臺灣博碩士論文加值系統

臺灣地區，由於山地地形關係平均坡度陡峻，豪雨季節容易山洪暴漲氾濫成災，乾旱季節則見旱象危機無水可用，因此興建水庫和構築堰壩，成為政府重要水資源政策，除了蓄水及兼具河川整治與防洪減災功能外，更可創造發電、觀光附加價值。
而在臺灣人口密度高居世界各國第2位情況下(國家人口數達1,000萬以上比較)，已構築大小不等的水庫及蓄水設施100餘座，各式堰壩更達200座以上。有鑑於國外曾發生多起水庫大壩潰堤，造成重大人員傷亡之慘劇，針對水壩安全性除了要求工程規劃、設計及施工之技術品質因素外，營運啟用的安全維護與管理更是永續水資源之不二法門，因此作好壩體變位量之監控，直接影響水庫設施運作之安全。
本研究蒐集翡翠水庫大壩歷年來監測資料，彙整分析可能影響壩體變位量之初步因子選項，利用統計軟體SPSS對初步因子與輸出變數(變位)進行相關性分析，客觀挑選出影響壩體變位量之重要因子作為研究模型的輸入參數，並應用不同的人工智慧理論，進行案例資料庫的學習訓練，再以各種推論模式進行測試，得到壩體變位量之預測成果值。
為驗證各種人工智慧推論模式之預測準確性，本研究模擬壩體變位曲線，展現各種模型預測成果比較，並分別以線性相關係數(Linear Correlation Coefficient， R)、均方根誤差 (Root Mean Square Error， RMSE)、平均絕對誤差(Mean Absolute Error， MAE)、平均絕對百分比誤差(Mean Absolute Percent Error， MAPE)進行預測準確性之誤差衡量，最後再以綜合性指數(Synthesis Index, SI)做為整體成果評估標準而得到驗證結論，結果以「自調適時間函數之演化式最小平方差支持向量機推論模式」(ELSIMT)最佳。

In Taiwan, the average gradient is steep owing to mountainous terrain; therefore, the occurrence of flash floods in the monsoon and drought crises in the dry season is common. Therefore, the construction of reservoirs and barrages has become an important facet of government water policy. In addition to functions like water storage, river regulation, and flood control, reservoirs and barrages provide additional value in the form of power generation and tourist attractions.
As Taiwan has the second-highest population density in the world (among countries with a population of 10 million or more), there are approximately 100 reservoirs and water storage facilities of different sizes and over 200 barrages of several types. Globally, several dam burst incidents have occurred, causing heavy casualties. Therefore, for dam safety, in addition to the requirements of factors such as project planning and design and technical quality of construction, security maintenance and management during operation is extremely essential for the sustainable utilization of water resources. Therefore, monitoring dam displacement directly influences the operational safety of the reservoir facilities.
This study involved collecting monitoring data of the Emerald dam over several years and analyzing preliminary factors that might influence the dam displacement. Thereafter, correlation analysis between preliminary factors and the output variable (displacement) was conducted using a statistical software (IBM SPSS) to select significant variables that influence the dam displacement as input parameters of research models. After the application of different artificial intelligence theories, learning and training in the case database, and testing by a variety of inference models, the predictions of the dam displacement were obtained.
To verify the prediction accuracy of various artificial intelligence inference models, this study simulated dam displacement curves and presented the comparison of prediction results among different models. The prediction accuracy was measured by Linear Correlation Coefficient (R), Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Mean Absolute Percent Error (MAPE). Finally, the Synthesis Index (SI) was used as the evaluation criteria of the overall results for verification. Adaptive Time-dependent Evolutionary Least Squares Support Vector Machine Inference Model (ELSIMT) was determined to be the best model.

