PSI - Issue 28

Abdoullah Namdar / Procedia Structural Integrity 28 (2020) 301–310 Author name / Structural Integrity Procedia 00 (2019) 000–000

302

2

1. Introduction The quality of the boundary condition of the modeled a geo-structure is one of the key factors in the numerical simulation and nonlinear analysis, and it significantly supports the accuracy of the seismic analysis. The accuracy level of the boundary condition of a geo-structure model is required to be close to the real environment for the prediction of the damage and collapse of a geo-structure.

Nomenclature E

Modulus elasticity Poisson’s ratio Unit weight Cohesion Friction angle Dilatancy angle

ν γ

C ϕ ψ

The linear regression analysis was applied in geotechnical engineering for the application of mixed soil in enhancement quality of the concrete footing design. And the statistical model was showed the prediction of the width of concrete footing with high accuracy economically supports geotechnical engineering projects (Namdar, 2020a). It has been reported that the application of the artificial neural network (ANN) for the forecast of soil foundation load response in considering the mechanical properties of the soil. Each element of the mechanical properties of the soil in the prediction of the safe bearing capacity of soil were analyzed. The R 2 and RMSE were used for interpretation of the safe bearing capacity of the soil based on the statistical modeling (Namdar, 2020b). The evaluation of fatigue mechanisms, strain energy density, damaging occurrences, cracks development, ductile of the materials, behavior of materials mechanical properties, stress analysis, structural seismic response, soil-structure seismic interaction, the failure mechanism of the soil foundation, settlement progress, etc. need the prediction process for enhancement analysis quality (Bellini et al., 2019; Berto and Lazzarin, 2009; Di Cocco et al., 2014; Iacoviello et al., 2008; Iacoviello et al., 2013a; Iacoviello et al., 2019; Iacoviello et al., 2014; Iacoviello, et al., 2013b; Johanson et al., 2018; Lazzarin and Berto, 2005; Lazzarin and Berto, 2010; Namdar et al., 2016; Namdar et al., 2019; Namdar, 2020c; Namdar, 2020d; Namdar, 2020e; Oplt et al., 2018; Viespoli et al., 2019). The soil load response is nonlinear, and it occurs because of the variation of the soil mechanical properties, soil mineralogy, soil morphology, the geometry and magnitude of the applied surcharge, and other type of loadings were applied on the soil foundation. On the other hand, with attention to the function of most finite element software, it is complicated to apply the seismic loading on a modeled the geo-structure. The limitation of using finite element software in the nonlinear seismic analysis is related to simulation of a boundary condition which requires to be remarkably close to the real environment. In this study, an attempt was made to develop a boundary condition for modeled geo-structure, and the seismic loading was subjected to the model in a single direction and multidirectional. To control the quality of the numerical simulation and to the predict seismic response of the geo-structure, the statistical analysis was applied to the assessment of stress-strain at considering the several stages of the numerical simulation. 2. Simulation and materials Figure 1 shows the geometry of the embankment-subsoil model was simulated and used in the nonlinear numerical simulation. The height of 18 (m), the width of 180 (m), and thickness 60 (m) are the geometry of the subsoil. The embankment is modeled with an equal thickness to subsoil and height of 18 (m) and 36 (m) crest. This big modeled the embankment-subsoil supports in the real environment to simulate an acceptable boundary condition for the nonlinear numerical analysis. Figure 1 shows the embankment-subsoil is symmetric, this shape of embankment-subsoil minimizes the eccentric gravity load. The simulate boundary condition in the initial stage of the numerical simulation need be accurate. When the seismic loading is applying on the embankment-subsoil model and the model is subjected to the differential displacement and nonlinear deformation the results of the numerical simulation exhibit with higher precision