臺灣博碩士論文加值系統

潮汐水位變化影響著沿海地區的生活型態與建設發展，因而若能適時的預測潮汐水位高程的變化，則對相關工程之興建應有相當的助益。本研究之首要重點在於整理台灣第一大港，即高雄港歷年來(1971至2002年)之潮汐觀測資料，並將之數位化。藉由視覺化的軟體設計，使得可以更加容易的觀察水位變化情況與查覺潮汐資料之錯誤。接著，本研究之重點旨在利用人工智慧中常用的倒傳遞類神經網路與遺傳演算法配合潮汐調和方程式，針對具有完整記錄資料的年度(2000年)給予訓練，以便做即時與長期之潮汐預測。從比較結果顯示類神經網路比遺傳演算法更能準確地描述此類問題。同時，從各種不同輸入層、隱藏層及輸出層之組合可找出一相對較佳之網路模式以做進一步之預測。從分析結果可知對各個月份與季節性之潮汐變化，本研究之類神經網路模式確實可適度地預測潮汐水位隨時間之變化情形，這些資料應可提供相關工程設計時之參考依據。

Variation of tidal levels has a significant influence on the living pattern and construction development along the costal region, and thus the prediction of tidal levels varied with time may become very useful in the design and an engineering project. The main purpose of this study, at first, was by using Visual Basic language to arrange the time history (1971-2002) tidal recorded data at Kaohsiung harbor, the largest harbor in Taiwan. From visualized software design, the variation of tidal levels can be seen more clearly and regarding the completeness of recording data can be detected more easily. In the next, the present study was focused on the training of a completed observation data set in the year 2000, with the use of harmonic tidal equations, to predict short-term and long-term tidal levels by back-propagation neural networks and genetic algorithm, which could be found popularly in artificial intelligence applications. The comparison results showed that the neural networks have a better performance than that of genetic algorithm in this type of problem. Meanwhile, from various combinations of input layer, hidden layer and output layer in neural networks, a relatively better model was obtained for further analysis including monthly and seasonly predictions. The analyzed results exhibited that the neural networks model did have the ability to predict the tidal levels varied with time with a moderately accuracy. The obtained information might provide a good reference for the relatively engineering design in the investigation area.

摘 要 I
Abstract II
誌謝 III
目錄 IV
圖索引 VI
表目錄 X
第1章 緒 論 1
1.1 研究動機與目的 1
1.2 文獻回顧 2
1.3 研究內容安排 4
第2章 類神經網路基本原理 6
2.1 類神經網路之發展緣起 6
2.2 類神經網路模式之分類 8
2.3 倒傳遞類神經網路 10
2.4 步進式倒傳遞網路 16
第3章 遺傳演算法基本原理 17
3.1 搜尋法的介紹 17
3.2 遺傳演算法的基本架構 17
3.3 限制條件的處理 23
3.4 步進式遺傳演算法 25
第4章 潮汐數位化軟體之設計開發 26
4.1 潮汐之調和分析 26
4.2 潮汐分析軟體之設計大綱 28
4.3 潮汐分析軟體功能說明 32
第5 章 潮汐預測結果分析 48
5.1 潮汐推估模式短期預測結果分析 48
5.2 潮汐推估模式長期預測結果分析 51
5.3 同月與同季節性潮汐預測分析 54
第6章 結論與建議 96
6.1 結論 96
6.2 建議 98
參考文獻 101
附錄一 即時潮汐推估程式碼 106
附錄二 預測即時潮汐畫面程式碼 112
附錄三 遺傳演算法預測長期潮汐程式碼 122
作者簡介 141

