本研究先進行理想化破壞曲面的建構與分析,利用分析出的破壞面來探討數值模擬中網格系統之差異。在理想化破壞曲面的分析,本研究選擇將參考橢圓作為初始可能崩塌範圍的選取參考基準,進行可能崩塌面的分析。嘗試使用窮舉法將地形因素會造成不同的結果都納入考量,雖然分析過程都在 CUDA 架構上完成,其耗時極短,但由於計算量龐大,造成整體計算時間變長,因此將參考橢圓法搭配基因演算法 (Genetic Algorithm, GA) 進行最佳化計算,利用基因演化的概念,保留表現較好的破壞面,大量減少考慮的橢圓數量,同時增加搜尋範圍,以此方式搜尋出最有可能的潛在破壞面,在最佳化計算中,本研究考量不同目標的函數,並比較對於結果的差異,最後將此方法應用於實際案例進行適用性的探討。
將最有可能的潛在破壞面作為起始條件,進行土砂運移行為的模擬,探討數值模擬中採用不同網格系統之差異,在進行土砂運移行為的數值模擬中,結構性網格為最廣為使用之方式,不過對於較崎嶇之區域並不一定能精準的描述地形變化,因此三角網格系統及非結構性網格系統等不同的網格系統也隨之產生。本研究在此部分將比較不同的網格系統於土石流運移行為的差異,以此分析各種網格在實際應用的適用性,並且加入 CUDA 架構進行 GPU 高效率計算,以此減少計算時間,增加此方法的應用性,使災害的預測流程能夠從崩塌的開始到停止,達到即時的預警效果。

The study is devoted to establishing an efficient methodology for hazard assessment by numerical scenario investigation, where the GPU-high-performance computation based on CUDA architecture is employed. It utilizes the genetic algorithm(GA) integrating with the ellipse-referenced idealized curved surface (ER-ICS) for preliminarily evaluating the plausible failure surface even when the initial prerequisites are limited. The ER-ICS is a smooth surface constructed with reference to an ellipse, where two constant curvatures are assigned in the down-slope and cross-slope directions, respectively, once the depth at the center is given. By translating, rotating, and side-tilting the reference ellipse, the most appropriate ICS is figured out regarding the specified conditions. Compared with the exhaustive method, even with a broader search range, the GA method may improve efficiency by reducing the number of candidate ICSs.
After determining the ICS location, the following essential information for disaster assessment is the simulation of the consequent flow paths. There are two typical mesh systems by numerical simulations: one is quadrilateral (the regular structured meshes), and the other one is in triangular form (unstructured meshes). According to the topographic features, each grid system has its intrinsic advantages and disadvantages. Hence, the numerical simulations are performed on both the quadrilateral meshes and the triangular mesh. The results confirmed the applicability of the GPU-accelerated implementation in the unstructured meshes. The hazard assessment can evaluate the disaster from the source area to the runout of the landslide mass.

[1] G. Bradski. The opencv library. Dr. Dobb’s Journal: Software Tools for the Professional Programmer, 25(11):120–123, 2000.
[2] E. K. Chong and S. H. Zak. An introduction to optimization. John Wiley & Sons, 2004.
[3] K. A. De Jong. An analysis of the behavior of a class of genetic adaptive systems.
University of Michigan, 1975.
[4] J. D. Foley, F. D. Van, A. Van Dam, S. K. Feiner, J. F. Hughes, and J. Hughes. Computer graphics: principles and practice, volume 12110. Addison-Wesley Professional, 1996.
[5] J. J. Grefenstette. Optimization of control parameters for genetic algorithms. IEEE Transactions on systems, man, and cybernetics, 16(1):122–128, 1986.
[6] F. Guzzetti, F. Ardizzone, M. Cardinali, M. Rossi, and D. Valigi. Landslide volumes
and landslide mobilization rates in umbria, central italy. Earth and Planetary Science Letters, 279(3-4):222–229, 2009.