第1章 緒論 1
1.1 研究動機 1
1.2 研究目的 3
1.3 研究範圍與限制 3
1.4 研究方法與流程 5
1.4.1 研究方法 5
1.4.2 研究流程 6
1.5 論文架構 9
第2章 文獻回顧 10
2.1 翡翠大壩變位監測系統 10
2.1.1 翡翠水庫大壩介紹 10
2.1.2 監測儀器 12
2.1.3 擺線儀(Plumbline) 12
2.1.4 變位量計算 15
2.2 早期推估大壩變位量方式 16
2.2.1 圖示法 16
2.2.2 重迴歸分析法 17
2.2.3 半迴歸分析法 17
2.2.4 即時分析監測法 21
2.3 卡方自動交互檢測（CHAID） 24
2.4 迴歸分析(Regression) 24
2.5 類神經網路(ANN) 25
2.6 支持向量機(SVM) 27
2.7 演化式支持向量機(ESIM) 29
2.7.1 快速混雜基因演算法(fmGA) 29
2.7.2 演化式支持向量機推論模式(ESIM) 34
2.8 演化式最小平方差支持向量機(ELSIM) 37
2.8.1 最小平方差支持向量機(LS-SVM) 38
2.8.2 差分進化演算法(DE) 39
2.8.3 ELSIM的架構流程 41
2.8.4 ELSIM特性與使用限制 44
2.9 生物共生演算法最小平方差支持向量機(SOS-LSSVM) 45
2.9.1 生物共生演算法(SOS) 45
2.9.2 生物共生演算法最小平方差支持向量機(SOS-LSSVM) 48
2.9.3 SOS-LSSVM特性與使用限制 51
2.10 自調適時間函數結合ELSIM推論模式(ELSIMT) 52
2.10.1 時間序列(Time Series) 52
2.10.2 自調適時間函數(ATF) 54
2.10.3 ELSIMT的架構流程 56
第3章 變位量影響因子確立 59
3.1 因子資料彙整 59
3.2 因子要項分析及篩選 60
3.2.1 因子要項分析 60
3.2.2 確立因子 61
3.3 正規化 65
3.4 建立案例資料庫 65
第4章 變位量推論模式建立與驗證 66
4.1 推論模式選用 66
4.1.1 資料採礦軟體SPSS Modeler 66
4.1.2 SVM推論模式 71
4.1.3 ESIM推論模式 73
4.1.4 ELSIM推論模式 76
4.1.5 SOS-LSSVM推論模式 80
4.1.6 ELSIMT推論模式 82
4.2 各種推論模式訓練與測試 84
4.2.1 各種推論模式執行環境 84
4.2.2 訓練案例與測試案例分組 85
4.2.3 各種推估模型測試成果 86
4.2.4 大壩變位模擬 89
4.3 誤差衡量 91
4.3.1 線性相關係數(R) 91
4.3.2 均方根誤差 (RMSE) 92
4.3.3 平均絕對誤差（MAE） 92
4.3.4 平均絕對百分比誤差（MAPE） 92
4.3.5 綜合性指數（SI） 93
4.4 預測成果驗證 94
4.4.1 各高程測點誤差驗證 94
4.4.2 模型性能比較 96
4.4.3 模型預測成果綜合評估 98
4.5 小結 100
第5章 結論與建議 101
5.1 結論 101
5.2 建議 102
參考文獻 103
附錄A 106
附錄B 137

1.國際大壩委員會(International Commission on Large Dams , ICOLD),全球資訊網,http:// www.icold-cigb.org/
2.高穆賓,「水庫水酷」,三聯技術雜誌,第74期,第2~7頁,2009。
3.陳振川、黃韶誠、林焜峰,「翡翠大壩壩體長期變位預測分析」,翡翠水庫管理局,1993。
4.翡翠水庫水庫管理局,「翡翠水庫安全檢查綜合評估年報」,2013。
5.吳育偉、鄭明淵,「支持向量機最佳化模式-應用於營建管理決策」,第11屆營建工程與管理學術研討會,2007。
6.Min-Yuan Cheng and Nhat-Duc Hoang, “Estimating Compressive Strength of Rubberized Concrete Using Evolutionary Least Squares Support Vector Machine”,Technical Report, CIC Lab, National Taiwan Univ. of Sci. and Tech,2012.
7.Min-Yuan Cheng, Doddy Prayogo and Yu-Wei Wu, (in preparation), “Predicting the Pavement Rutting Behavior of Asphalt Mixtures Using Symbiotic Organisms Search - Least Squares Support Vector Machine Inference Model”, Construction and Building Materials.
8.Nhat-Duc Hoang, “Decision Support System for Construction Management Based on Evolutionary Least Squares Support Vector Machine”, PhD Dissertation,CIC Lab, National Taiwan Univ. of Sci. and Tech,2013.
9.李延財,「翡翠水庫定期安全評估重點與案例簡介」,翡翠水庫管理局簡報,2013。
10.張森源,「大地工程四十載」,大地工程技師公會、科技圖書股份有限公司,2008。
11.傳重焌、余興河,「翡翠大壩變位觀測系統之介紹」,地工技術雜誌,第10期,第1~5頁,1985。
12.高穆賓,「監測儀器介紹與監測資料研判重點」,三聯科技簡報資料,2010。
13.黃韶誠,「拱壩安全預警系統之建立-以翡翠大壩為例」,國立臺灣大學土木工程學研究所碩士論文,1992。
14.蕭文龍,「多變量分析最佳入門實用書」,碁峰資訊股份有限公司,2009。
15.葉怡成,「類神經網路模式應用與實作」,9 ed.:儒林出版社, 2009。
16.Cortes and V. Vapnik, “Support-vector network”, Machine Learning, Vol.20, No.3, pp.273-297, 1995.
17.D.E. Goldberg, K. Deb, H. Kaegupta and G. Harik, “Rapid Accurate Optimization Of Difficult Problems Using Fast Messy Genetic Algorithms”, Proceedings of the Fifth International Conference on Genetic Algorithms, pp. 56– 64, 1993.
