenomenon”, The 12th International Rainwater Catchment Systems Conference, New Delhi, India.
34. Mcmahon, T.A. and Mein, R.G., 1978, “Reservoir Capacity and Yield”, pp.6-68, Elsevier Scientific Publishing Company.
35. Mirchandani, G., & Cao, W., 1989, “On hidden nodes for neural nets”, IEEE Transactions on Circuits and Systems, Vol. 36(5), pp661–664. (SCI)
36. Rajurkar, M.P., Kothyari, U.C., Chaube, U.C., 2002, “Artificial neural networks for daily rainfall-runoff modeling”, Hydrological Sciences-Journal-des Sciences Hydrologiques, Vol., 47(6), pp. 865-877. (SCI)
37. Ree-Ho Kim, Sang-ho Lee, Jin-woo Jeong, Jung-Hun Lee, Yeong-Kwan Kim, 2007, “Reuse of greywater and rainwater using fiber filter media and metal membrane” , Desalination., Vol., 202, pp. 326–332. (SCI)
38. Ruby L Kennedy, Yuchun Lee, Benjamin Van Roy, Christopher D. Reed, Richard P. Lippmann, 1998, “ Solving Data Mining Problems through Pattern Recognition” , pp.10-5, Unica Technologies, Inc. published, USA.
39. Sartor, J., & Boyd, G., 1972, 〈Water Pollution Aspects of Street Surface Contaminants〉, U.S. Environmental Protection Agency.
40. Shih-Chi Lee, Hsien-Te Lin, 2006, “Artificial Neural Network Analysis for Reliability Prediction of Regional Runoff Utilization”, CIB 32th International Symposium: Water Supply and Drainage for Buildings, Taipei, Taiwan.
41. Shih-Chi Lee, Hsien-Te Lin, 2009, “First-flush Rainfall Operation Strategy for the Rainwater Utilization System Design”, The 14th International Rainwater Catchment Systems Conference, Kuala Lumpur, Malaysia.
42. Shih-Chi Lee, Hsien-Te Lin, 2010, “Artificial Neural Network Analysis for Reliability Prediction of Regional Runoff Utilization”, Environmental Monitoring and Assessment, Vol. 161, pp. 315-326. (SCI)
43. Stephen N. Ngigi, 2003, “What is the limit of up-scaling rainwater harvesting in a river basin?”, Physics and Chemistry of the Earth, Vol.28, pp. 943–956. (SCI)
44. Warren Viessman, 1989, Jr., Gary L. Lewis, Knapp, John W, “Introduction to hydrology”, Harper & Row published, USA.