Landslide risk analysis subject to geological uncertainty- a viewpoint from a simplified model

NCUIR > Engineering College > Graduate Institute of Civil Engineering > Electronic Thesis & Dissertation > Item 987654321/88029

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/88029

Title:	Landslide risk analysis subject to geological uncertainty- a viewpoint from a simplified model
Authors:	Fitra, Joni;Fitra, Joni
Contributors:	土木工程學系
Keywords:	地質模型不確定性;邊坡穩定;滑動面位態;節理分布;風險;Geological model uncertainty;Slope stability;Plane orientation;Joint distributions
Date:	2022-01-24
Issue Date:	2022-07-13 15:32:40 (UTC+8)
Publisher:	國立中央大學
Abstract:	台灣夏季多雨且處於地震頻繁區域，山區易有土石流、山崩等災害事件，對於台灣來說，順向坡的災害更是常見，一旦坡趾受到施工或是河流沖蝕而見光，容易因外在因素影響而向下坍滑，造成坡趾處保全對象的嚴重損失及傷亡。對於大地工程或工程地質相關災害的分析來說，其影響因子除了大地工程材料的一般物理及力學性質外，地質模型或地質參數也是屬於相當關鍵的因子，對於順向坡滑動，其中較為關鍵的地質參數包括了地下水位的分布，以及潛在滑動面的位態。近年來，大地工程或工程地質的分析愈來愈強調考慮參數變異性以及風險的重要性，而相關研究亦指出地質參數或模型不確定性對於工程地質災害的風險評估有其必要性。因此，本研究以簡化順向坡為基本模型，考慮潛在滑動面傾角以及滑動塊體節理分布不確定性，探討順向坡滑動的機率以及其崩滑塊體運移距離之機率分布，了解地質模型不確定性對於順向坡滑動及崩滑之風險。本研究基本模型中的潛在滑動面傾角來源，主要參考Yeh et al.(2021)所探討的台灣國道三號順向坡滑動案例，滑動面傾角來源主要有四大類，共8項，四大類包含臨近案例坡地之地質圖、災前現場量測值、災後滑動面現場量測值以及災前光達推估資料，每一類資料均有其滑動面之走向及傾角，因此各類資料再依本案例滑動面之傾向投影至滑動面，而得到滑動面上之傾角，並利用蒐集而得之資料計算其統計參數，以利後續蒙地卡羅分析。對於滑動塊體之節理分布來說，本研究將滑動塊體內部之節理分為兩類，分別為與滑動面平行之層面以及與走向節理平行之高角度節理兩組節理，由於缺乏適當之節理分布資料，因此本研究假設兩組節理之統計參數，並建立分離元素模型，再以點估計法計算其塊體崩落運移距離之機率分布。研究結果顯示，以極限平衡法再搭配蒙地卡羅分析所得的滑動機率，與傾角來源有極大的關連性，光達資料所推估之滑動面傾角為較大範圍之平均，因此其變異性較低，平均值也較小，因此其對應之破壞機率最低(16.9)，相反的，由地質圖所推估之傾角，其平均值與變異程度均較高，因此對應之破壞機率為最高(78%)。由運移距離的分析看來，若以光達資料所推估之傾角來進行分析，其塊體運移距離為最短，到達距離坡趾50公尺之機率為17%，而200公尺之機率為0%，研究結果中節理分布對於塊體之運移距離較無法看出整體趨勢。由以上結果可以看出，順向坡的分析及設計，受到地質模型不確定之影響相當大，若以地質圖之傾角來設計，則可以了解在將分析的破壞機率降到一定可接受機率(如10-3)以下，則以地質圖傾角來設計所需之工程成本，會比以光達推估傾角所需之工程成本要高得多，說明了在工程中考慮地質模型不確定性的重要性。 ;Landslide is the primary driver of the denudational process and sediment source dominantly onsite. Landslides are one of the most disastrous effects in Taiwan; groundwater or flood erosion is highly attributed to the landslide. The slope dip angle and water-induced slope cause increased driving force and decreased resisting force, causing a slope landslide. This condition generally affects slope stability to the understanding effect of dip angle attributed to the landslide. In this study, we attempt to consider the uncertainty of the dip angle in slope stability analysis. The dip angles employed in this study were based on Highway no. 3 sliding events in Taiwan. Four different measurement sources, i.e., Central Geological Survey (C.G.S., Taiwan), Compass measurement before the sliding event, Surface measurement after the event, and LiDAR-derived data, were employed in this study. Four methods were employed to analyze attributed dip angle to the landslide, i.e., Limit Equilibrium, monte Carlo simulation, Point estimate method (P.E.M), and 3DEC software simulation. In this research, the Monte Carlo simulation was used to quantify the effect of the geological uncertainty. Various dip angles (with mean and standard deviation) were employed to generate 100000 dip angle samples. The limit equilibrium was used to quantify the safety factor of each measurement dip angle in the condition of no cohesion, no anchor, and no water pressure effect. The point estimate method was used to predict the probability of displacement length, displacement width, and debris volume. The point estimates method also predicts the risk of displacement length to the building in two scenarios applied. 3DEC software was used to simulate the simplified model 100m by 100m by 100m were cut off by dip angle of each measurement and various joint spacing. Limit equilibrium results show that the LiDAR measurement source provides the highest safety factor of the slope due to the lowest dip angle. Monte Carlo simulation results show LiDAR Measurement Source provides the lowest failure probability of 16.9%, and Central Geological Survey (C.G.S., Taiwan) Measurement provides the highest failure probability of 78%. 3DEC simulation shows the effect of damping factor, dip angle, and joint spacing on the displacement length, displacement width, debris volume, and mechanical time of debris to stabilize. The point estimate method (PEM) shows that the LiDAR measurement source provides the lowest probability of 17% of the debris reaching the building at a distance of 50 meters and probability 0% of the building at a distance of 200 meters for all measurement methods. Finally, the risk might be unacceptable for the building at a distance of 50 meters from the slope toe because probability has not reached 0% and expense much money. The risk might be acceptable for the building′s minimum distance of 150meters. Further, the measured dip angles were converted to the projected dip angle based on the plane′s strike. Therefore, based on the engineering design concept, if the design is performed using the C.G.S. data, the engineering design must be very conservative compared to the design using the LiDAR data
Appears in Collections:	[Graduate Institute of Civil Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	44	View/Open

社群 sharing

Loading...