台灣四面環海，位於西太平洋熱帶氣旋盛行地區，氣候深受季風、颱風及洋流的影響，每逢夏秋兩季之時，常受到颱風的強風暴雨侵襲，平均年降雨量約為 2,510 公釐，其中約有 78%集中在 4~10 月間的梅雨及颱風季節，颱風帶來的大量雨水，常導致淹水災害，而台灣跨河橋梁每逢颱風來襲時常有斷橋之疑慮，為了確保橋梁上車輛安全，橋梁預警的發展越來越重要。在橋梁預警中，水位預報是重要的一大環節，因為本研究目的為建立一套有系統且完整的河川橋樑水位預測模式。本研究蒐集國內河川水位預報模式相關文獻，找出影響水位之因子，並以有完整橋樑水位資料的大安溪流域作為案例，蒐集2011-2017侵台颱風資料、雨量以及水位資料作為模式輸入因子，輸出資料為河川下游橋樑水位。利用SOS-NNLSTM演算法建立國道橋樑水位模式，進行未來1小時至未來2小時水位預報。
研究結果顯示，SOS-NNLSTM於預測河川水位上有不錯的表現，未來可根據本研究之模式搭配即時觀測系統，協助相關管理單位擬訂適當的防災策略。預測未來水位作為防災參考依據，以期達到減災的目的，研究成果可望為橋梁安全預警所使用。

Surrounded by the sea, Taiwan is located in the prevailing tropical cyclone in the western Pacific. The climate is deeply affected by monsoons, typhoons and ocean currents. During the summer and autumn, it is often hit by strong storms of typhoons. The average annual rainfall is about 2,510 mm. 78% concentrated in the plum rain and typhoon seasons from April to October. The heavy rain caused by typhoons often caused flooding disasters. Taiwan’s cross-river bridges often have doubts about broken bridges when typhoons strike, in order to ensure vehicles on the bridges. Safety, the development of bridge warnings is becoming more and more important. Water level forecasting is an important part of bridge warning, because the purpose of this study is to establish a systematic and complete bridge water level prediction model. This study collects relevant literatures on domestic water level forecasting models, finds the factors affecting the water level, and uses the Da'an River Basin with complete national bridge water level data as a case to collect typhoon data, rainfall and water level data for 2011-2017 as model input factors. The output data is the water level of the bridge downstream of the national highway. The SOS-NNLSTM algorithm is used to establish the water level model of the national bridge, and the water level forecast for the next hour to the next 2 hours
The results show that SOS-NNLSTM has a good performance in predicting the water level. In the future, the real-time observation system can be matched according to the model of this study to assist the relevant management units to formulate appropriate disaster prevention strategies. Predicting the future water level as a reference for disaster prevention, in order to achieve the purpose of disaster reduction, the research results are expected to be used for bridge safety warning

[1] 李戎威, 淡水河即時洪水預測模式之建立及實務應用, 水利會訊 頁1-36, 2006.
[2] 陳文信，次集水區降雨觀測於橋梁防災預警之研究，公路總局公路防災中心，2015.
[3] 陳俐穎、李家順、楊秉順、孟伯鈞、陳進發，2014，公路總局地工技術，第 141 期，第 61-70頁，2014.
[4] 陳進發，從橋梁之流域管理與山區公路之風險管理談公路總局之百年防汛，臺灣公路工程，37卷，6期， 2-21，2011
[5] Seibert, J.，. Multi-criteria calibration of a conceptual runoff model using a genetic algorithm，Hydrology and Earth System Sciences 4，215–224，2000.
[6] 張逸凡，支撐向量機在即時河川水位預報之應用，國立成功大學，碩士論文，2005.
[7] 朱芫慧，利用MARS法進行即時河川水位預測之研究，輔仁大學，碩士論文，2008.
[8] 林洙宏，水文即時監測資料，2015應用在河川洪水預報之研究，國立臺灣大學，碩士論文，2010.
[9] 蕭閔鍾，類神經網路應用於颱風時期集水區流量預測之研究，淡江大學，碩士論文，2012.
[10] 紀昀，輻狀基底函數類神經網路結合遺傳演算法於河川水位預測之應用，中原大學，碩士論文，2013.
[11] 李丞倫，以支持向量機結合退火演算法及局部線性內嵌法模式推估降雨事件和河川警戒水位關係–以八掌溪為例，國立臺灣大學，碩士論文，2014.
[12] Assem, H., Ghariba, S., Makrai, G., Johnston, P., Gill, L., Pilla, F. Urban Water Flow and Water Level Prediction Based on Deep Learning. Machine Learning and Knowledge Discovery in Databases Lecture Notes in Computer Science 317–329. 2017.
[13] 洪逸鈞，結合水文及數值模式應用於河川水位預報—以高屏溪為例，國立中央大學，碩士論文，2015.
[14] 林焜詳，支撐向量機與隨機森林應用於颱風時雨量預報之比較，國立成功大學，碩士論文，2016.
[15] 張哲豪、沈志全、吳祥禎、連和政、許至璁、謝孟益、蕭健雄，QPESUMS於水利署FEWS_Taiwan平台的應用，國立臺北科技大學土木工程系，國家高速電腦與計算中心，2015.
[16] 經濟部水利署，中華民國九十六年水文年報，2007.
