柔性加勁基礎抗斷層錯動引致地表變形之研究

Abstract

1999年集集地震造成車籠埔斷層帶附近大量房屋與結構物因地表變形損壞與倒塌，目前國內已於建築技術規則中明訂，活動斷層兩側禁建範圍內不得從事建築開發，然而由於臺灣活動斷層分布密集，部分線性構造物例如高速公路、隧道或擋土結構等公共基礎建設，仍無法避免由斷層帶通過，大幅增加斷層錯動地表變形災害之風險。本研究以國道4號臺中環線豐原-潭子路段加勁擋土牆橫跨車籠埔斷層為啟發，透過模型試驗(1g model tests)與有限元素法分析(Finite element analysis)評估平面加勁基礎(Geosynthetic reinforced soil foundation)與外包加勁砂柱(Geosynthetic encased granular column, GEC)對於減緩正逆斷層錯動引致地表變形之效果。 本研究首先探討平面加勁基礎受正斷層作用下，其減緩最大地表角變量之效果，評估地表變形、加勁材應變量、剪裂帶發展與力學機制，比對模型試驗與數值模擬結果，驗證數值分析工具，並透過全尺度數值模擬討論平面加勁材鋪設長度、勁度、張力強度與基礎厚度等之影響，提出正斷層作用下加勁材鋪設長度建議。接著，本研究探討外包加勁砂柱基礎受逆斷層作用下，其減緩最大地表角變量之效果，評估地表變形、剪裂帶之發展與力學機制，並與平面加勁基礎比較。透過模型參數試驗討論外包加勁砂柱水平間距之影響，找出最佳化水平間距，並藉由三維全尺度數值模擬評估外包加勁砂柱變形與外包加勁材之應變量。 研究結果顯示平面加勁基礎可有效減緩正斷層錯動引致之地表變形，透過阻斷效應(Shear rupture interception effect)阻止剪裂帶發展至地表，並藉由發展加勁材張力膜效應(Tensioned membrane effect)將斷層錯動量分散至較寬之影響範圍，降低最大地表角變量，當正斷層垂直錯動比達30%時，最大地表角變量減緩60%，大幅降低正斷層錯動地表變形災害之風險。全尺度數值模擬結果也顯示基礎厚度、加勁材勁度與間距對於地表角變量與加勁材應變量具有顯著之影響，且當加勁材鋪設長度超過正斷層錯動引致之張力區影響範圍時，加勁材將有效減緩主要沉陷(Primary settlement)之地表角變量，並防止加勁材拉出破壞與次要沉陷(Secondary settlement)產生;而當加勁材張力強度較低時，加勁材發生拉斷破壞，最大地表角變量顯著增加。外包加勁砂柱基礎之研究結果則顯示當逆斷層錯動時，外包加勁砂柱透過剪裂帶分散(Diffusion)與偏轉(Diversion)效應之作用，降低最大地表角變量，當逆斷層垂直錯動比達30%時，最大地表角變量約減緩23.3%。此外，模型參數試驗結果指出外包加勁砂柱之水平間距會影響剪裂帶發展與其力學機制，且當水平間距與樁徑比為3.3時，剪裂帶分散與偏轉效應可有效發展，降低地表角變量之效益最高。三維全尺度數值模擬之結果亦指出當逆斷層錯動時，上盤之加勁砂柱未受到影響，下盤之加勁砂柱發生傾斜變形，外包加勁材之最大張應變大約沿著斷層剪裂帶之路徑發展，下盤最靠近斷層面之加勁砂柱因受較大之土壓力與土壤-加勁材互制作用，其外包加勁材於斷層垂直錯動比達22.5%時，發生拉斷破壞。

