台灣為海島國家，又處於地震活躍區環太平洋火山帶中，菲律賓海板塊和歐亞板塊交界上，近幾十年來，有鑒於巴布亞新幾內亞及311福島大海嘯造成的核電廠災難，台灣的防海嘯設施是否完善值得我們省思。
海嘯衝擊相關基礎設施(如海嘯牆)之安全評估過程中，難以避免的，包含許多不確定性的因素在內，因此，為了更進一步掌握分析的成果，除了傳統的定然式分析，本研究擬導入可靠度分析的概念以增加成果展示的可信度。
由於目前國內尚無防海嘯牆的結構體，相關的定然式分析較為缺乏，因此，本研究將參考國外相關的設計資料與分析方法，再輔以本研究所開發的可靠度分析法進行分析。又由於海嘯數據之蒐集之困難，故造成資料過少而無法進行可靠度分析，在此應用馬可夫練蒙地卡羅演算法增加數據量，並使用最小平方向量機及卡方檢驗來測試演算法之精確度，以便判斷資料之分佈型態。搭配以上判斷後，即可以用易損曲線的方式呈現海嘯牆結構物對於不同條件下之可靠度。
本研究以上部、下部以及整體結構分別分析結構物，上部結構分析顯示，海嘯波高越高則需較厚的海嘯牆厚度；下部結構分析顯示，當基樁數目越多，系統破壞機率越小；而整體系統分析，將同時考慮牆高、牆厚及基樁數目，並判斷控制因子。雖然所分析的結構物並非國內目前既有的海嘯牆，但本研究所建立的分析演算流程，仍對未來國內相關結構物之分析過程具有重要的參考價值。

Taiwan is a sea island which is located at Pacific-rim volcanic belt and the boundary of Philippine Sea Plate and Eurasian Plate. In recent decades, due to the disaster of the nuclear power plant that caused by the tsunami in Papua New Guinea in Japan, it is worth thinking to us whether Taiwan's anti-tsunami facilities have well protection.
The safety evaluation process for the basic facilities related to tsunami attack (e.g., Tsunami Wall), inevitably, contains several uncertain factors. Therefore, in order to capture the uncertainty influences, apart from the traditional deterministic analysis, the study adopts the concept of reliability analysis to enhance the analysis credibility.
Because of the lack of the domestic structure of anti-tsunami wall, a related design from Japan is used as a baseline structure in our deterministic analysis. Together with the reliability analysis, we developed a reliability safety evaluation process. Since collecting tsunami data is not an easy task, as a result, only few historic data are available for reliability analysis. In view of this, Markov Chain Monte Carlo (MCMC) is proposed to increase the quantity of data, Least Square-Support Machine (LS-SVM) and Chi-squared Test are used to ensure the accuracy of algorithm, so that we can determine the distribution of the data. With the assessments above, a fragility curve is used to present the credibility of tsunami wall under different conditions.
Although the analysis of structure in this study isn't a domestic tsunami wall, the algorithm we built still has important referential value for future study.

[1] Fukuyama, H., Kato, H., Ishihara, T., Tajiri, S., Tani, M., Okuda, Y., Kabeyasawa, T. and Nakano, Y., Structural design reuirement on the tsunami evacuation buildings. U. Kirsch, “Structural Optimization”, Springer-Verlag, 1993.
[2] Asakura R., Iwase K., Ikeya T., Takao M., Kaneto T., Fujii N. and Omori M., 2000, ”Experimental research on wave power of tsunami surged over the land beyond the upright seawalls”, Coastal engineering Journal, Japan Society of Civil Engineering, No.47, 911-915.
[3] Nouri, Y., Nistor, I., Palermo, D., and Cornett, A. 2007. "Structural Analysis for Tsunami-Induced Force and Debris Impact." Coastal Structures 2007, Venice, Italy.
[4] FEMA P55, "Determining Site-Specific", Capter 8.
[5] Como, A. and Mahmoud, H., 2013, "Numerical evaluation of tsunami debris impact loading on wooden structural walls." Engineering Structures 56, 1249–1261.
[6] Chakraborty, D. and Choudhury, D., 2014, "Stability of non-vertical water front retaining wall supporting inclined backfill under earthquake and tsunami." Ocean Engineering 78, 1–10.
[7] Charusrojthanadech, N., 2011, "Trial of damage evaluation by Indian ocean tsunami in Patong beach of Thailand." Coastal Engineering Journal, Vol. 53, No. 4, 285–317.
[8] Ang, A.H. and Tang, W.H. 1984, " Probability concepts in engineering planning and design Vol. 2: decision, risk and reliability, John Wiley, New York.
[9] Naess, A., Leira, B.J. and Batsevych, O. 2009, " System reliability analysis by enhanced Monte Carlo simulation." Structural Safety 31, 349–355.
[10] 濱岡電廠工程現況。
[11] FEMA P646, June 2008, "Guidelines for Design of Structures for Vertical Evacuation from Tsunamis",6.5, 69-99.
[12] Ramsden, J.D., 1993, Tsunamis: Forces on a Vertical Wall Caused by LongWaves, Bores, and Surges on a Dry Bed, Report No. KH-R-54,W.M. Keck Laboratory, California Institute of Technology,Pasadena, California.
[13] 廖翊鈞、許長安、廖哲民，2011，「橋墩劑沖刷計算模式建立研究」，跨河橋梁安全預警系統暨橋梁沖刷試驗室研討會論文集。
[14] 李昭興、許明光、鄭文彬、蕭士俊、林欽隆, “我國東北角海嘯警報系統建立之初始研究”
[15] Support Vector Machine，WIKIAPEDIA http://en.wikipedia.org/wiki/Support_vector_machine
[16] J.A.K. Suykens, J. Vandewalle, 1998, ”Least Squares Support Vector Machine Classifiers”.
[17] 林冠中，2005，「漸進式支持向量機於人臉辨識之應用」，國立成功大學資訊工程學系碩士論文
[18] Amaury Lendasse, Yongnan Ji, Nima Reyhani, and Michel Verleysen, “LS-SVM Hyperparameter Selection with a Nonparametric Noise Estimator”, Neural Network Research Centre, Helsinki University of Technology.
[19] J. Kennedy and R. C. Eberhart, “Particle swarm optimization, 1995,” in: Proc. IEEE Int. Conf. on Neural Networks, Perth, Australia, vol. 4, pp. 1942-1948.
[20] 生物統計學，第十章，卡方分佈
[21] Pearson’s chi-squared test , WIKIAPEDIA, https://en.wikipedia.org/wiki/Pearson%27s_chi-squared_test
[22] Braverman, J.l. (BNL), Miller, C.A. (BNL), Ellingwood, B.R. (GIT), Naus, D.J. (ORNL), Hofmayer, C.H. (BNL), Shteyngart, S. (BNL), Bezler, P. (BNL), U.S. Nuclear Regulatory Commission, "Probability-Based Evaluation of Degraded Reinforced Concrete Components in Nuclear Power Plants", Washington, DC 20555-0001.
[23] 吳祚任，2001，「台灣潛在高於預期之海嘯模擬與研究」，國立中央大學水文與海洋科學研究所，地震海嘯模擬結果報告書