An earthquake can potentially damage pipelines lying near a fault, thus disrupting the water supply system. Because of this, the structural behavior of buried pipelines under the effects of large fault movements is a topic gaining increased attention from engineers and researchers. O’Rourke et al. addressed large-scale testing of pipeline responses to earthquake-induced ground ruptures and a pipeline system’s performance after an earthquake. Vazouras et al.presented a rigorous finite element method (FEM) model to investigate the mechanical behavior of buried steel pipelines under strike-slip faults. The interacting soil–pipeline system is modeled by finite elements, which can account for large strains and displacements; nonlinear materials; andspecial contacts at the soil–pipeline interface. While large-scale experiments, have high costs and are time consuming to conduct, small-scale experiments are relatively convenient to set up and conduct as pre-experiments to large-scale experiments.
In this study, small-scale experiments were conducted in the National Center for Research on Earthquake Engineering (NCREE) laboratory. Strike-slip fault movements were applied using the shaking table. The selection of pipelines was limited by material and size due to the scale of the experiments. The axial and bending strains of the pipelines and the fault movement displacements were measured during the experiments. The responses of buried pipelines subjected to large oblique-slip fault movements were investigated using finite element analysis (with ABAQUS 2011). Geometric and material nonlinearities were taken into account in the simulation.
Fig. 1 shows one of the shear boxes placed on a small shaking table with a small hydraulic structural actuator (with an allowed maximum force of 15 kN), which can provide a one-way stroke displacement of 14 cm. Another shear box was set up on a small platform that could be adjusted to the same height as the shear box on the shaking table. The two shear boxes contacted closely to prevent sand from leaking, but could also move freely under fault displacements. The shear boxes were initially offset to allow for a maximum soil pressure acting on the pipelines, as shown as Fig. 2.
All tests were conducted with partially saturated silica sand imported from Vietnam. During production, a full test setup was completed by the following steps: 1) pipeline installation in the test basin, 2) structure adjustment in the boxes, 3) soil placement and compaction, 4) adding additional weight on top of the soil surface. To allow the ends of the pipeline to freely move and to eliminate the shear box effect at the ends, clay was filled in between the pipeline and the end plate, as shown in Fig. 3. Fig. 4 shows obvious local buckling behavior occurring in the pipeline on both sides of the fault, which are located, respectively, at 20 cm and 16 cm from the fault.
The commercial finite element package software ABAQUS was adopted to numerically simulate the mechanical behavior of the buried pipe, the surrounding soil medium and the soil–pipeline interaction in a rigorous manner, while considering the nonlinear geometry of the soil and the pipe. An elongated prismatic model is considered in Fig. 5, in which the pipeline is embedded in the soil. The eight-node reduced-integration brick elements (type C3D8R) are used to simulate the surrounding soil. The seismic fault plane divides the soil into two equal parts and is considered to be perpendicular to the pipeline axis at the pipeline’s middle section. The pipeline is modeled by four-node reduced-integration shell elements (S4R), which easily express the local buckling behavior of the pipeline. The pipeline axis is assumed to be horizontal and normal to the fault plane.
Fig. 6 depicts the shape of a deformed pipeline with fault displacement of d = 8 cm located in an area near the fault, where a localized deformation at point A indicates local buckling. Due to the skew-symmetry of the problem, a similar local deformation occurs at point B, on the other end of the pipeline.
For example, in a pipeline with a D/t ratio equal to 80, buried in soft sand at a depth of 1.25 m, the relationship between axial strain and distance to the faults under different fault displacements is shown in Fig. 7. The obviously local buckling occurs 1.5 m from the center of the pipeline when the fault movement reached to 0.3 m.
The critical axial strain, defined by Vazouras, is due to the pipe wall developing wrinkling, associated with significant tensile strains at the ridge of the buckle, so that the compressive strains at this location on the outer surface of the pipe wall start decreasing, as shown in Fig. 8. However, the estimate of the critical axial strain proposed by Vazouras might be too conservative, because when the pipe body reaches the critical axial strain, the pipeline may still retain its functionality.
to the fault under different fault displacements
By conducting small-scale soil–pipeline interaction tests, the structural behavior of buried pipelines crossing strike-slip faults was investigated. Using commercial FEM software, the effects of shear box sizes, soil material properties, and boundary conditions, which might affect the behavior of pipelines, can be considered. In addition, the parameters of the numerical models can be calibrated according to experimental data.
No comments:
Post a Comment