عنوان مقاله [English]
High flexibility and stability of reinforced soil walls make them very useful structures and cause their extended applications as retaining structures at side embankments of roads, slopes, and as abutments especially at regions with high seismic risk. So, such structures could be recommended for Iran as a country with high seismic risk. Reinforced soil structures may present significant deformations under strong earthquake motions. In this regard, they will not provide expected functionality. Therefore, necessity for extension in application of reinforced soil walls especially high geosynthetic reinforced soil walls and significance of plastic displacement in these structures motivated researchers to give special attention to prediction of reinforced soil walls displacements experimentally and numerically.
Two techniques are available for numerical simulation of the reinforced soil masses. In the first method, the soil and inclusion are considered separately in a layered or discrete analysis. This procedure is very time consuming. The second approach is a homogenization method in which reinforced soil is replaced with an equivalent homogeneous but anisotropic medium. Layer-by-layer modeling is not needed in the homogenization methods, so modifying the arrangement of inclusions is easy. Two-phase model is an extension of classical homogenization methods and was developed in the recent two decays.
This approach is actually a mechanical framework based on the virtual work method. It is a macroscopic description of a composite medium, which is the superposition of individual continuous media (phases). The matrix phase (soil) and reinforcement phase (inclusion) are geometrically coincident at any given point in the multiphase material. The proposed model introduces a two-phase model to simulate the nonlinear dynamic behavior of geosynthetic reinforced soil walls. A modified generalized plasticity model for granular materials was used in the proposed two phase model.
The approach was validated by comparison of the results with those of eight reduced-scale reinforced soil walls subjected to seismic loading in shaking tables. Predicted lateral displacement showed good agreement with the test results. The two phase model predicted critical acceleration amplitudes similar to that observed in the experiments. The predicted potential failure surfaces in the two phase model were consistent with the observed deformation patterns.