[7] R. M. Haralick, S. R. Sternberg, and X. Zhuang. Image analysis using mathematical morphology. IEEE transactions on pattern analysis and machine intelligence, (4):532–550, 1987.
[8] F. Hecht. New development in freefem++. Journal of numerical mathematics, 20(3-4):251–266, 2012.
[9] J. H. Holland. Adaptation in natural and artificial systems: an introductory analysis with applications to biology, control, and artificial intelligence. MIT press, 1992.
[10] X.-B. Hu and E. Di Paolo. An efficient genetic algorithm with uniform crossover for air traffic control. Computers & Operations Research, 36(1):245–259, 2009.
[11] M. Jaboyedoff, F. Baillifard, R. Couture, J. Locat, P. Locat, and J. Rouiller. Toward
preliminary hazard assessment using dem topographic analysis and simple mechanic modeling. Landslides Evaluation and Stabilization, London, pages 191–197, 2004.
[12] M. Jaboyedoff, M. Chigira, N. Arai, M.-H. Derron, B. Rudaz, and C.-Y. Tsou. Testing a failure surface prediction and deposit reconstruction method for a landslide cluster that occurred during typhoon talas (japan). Earth Surface Dynamics, 7(2):439–458, 2019.
[13] C.-J. Ko, C.-L. Wang, H.-K. Wong, W.-C. Lai, C.-Y. Kuo, and Y.-C. Tai. Landslide scarp assessments by means of an ellipse-referenced idealized curved surface. Frontiers in Earth Science, page 862, 2021.
[14] C.-Y. Kuo, P.-W. Tsai, Y.-C. Tai, Y.-H. Chan, R.-F. Chen, and C.-W. Lin. Application
assessments of using scarp boundary-fitted, volume constrained, smooth minimal surfaces as failure interfaces of deep-seated landslides. Frontiers in Earth Science, 8:211, 2020.
[15] S. Mehrotra. On the implementation of a primal-dual interior point method. SIAM Journal on optimization, 2(4):575–601, 1992.
[16] C. Meier, M. Jaboyedoff, M.-H. Derron, and C. Gerber. A method to assess the probability of thickness and volume estimates of small and shallow initial landslide ruptures based on surface area. Landslides, 17(4):975–982, 2020.
[17] M. E. Reid, S. B. Christian, D. L. Brien, and S. Henderson. Scoops3d—software to
analyze three-dimensional slope stability throughout a digital landscape. US Geological Survey Techniques and Methods, book, 14, 2015.
[18] Y.-C. Tai, C.-J. Ko, K.-D. Li, Y.-C. Wu, C.-Y. Kuo, R.-F. Chen, and C.-W. Lin. An
idealized landslide failure surface and its impacts on the traveling paths. Frontiers in
Earth Science, 8:313, 2020.
[19] Y.-C. Tai, J. Vides, B. Nkonga, and C.-Y. Kuo. Multi-mesh-scale approximation of thin geophysical mass flows on complex topographies. Communications in Computational Physics, 29(1):148–185, 2021.
[20] F. E. Taylor, B. D. Malamud, A. Witt, and F. Guzzetti. Landslide shape, ellipticity and length-to-width ratios. Earth Surface Processes and Landforms, 43(15):3164–3189, 2018.
[21] R. Van Den Boomgaard and R. Van Balen. Methods for fast morphological image transforms using bitmapped binary images. CVGIP: Graphical Models and Image Processing, 54(3):252–258, 1992.
[22] A. Y. Zomaya and Y.-H. Teh. Observations on using genetic algorithms for dynamic load-balancing. IEEE transactions on parallel and distributed systems, 12(9):899–911, 2001.
[23] 柯奇均. 在 CUDA 架構下以理想曲面估算崩塌破壞面及地形座標系統上的二相
流 GPU 平行模式應用. 國立成功大學水利及海洋工程學系碩士論文 , 2021.