PSI - Issue 39

Abdoullah Namdar / Procedia Structural Integrity 39 (2022) 47–56 Author name / Structural Integrity Procedia 00 (2019) 000–000

49

3

model, the crack on soil is controlled. To verifying the results of the numerical simulation, the statistical model was developed. 2. Crack mechanism simulation The seismic load acts to infrastructure in form nonlinear load in multi-direction, the important parts of the seismic loadings on soil tensile strength is X, Y, and Z seismic load components. Concerning to literature (Namdar, 2021b), in the numerical simulation X, Y, and Z, components of the seismic load need to apply to the model for the numerical simulation. Namdar, 2021b, was indicated the model is subjecting to the seismic loading in the X, Y, and Z directions, has higher results accuracy compared model that is only subjecting to seismic load at a single direction. Figure 1 illustrates the embankment-subsoil model. Figure 1 shows the scaled embankment-subsoil model. The model has medium size embankment. The height of 30 (m), the width of 160 (m), and the thickness of 90 (m) are the geometry of the subsoil. The embankment is modeled with an equal thickness to subsoil and height of 20 (m) and 20 (m) crest. Both embankment and subsoil are unsaturated. According to Figure 1, the geogrid is installing under the toe of the embankment at both sides of the subsoil. With using geogrid, the subsoil considering as a reinforced soil. It assumed the crack is developing on subsoil, under the toe of the embankment. The geogrid is supposing to minimize the crack extension at the subsoil. To analyze the objective of the present study the reinforced and unreinforced subsoil modeled.

(20 (m)

50

20 (m) Embankment

20 (m)

40

The model is unsaturated

30

Subsoil

20

30 (m)

Geogrid

10 Elevation (m)

0

0

10

20

30

40

50

60

70

80

90

100

110

120

130

140

150

160

Length (m)

Fig. 1. The model is using in numerical simulation Figure 2 illustrates the acceleration data relates to the earthquake of 6.2 magnitudes at the Norcia. The seismic data was collecting from the Amatrice station and is using in the numerical simulation. At the numerical simulations X, Y, and Z, components of the seismic load are applying to the embankment-subsoil model. The seismic load is applying to the base of the model. In addition, the impact of the pore water pressure on the model seismic response is not considered. Figure 2 shows the critical part of earthquake data are using in the numerical simulation. From 2 seconds to 6 seconds of the earthquake is maximum acceleration fluctuation, the peaks accelerations are happening in this range, this phenomenon occurs at three directions seismic loading. The remaining parts of the recorded seismic data exhibit low magnitude and fluctuation. Because of this phenomenon, only from 2 seconds to 6 seconds of the earthquake data is using in the numerical simulation.