[17] 田璦菁，颱洪期間區域總雨量估計最佳化之研究，國立中央大學，碩士論文，2003.
[18] 葉怡成，類神經網路模式應用與實作，儒林出版社，2009.
[19] 賀志強，馮寅，2007，長的短時記憶(LSTM)在五聲調適合聲的應用，福建電腦, no. 4, pp. 101-102
[20] S. Hochreiter and J. Schmidhuber, "Bridging long time lags by weight guessing and “Long Short-Term Memory”," Spatiotemporal models in biological and artificial systems, vol. 37, pp. 65-72, 1996.
[21] F. A. Gers and J. Schmidhuber, "Recurrent nets that time and count," in Neural Networks, 2000. IJCNN 2000, Proceedings of the IEEE-INNS-ENNS International Joint Conference on, vol. 3, pp. 189-194: IEEE,2000.
[22] W. Zaremba, I. Sutskever, and O. Vinyals, "Recurrent neural network regularization," arXiv preprint arXiv:1409.2329, 2014.
[23] M.-Y. Cheng and D. Prayogo, "Symbiotic organisms search: a new metaheuristic optimization algorithm," Computers & Structures, vol. 139, pp. 98-112,2014.
[24] D. Prayogo, "An Innovative Parameter-Free Symbiotic Organisms Search (SOS) for Solving Construction-Engineering Problems," PhD thesis, Department of Construction Engineering, National Taiwan University of Science and Technology, 2015.
[25] 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, vol. 30, no. 3, p. 04015036, 2015.
[26] 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, 2014.
[27] M.-Y. Cheng et al., "SOS optimization model for bridge life cycle risk evaluation and maintenance strategies," Journal of the Chinese Institute of Civil and Hydraulic Engineering Volume 26, no. 4, 2017.
[28] Jason Ghorman Herianto, SOS Tuned NN-LSTM Inference Model For Construction Cash Flow Forecasting Considering Complexity Of The Project, National Taiwan University of Science and Technology, 2019.
[29] M.-Y. Cheng and C. Ko, "Computer-aided decision support system for disaster prevention of hillside residents," in Proceedings of the 17th IAARC/CIB/IEEE/IFAC/IFR International Symposium on Automation and Robotics in Construction, Taipei, Taiwan, 2000, pp. 587-592.
[30] K. Fukahori and Y. Kubota, "Consistency evaluation of landscape design by a decision support system," Computer‐Aided Civil and Infrastructure Engineering, vol. 15, no. 5, pp. 342-354, 2000.
[31] S. Sundin and C. Braban‐Ledoux, "Artificial intelligence–based decision support technologies in pavement management," Computer‐Aided Civil and Infrastructure Engineering, vol. 16, no. 2, pp. 143-157, 2001.
[32] M.-Y. Cheng and Y.-W. Wu, "Evolutionary support vector machine inference system for construction management," Automation in Construction, vol. 18, no. 5, pp. 597-604, 2009.
[33] M.-Y. Cheng, P. M. Firdausi, and D. Prayogo, "High-performance concrete compressive strength prediction using Genetic Weighted Pyramid Operation Tree (GWPOT)," Engineering Applications of Artificial Intelligence, vol. 29, pp. 104-113, 2014.
[34] M.-Y. Cheng and M.-T. Cao, "Evolutionary multivariate adaptive regression splines for estimating shear strength in reinforced-concrete deep beams," Engineering Applications of Artificial Intelligence, vol. 28, pp. 86-96, 2014.
[35] M.-Y. Cheng, D. K. Wibowo, D. Prayogo, and A. F. Roy, "Predicting productivity loss caused by change orders using the evolutionary fuzzy support vector machine inference model," Journal of Civil Engineering and Management, vol. 21, no. 7, pp. 881-892, 2015.
[36] M.-Y. Cheng, N.-D. Hoang, and Y.-W. Wu, "Hybrid intelligence approach based on LS-SVM and Differential Evolution for construction cost index estimation: A Taiwan case study," Automation in Construction, vol. 35, pp. 306-313, 2013.
[37] M.-Y. Cheng and A. F. Roy, "Evolutionary fuzzy decision model for cash flow prediction using time-dependent support vector machines," International Journal of Project Management, vol. 29, no. 1, pp. 56-65, 2011.
[38] D. Prayogo, "A Novel Genetic Algorithm-Based Evolutionary Support Vector Machine for Optimizing High-Performance Concrete," Department of Construction Engineering, National Taiwan University of Science and Technology, Taiwan, 2012.
[39] M.-Y. Cheng, D. Prayogo, and Y.-W. Wu, "Novel genetic algorithm-based evolutionary support vector machine for optimizing high-performance concrete mixture," Journal of Computing in Civil Engineering, vol. 28, no. 4, p. 06014003, 2013.
[40] 楊智民、劉廣利, 不確定性支持向量機原理及應用. 中國: 科學出版社, 2006.
[41] 賀志強，馮寅, "長的短時記憶(LSTM)在五聲調適合聲的應用," 福建電腦, no. 4, pp. 101-102, 2007.
[42] 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.
[43] M.-Y. Cheng, D. Prayogo, and Y.-W. Wu, "Predicting the Pavement Rutting Behavior of Asphalt Mixtures Using Symbiotic Organisms Search-Least Squares Support Vector Machine Inference Model," Construction and Building Materials.