本研究以2018至2019年於臺20線52公里處之邊坡場址所架設的物聯網自動測站，蒐集監測之數據以進行邊坡因降雨引致地下水位升降分析。研究以為期兩年之監測成果訓練類神經網路，以監測到的雨量、水位資訊做為輸入特徵來預測未來特定時間段內之地下水位變化為目的。利用場址監測的雨量、水位、含水量計，並擷取特定時間點下的降雨序列、累積雨量(特定深度的含水量值)、水位高程變化和非飽和層厚度作為模型的特徵輸入，各因子分別對應至降雨總水量、非飽和層平均含水量、上游滲流補注變化與垂直入滲逕流長度。以適用於高度非線性關係且具有明顯對應值的倒傳遞神經網路架構進行水位預測模型的建立，並利用監測數據驗證，界定適用於水位預測的特徵因子範圍、模型架構的神經元個數及網路修正權重的演算方式。此類神經網路可供此場址以即時監測成果做水位的升降預測，後續可結合極限平衡分析以評估邊坡安全係數變化，提供相關負責單位進行預警。

Ground-water levels are a principal factor of slope instability and of significance in other geotechnical engineering problems. In this study, we propose a backpropagation neural network with 10-fold cross validation method for predicting the ground-water level induced by rainfall, the site was selected at South Cross-Island Highway 52K and use the value of monitoring system with ground-water level, soil water content and rainfall measurement. Furthermore, correlation analysis is used to find the potential input variables range for the predicting model, such as Pearson correlation coefficient and partial auto-correlation are applied. In this study, we choose four factor as the model input, such as hourly time series of rainfall and borehole water-level, depth of unsaturated zone and unsaturated zone moisture content via 4 different algorithms like the one step algorithm, the scaled conjugate gradient method, the Bayesian algorithm and the Levenberg-Marquardt method to find the best combination for predicting future 12 hours ground-water level. It was found that to obtain satisfactory results, the network using 4 hours of antecedent precipitation, 3 hours of antecedent borehole water-level, the time unsaturated zone thickness and turning 48 hours effective accumulated precipitation into unsaturated zone moisture content performed well in simulating the future 12 hours water-level.

李佳勳，2004，類神經網路應用於高屏溪洪水流量資料推估之研究，國立屏東大學土木工程學研究所，碩士論文。
李宗霖；林宏明；盧育聘，2004年12月，類神經網路在南横公路邊坡破壞潛能之預測. 立德學報, 2(1), 頁 97-115。
李明熹，2006，土石流發生降雨警戒分析及其應用國立成功大學水利及海洋工程研究所，博士論文。
周仕勳，2019，水力力學耦合分析應用於降雨引致土壤邊坡破壞之依時預警，國立成功大學資訊管理研究所，博士論文。
林巧盈，2014，探討K等分交叉驗證法對於分類器錯選率之研究，國立成功大學資訊管理研究所，碩士論文。
林承賢，2012，以類神經網路建構濁水溪流域地下水位推估模式. 國立臺灣大學生物資源暨農學院生物環境系統工程學研究所，碩士論文。
林勝義，2005，阿里山五彎仔邊坡滑動因子相關性與滑動基準值之初步探討. 國立雲林科技大學營建工程研究所，碩士論文。
林進國，2003，降雨和地下水位變化之關聯性分析，國立成功大學水利及海洋工程研究所，碩士論文。
青山工程顧問股份有限公司，2018年1月，台 20 線 52k+150 路基保護工程委託地滑調查、測量及設計服務監測總成果報告，交通部公路總局第五區養護工程處。
洪芯琦，2018，崩積層邊坡滑動即時雲端監測系統研發. 國立成功大學土木工程研究所，碩士論文。
胡永國，2002，應用類神經網路推估地下水位洩降所致地層沉陷之研究，國立屏東大學土木工程研究所，碩士論文。
張舜孔，2003，類神經網路應用在阿里山公路邊坡破壞因子之分析研究，國立成功大學土木工程研究所，碩士論文。
陳宗顯，2006，降雨引致地下水位變化之研究-以那菝、六甲與東和地下水位站為例，國立成功大學水利及海洋工程學系，博士論文。
陳柏丞；葉富鈞，2016年4月，以類神經網路分析台北市交通死亡車禍之關鍵因素及提出解決方法. Journal of China University of Science and Technology, 66.