This research presents a series of 1g model tests and finite element (FE) analyses to investigate the performance of geosynthetic-reinforced soil (GRS) and geosynthetic encased granular columns (GEC) reinforced foundations subjected to fault movement. The aim is to evaluate the performance of GRS and GEC reinforced foundations as a mitigation measure for surface faulting hazards. This research first conducts a series of 1g model tests on GRS foundations across a normal fault. A 3-m thick foundation in prototype subjected to a normal fault vertical displacement up to 90 cm was modeled in the 1g model tests. Digital image analysis (DIA) techniques were applied to determine the surface settlement profile, angular distortion, shear rupture propagation, and mobilized reinforcement tensile strain at various magnitudes of fault offset. Finite element (FE) analyses were developed to investigate the reinforcing mechanism of GRS foundations subjected to normal fault movement. Experimental and numerical results of unreinforced and GRS foundations were compared for model validation. Parametric studies were conducted to evaluate the influence of soil and reinforcement parameters on the performance of GRS foundations. Design method was also developed for determining the reinforcement length against significant pullout. Secondly, a series of 1g model tests on GEC reinforced foundations across a reverse fault were conducted in this research. The performance of GEC reinforced foundations under reverse faulting and the reinforcing mechanism of the geotextile encasement were investigated. The DIA techniques were also adopted in the GEC reinforced foundation tests to determine the surface displacement profile, angular distortion, and shear rupture propagation at various reverse fault offsets. The influence of horizontal spacing of GECs on the effectiveness and reinforcing mechanism of GEC reinforced foundations was discussed. Three-dimensional FE analyses were also carried out to explore the deformation behavior of the GECs embedded in the GEC reinforced foundation and the mobilized reinforcement tensile strain developed in the geotextile encasement. For the GRS foundation subjected to normal fault movement, a smooth surface settlement profile was observed. The reinforcement inclusions effectively prevented the shear rupture propagating to the ground surface and also spread the differential settlement to a wider influential zone, resulting in an average reduction of 60% in the fault-induced angular distortion at the ground surface as compared to the unreinforced foundation. Two main reinforcing mechanisms, the tensioned membrane and shear rupture interception effects, were identified. The numerical results of the parametric study showed that reinforcement pullout in the top reinforcement layer could occur in the case of the short reinforcement. Due to the impact of reinforcement pullout, the shear rupture propagated upward and passed through one end of the reinforcements, resulting in the development of a secondary settlement at the ground surface. The numerical results of the parametric study also showed that reinforcement breakage could occur in the cases of the reinforcement with a low ultimate tensile strength. The foundation height and reinforcement stiffness had considerable influence on βmaxandmax. To ensure adequate reinforcement anchorage against significant pullout, the reinforcement length should be longer than the fault influence length at the free field ground surface. For the GEC reinforced foundations subjected to reverse fault movement, the maximum angular distortion at the ground surface was effectively mitigated. A reduction of 15-23.3% in the max values was observed at S/H = 30% as compared to the unreinforced foundation. Two reinforcing mechanisms, the shear rupture diffusion and diversion effects, were identified. Complex reinforcing mechanisms between diversion and diffusion of the shear rupture were observed in the GEC foundations with different horizontal spacing. The GEC reinforced foundation with Sh /dc = 3.3 shows the most significant effect in reducing max values. The results of FE analysis show that the predicted displacement at the surface of the hanging wall for the unreinforced foundation was overestimated. The predicted βmax values were also slightly underestimated because of the less localized surface displacement occurring at the outcrop. The predicted and measured βmax values for the GEC reinforced foundations are in good agreement, except that the βmax at S/H = 15% was slightly underestimated in FE analyses. The GEC group in the hanging wall remained vertical and undisturbed as reverse fault displaced, while considerable tilting deformation was observed in the GEC group in the footwall. The locations of the maximum tensile strain for each GEC were approximately on the propagation path of the shear rupture. Reinforcement breakage may occur in the geotextile encasement of the GEC closest to the fault tip due to the high lateral earth pressure acting on the geotextile encasement, which leads to strong soil–reinforcement interaction and higher εmax values.