以有效應力參數評估含細粒料砂土不排水之力學行為

Abstract

現今面對水位狀態不易描述之邊坡或深開挖工程時，以不排水狀態之剪力強度進行設計，結果趨向保守且便利，故目前已成為大地工程界最倚賴的強度指標。但相較於有效應力參數，可具體以摩擦力表示其力學行為，不排水剪力強度僅能表示土壤在不排水狀態下之承載能力，並不具物理意義，因此本研究先藉由三軸試驗獲得砂土之有效應力參數，在透過有限元素法軟體PLAXIS 2D模擬土壤不排水試驗，希望藉此找出不排水強度與有效應力參數之關係，並定義其力學現象。 高細粒料含量砂土為最常釀成液化災害之土壤，對於含細粒料砂土性質與強度有其研究之重要性。因此本研究採用粒徑尺寸單ㄧ之粗細顆粒混合砂土，作為試驗土樣，以細粒料含量、排水條件作為試驗變因，並加入多段式試驗與解壓-再壓試驗，探討細粒料含量對砂土排水與不排水行為之影響，以及檢核多段式試驗對本研究土樣之適用性。數值模擬方面，先透過三軸試驗獲得模擬所需之有效應力參數，再配合排水試驗曲線進行參數優化，以優化後之有效應力參數，模擬砂土不排水行為，最後再將模擬結果與真實試驗作比較，探討有效應力參數與不排水強度之關係。 試驗結果顯示：砂土之細粒料含量是影響三軸試驗結果重要的試驗變因，尖峰強度、孔隙水壓激發量、彈性模數、摩擦角等皆受其影響。多段式試驗在本研究中，試驗強度均低於傳統試驗，推測是受到相對密度與土壤產生剪脹行為所影響，導致結果不如預期。數值模擬結果顯示：在排水試驗模擬中，硬化土壤模式相較於莫爾-庫倫模式，更能反應出真實土壤行為。以硬化土壤模式預測易剪脹土壤不排水行為時，建議將m與Rf參數皆設定為0，如此較能反應出因正負孔隙水壓變化，所導致不同加載斜率之現象。透過本研究也證實了，要在缺少不排水試驗曲線下，預測砂土不排水行為，目前還無法做到。

Nowadays, in the face of slopes or deep excavation projects where the water level is not easy to describe, the shear strength of the undrained state is used for design. The result tends to be conservative and convenient. Therefore, it has become the most important strength indicator for the geotechnical engineering community. However, compared with the effective stress parameters, the mechanical behavior can be expressed in terms of friction. The undrained shear strength can only represent the carrying capacity of the soil under undrained conditions, and has no physical meaning. Therefore, this study was first conducted by the triaxial test. The effective stress parameters of sand were obtained and the soil undrained test was simulated by the finite element software PLAXIS 2D. It is hoped that the relationship between the undrained strength and the effective stress parameters can be found and the physical phenomena can be defined. High fines content sand is the soil most commonly responsible for liquefaction, and it is important for the study of the properties and strength of sand containing fine-grained . Therefore, in this study, coarse-grained mixed sand with particle size was used as the test soil sample, and the fines content and drainage conditions were used as test variables, and the multistage triaxial test and unloading-reloading test were added. To investigate the effect of fines content on drainage and undrained behavior of sand and the applicability of the multistage triaxial test to the soil sample. In numerical simulation, the effective stress parameters required for the simulation are obtained through the triaxial test, and the effective stress parameters are optimized with the drainage test curve. Based on the optimized effective stress parameters, the undrained behavior of the sand was simulated. Finally, the simulation results were compared with the actual test to explore the relationship between the effective stress parameters and the undrained strength. The test results show that the fines content of sand is an important experimental factor affecting the triaxial test results, and the peak strength, pore water pressure excitation amount, elastic modulus and friction angle are all affected. Multistage triaxial test in this study, the test intensity was lower than the traditional test, which was supposed to be affected by the relative density and the dilatancy behavior of the soil, resulting in less than expected results. The numerical simulation results show that in the simulation of drainage test, the hardening soil model is more reflective of the true soil behavior than the Mohr-Coulomb model. When predicting the undrained behavior of the easily dilatable soil by the hardening soil model, it is recommended to set the m and Rf parameters to 0, which is more likely to reflect the phenomenon of different loading slopes due to pore water pressure changes. It has also been confirmed through this study that it is impossible to predict the undrained behavior of sand in the absence of an undrained test curve.