葉怡成，1997，應用類神經網路，儒林圖書公司。
葉怡成；吳沛儒，2009，基於類神經網路與交叉驗證法之田口方法. 品質學報，16(4)，頁 261-279。
龔文瑞，2007年9月，以地下水位變動法結合消退曲線位移法評估地下水位，農業工程學報，53(3)。
江昭輝，2015，台灣地區土壤含水率觀測網之建置與模擬，國立中興大學環境工程系研究所，碩士論文。
詹連昌，2003，梨山地滑區地下水水位管理基準建立方法研究，國立中興大學水土保持學系，博士論文。

Bustami, R., Bessaih, N., Bong, C., & Suhaili, S. (2007, 11 17). Artificial Neural Network for Precipitation and Water Level Predictions of Bedup River. International Journal of Computer Science.
Bazartseren, B., Hildebrandt, G., Holz (2003).Short-term water level prediction using neural networks and neuro-fuzzy approach,55(4),pp. 439-450.
Campbell, R. H. (1975). Soil Slips, Debris Flows, and Rainstorms in the Santa Monica Mountains and Vicinity, Southern California. Geological Survey(851).
Campolo, Marina; Andreussi, Paolo; Soldati, Alfredo (1999, 4). River flood forecasting with a neural network model. WATER RESOURCES RESEARCH, 35(4), pp.1191-1197.
Caudill, M., & Butler, C. (1992). Understanding Neural Networks. The MIT Press.
Evans, J. D. (1995). Straightforward stastics for the behavioral sciences. Brooks/Cole.
Fetter, C. W. (1993). Contaminant Hydrogeology. pp. 168-175.
Fredlund, D.G., and Rahardjo, H.(1993). Soil Mechanics for Unsaturated Soils. Wiley-Interscience Publication.
Finch, J. W. (2001, 2). Estimating change in direct groundwater recharge using a spatially distributed soil water balance model. Quarterly Journal of Engineering Geology and Hydrogeology(34), 71-83.
Freeze, R. A. (1969, 2). The Mechanism of Natural Ground-water Recharge and Discharge 1. One-dimensional, Vertical, Unsaturated Flow above a Recharging or Discharging Ground-water Flow System. Water Resource Research, 5(153-171).
Gay, L. R. (1992). Educational research competencies for analysis and application. Macmillan. New York.
GholamiV., ChauW.K., FadaeeF., TorkamanJ., & GhaffariA. (2015, 9). Modeling of groundwarer level fluctuation using denrochronology in alluvial aquifers. Journal of Hydrology, pp. 1060-1069
Green, W.H. and Ampt, G.A. (1911). Studies on soil physics, part 1, the flow of air and water though soils. Journal of Agricultural Science, (4),pp. 11-24.
GurjeetSingh, & RabindraPandaK. (2011, 10). Daily Sediment Yield Modeling with Artificial Neural Network using 10-fold Cross Validation Method: A small agricultural watershed, Kapgari, India. International Journal of Earth Sciences and Engineering, 4(6), pp. 443-450.
Hagan, M. T., & Menhaj , M. B. (1994, 11). Training Feedforward Networks with the Marquardt Algorithm. TRANSACTIONS ON NEURAL NETWORKS, 5(6), pp. 989-993.
Hagan, M. T., Demuth, H. B., & Beale, M. (1995). Neural Network Design. Boston, USA: PWS Publishing Company.
Hamilton, J. D. (1994). Time series analysis. Princeton University Press, Princeton, NJ.
Healy W.Richard, & CookG.Peter. (2002, 10). Using groundwater levels to estimate recharge. Hydrogeology Journal, pp. 91-109.
Hebb, D. O. (1949). The organization of behavior; a neuropsychological theory. Wiley.
HortonE.R. (1939). Analysis of runoff-plot experiment with varying infiltration capacity. Trans. Amer. Geophys.
Hush, D., & Horne, B. (1993). Progress in Supervised Neural Networks. IEEE Signal Processing Magazine(10), pp. 8-39.