1. 王皓正，(2001)，”應用遺傳演算法於長期潮汐預報之研究”，國立交通大學土木工程研究所碩士論文。
2. 朱執均，(2000)，”類神經網路之應用-南中國海海域潮汐預報及補遺”，國立中山大學海洋環境及工程學系研究所碩士論文。
3. 葉怡成，(1993) ，”類神經網路模式應用與實作”，儒林圖書有限公司。
4. 葉怡成，(1997) ，”應用類神經網路”，儒林圖書有限公司。
5. 張國棟、林維揚、曾相茂、何崇華，(2000),”潮汐預報時間幅度之探討”，第22屆海洋工程研討會論文集,pp.547-554.
6. 曾彥，(2001)，”利用類神經網路於長期潮汐預報之研究”，國立交通大學土木工程研究所碩士論文。
7. 黃明哲、吳祥雲，(1996)，”潮汐預報作業與基準面探討”，海下技術季刊第六卷第一期, pp.29-38.
8. 劉文俊，(1999)，”台灣的潮汐”，文英出版社。
9. Amir F. Atiya, Suzan M EI-Shoura, Samir I.Shaheen and　Mohamed S. EI-Sherif (1999), ”A Comparison Between Neural-Network Forecasting Techniques─Case Study: River 　Flow Forecasting,” IEEE Transactions On Neural Networks, 　Vol.10, No.2, pp.402-409.
10. Amaury Lendasse, Michel Verleysen, Eric de Bodt, Marie Cottrell and Philippe Gregoire (1998), ”Forecasting Time-Series by Kohonen Classification,” European Symposium on Artificial Neural Networks, pp.221-226.
11. Bernard B. Hsish and Thad C. Pratt (2001), ”Filed Data Recovery in Tidal System Using Artificial Neural Networks,”　 US Army Corps of Engineers.
12. B.Bhattacharya and D.P. Solomatine (2000), ”Application of Artificial Neural Network in Stage-Discharge Relationship,” Proc. 4th International Conference on Hydroinformatics, Iowa City, USA.
13. Ching-Piao Tsai and Tsong-Lin Lee (1999), ”Back-Propagation Neural Network in Tidal-Level Forecasting,” Journal of Waterways, Port, Coastal, and Ocean Engineering, Vol.125, No.4, pp.195-202.
14. Donghui Yi, Jean-Bernard Minster and Charles Bentley (1999),” Ocean Tidal Loading Corrections, ”Geoscience Laser Altimeter System (GLAS).
15. G.B. Sheble, T.T. Maifeld, K. Brittig, G. Fahd and S. Fukurozaki-Coppinger (1994), ”Unit Commitment by Genetic Algorithm with Penalty Mthods and A Comparison of Lagrangian Search and Genetic Algorithm-Economic Dispatch Example,” Electrical Power & Energy Systems, Vol.18, No.6, pp.339-346.
16. Hector Allende, Claudio Moraga and Rodrigo Sales, ”Artificial Neural Networks in Time Series Forecasting: A comparative Analysis.”
17. Hojjat Adeli (2001), ”Neural Networks in Civil Engineering: 1989-2000,” Computer-Aided Civil and Infrastructure Engineering, Vol 16, pp.126-142.
18. Hans-Georg Wittkemper, Manfred Steiner (1995), ”Using Neural Networks to Forecast The Systematic Risk of Stocks,” European Journal of Operational Research, Vol.90, pp.577-588.
19. Imran Tasadduq, Shafiqur Rehman and Khaled Bubshait (2002), ”Application of Neural Networks for The Prediction of Hourly Mean Surface Temperatures in Saudi Arabia, ” Renewable Energy, Vol. 25, pp.545-554.
20. Jin Li and Edward P.K. Tsang (1999), ”Improving Technical Analysis Predictions: An Application of Genetic Programming,” Artificial Intelligence Research Symposium, USA
21. Pauline Kneale, Linda See and Andrew Smith (2001), ”Towards Defining Evaluation Measures for Neural Network Forecasting Models.”
22. P.R. Sutcliffe (2000), ”The Development of A Regional Geomagnetic Daily Variation Model Using Neural Networks,” Annales Geophysicae, Vol.18, pp.120-128.
23. R.B. Boozarjomehry, W.Y. Svrcek (2001), ”Automatic Design of Neural Network Structures,” Computers and Chemical Engineering, Vol.25, pp.1075-1088.
24. R.G. Song and Q.Z. Zhang (2001), ”Heat Treatment Optimization for 7175 Aluminum Alloy by Genetic Algorithm,” Materials Science and Engineering, Vol.17, pp.133-137.
25. Safaai Deris, Sigeru Omatu, Hiroshi Ohta and Puteh Saad (1999), ”Incorporating Constraint Propagation in Genetic Algorithm for University Timetable Planning,” Engineering Applications of Artificial Intelligence, Vol.12, pp.241-253.
26. Sameer Singh (1998), ”Forecasting Using a Fuzzy Nearest Neighbour Method,” Proc.6th Internation Conference on Fuzzy Theory and Technology, Vol.1, pp.80-83.
27. Sameer Singh (1998), ”Fuzzy Nearest Neighbour Method For Time-Series Forecasting,”Proc. 6th European Congress on Intelligent Techniques and Soft Computing, Vol. 3, pp.1901-1905.
28. T. Kerh and Y.C. Yee (2000), ”Analysis of A Deformed Three-Dimensional Culvert Structure Using Neural Networks,” Advances in Engineering Software, Vol.31, pp.367-375.
29. T.L. Lee, C.P. Tsai, D.S. Jeng and R.J. Shieh (2002), ”Neural Network for The Prediction and Supplement of Tidal Record in Taichung Habor, Taiwan,” Advances in Engineering Software, Vol.33, pp.329-338.
30. T.L. Lee and D.S. Jeng (2002), ”Application of Artificial Neural Networks in Tide-Forecasting,” Ocean Engineering, Vol.29, pp.1003-1022.
31. T. Tingsanchali (2000), ”Forecasting Model of Chao Phraya River Flood Levels at Bangkok,” Asian Institute of Technology.
32. T. Tchaban, J. P. Griffin and M.J. Taylor (1998), ”A Comparison between Single and Combined Backpropagation Neural Networks in The Prediction of Turnover,” Engineering Applications of Artificial Intelligence, Vol.11, pp.41-47.
33. W. Huang and S. Foo (2002), ”Neural Network Modeling of Salinity Variation in Apalachicola River,” Water Research, Vol.36, pp.356-362.