18.D.Knjazew and G.A. Ome, “A Competent Genetic Algorithm for Solving Permutation and Scheduling Problems”, Kluwer Academic Publishers, Boston, 2003.
19.許淑婷,「利用快速混雜基因演算法與模擬機制建立設計專案作業程序最佳化之研究」,國立成功大學碩士論文,2006。
20.簡崑棋,馮重偉,「結合模糊類神經模式與快速混雜基因演算法於專案工期之預測」,國立成功大學碩士論文,2003。
21.Thorsten Joachims, http://svmlight.joachims.org/svm_struct.html
22.Suykens, J. and et al., “Least Square Support Vector Machines”. World Scientific Publishing Co. Pte. Ltd., 2002.
23.Min-Yuan Cheng and Nhat-Duc Hoang, “Evolutionary Least Squares Support Vector Machine – Userguide”, Technical Report, CIC Lab, National Taiwan Univ. of Sci. and Tech,2012.
24.Storn, R. and K. Price, “Differential Evolution - A Simple and Efficient Heuristic for Global Optimization over Continuous Spaces”. J. Global Optim.11: 341–359,1997.
25.Price, K.V., R.M. Storn and J.A. Lampinen, “Differential Evolution A Practical Approach to Global Optimization. Springer-Verlag”, 2005.
26.Min-Yuan Cheng and Doddy Prayogo, “Symbiotic Organisms Search: A new metaheuristic optimization algorithm”, Computers & Structures, Vol. 139, pp. 98-112, 2014.
27.G.G. Tejani, V.J. Savsani and V.K. Patel, “Adaptive symbiotic organisms search (SOS) algorithm for structural design optimization”, Journal of Computational Design and Engineering, 2016.
28.M.-Y. Cheng, D. Prayogo and D.-H. Tran, “Optimizing Multiple-Resources Leveling in Multiple Projects Using Discrete Symbiotic Organisms Search”, Journal of Computing in Civil Engineering, 2015.
29.D.-H. Tran, M.-Y. Cheng and D. Prayogo, “A novel Multiple Objective Symbiotic Organisms Search (MOSOS) for time–cost–labor utilization tradeoff problem”, Knowledge-Based Systems, vol. 94, pp. 132-145, 2016.
30.M.-Y. Cheng, C.-K. Chiu, Y.-F. Chiu, Y.-W. Wu, Z.-L. Syu, D. Prayogo and C.-H. Lin, “SOS optimization model for bridge life cycle risk evaluation and maintenance strategies”, Journal of the Chinese Institute of Civil and Hydraulic Engineering, vol. 26(4), pp. 293-308, 2014.
31.S. Duman, “Symbiotic organisms search algorithm for optimal power flow problem based on valve-point effect and prohibited zones”, Neural Computing and Applications, 2016.
32.H. Kamankesh, V.G. Agelidis and A. Kavousi-Fard, “Optimal scheduling of renewable micro-grids considering plug-in hybrid electric vehicle charging demand”, Energy, vol. 100, pp. 285-297, 2016.
33.E. Ruskartina, V.F. Yu, B. Santosa and A.A.N.P. Redi, “Symbiotic Organism Search (SOS) for Solving the Capacitated Vehicle Routing Problem”, International Journal of Mechanical, Aerospace, Industrial, Mechatronic and Manufacturing Engineering, vol. 101, pp. 857-861, 2015.
34.A. Panda and S. Pani, “A Symbiotic Organisms Search algorithm with adaptive penalty function to solve multi-objective constrained optimization problems”, Applied Soft Computing, vol. 46 pp. 344-360, 2016.
35.羅俊雄,「翡翠水庫大壩微振動量測及長期監測資料分析研究」,翡翠水庫管理局,2009。
36.Cohen,J,Statistical Power Analysis for the Behavioral Sciencees(2nd ed).Hillsdale,NJ:Lawrence Erlbaum Associates,1988.
37.王国平、郭伟宸、汪若君, 「数据与文本挖掘実战」,清华大学出版社,中国北京,2014。
38.Lewis,C. D. Industrial and Business Forecasting Method. London: Butterworth-Heinemann,1982.
39.Jui-Sheng Chou, Kuo-Hsin Yang, Jusieandra Pribadi Pampang and Anh-Duc Pham, Evolutionary metaheuristic intelligence to simulate tensile loads in reinforcement for geosynthetic-reinforced soil structures, Computers and Geotechnics 66 , 1-15,2015.
40.吳演聲,「翡翠大壩變位監測警戒值及其視窗化展示查詢系統」,技術學刊第22卷第3期,2007。
41.陳賜賢,「淺談臺灣水庫之安全及管理」,水利技師公會資訊會訊,2014。