Iverson, R. M.(2000, 7 1). Landslide triggering by rain infiltration. WATER RESOURCES RESEARCH, 36(7), pp. 1897-1910.
Kearns, M. (1997). A bound on the error of cross validation using the approximation and estimation rates, with consequences for the training-test split. (9), pp. 1143-1161.
Kecman, V. (2001). Learning and Soft Computing:Support Vector Machines, Neural Networks, and Fuzzy Logic Models. The MIT Press.
Lee, L. E., Lawrence, D. L., & Price, M. (2006, 4 17). Analysis of water-level response to rainfalland implications for recharge pathways in theChalk aquifer, SE England. Journal of Hydrology, pp. 604-620.
LeeL.T. (2004). Back-Propagation Neural Network for long-Term Tidal Predictions (第 31 冊). Ocean Engineering .
Lourakis, M. A. (2005, 2 11). A Brief Description of the Levenberg-Marquardt Algorithm Implemented by levmar. Computer Science.
MacKay, D. J. (1992). Bayesian interpolation. Neural computation, 4(3), pp. 415-447.
Maier, H. R., & Dandy, G. C. (2000). Neural networks for the prediction and forecasting of water resources variables: a review of modelling issues and applications. Environmental Modelling & Software(15), pp. 101-124.
Marquardt, D. W. (1963, 6). An Algorithm for Least-Squares Estimation of Nonlinear Parameters. the Society for Industrial and Applied Mathematics, 11(2), pp. 431-441.
McCullochS.W., & PittsW. (1943). A logical calculus of the ideas immanent in nervous activity (第 5 冊). Bulletin of Mathematical Biophysics.
Mein, R. G., & Larson, C. L. (1973). Modeling infiltration during a steady rain.
Mohanty, S., Jha, M. K., & Raul, S. K. (2015). Using Artificial Neural Network Approach for Simulataneous Forecasting of Weekly Groundwater Levels at Multiple Sites. Water Resource Manage.
Moller Fodslette Martin. (1993). A scaled conjugate gradient algorithm for fast supervised learning. Neural Networks, 6(4), pp. 525-533.
Richards, A.L. (1931). Capillary conduction of liquids through porous mediums. pp. 318-334.
Rodriguez J. D., Perez A. and Lozano J. A. (2010), Sensitivity Analysis of k-Fold Cross Validation in Prediction Error Estimation, Transactions on Pattern Analysis and Machine Intelligence, 32(3), pp. 569-575
Rumelhart, D. E., & McClelland, J. L. (1986). Parallel distributed processing: explorations in the microstructure of cognition, vol. 2: psychological and biological models. Cambridge: MIT Press.
Sangrey, D. A., Harrop‐Williams, K. O., & Klaiber, J. A. (1984). Predicting Ground‐Water Response to Precipitation. J. Geotech. Engrg., 110(7), pp. 957-975.
Shiri, J., Kisi, O., Yoon, H., Lee, K. K., & Nazemi, A. H. (2013, 11 2). Predicting groundwater level fluctuations with meterological effect implications-A compariative study among soft computing techniques. Computers and Geosciences, pp. 32-44.
Stephens, D. B. (1996). Unsaturated Zone Hydrology. Lewis Publishers, Boca Raton: Florida, pp. 347.
Sakellariou, M.G., Ferentinou, M.D.(2005).A study of slope stability prediction using neural networks. Geotrchnical and Geological Engineering, 23(4).pp.419-445
Sudheer Goasin, A,K,, Ramasastri, K.S.K.P.,. (2002). A data-dricen algorithm for constructing artificial neural network rainfall-runoff models., pp. 1325-1330.
Van Genuchten, M. T. (1980, 9). A Closed-form Equation for Predicting the Hydraulic Conductivity of Unsaturated Soils1. Soil Science Society of America Journal, 44(5), pp. 892-898.
Villiers, J. d., & Barnard, E. (1993, 2). Backpropagation Neural Nets with One and Two Hidden Layers. 4(1), pp. 136-141.
Werbos Paul. (1974). Beyond regression : new tools for prediction and analysis in the behavioral sciences. Harvard University.
Williams, J.P. (1982). The surface of the Earth, an introduction to geotechnical science. New York.