Two finite element approaches for the dynamic analysis of seepage flows are discussed. This represents the first step of a study on the effects of earthquakes on retaining or embedded structures in saturated granular soils. The equations governing the flow of a liquid within a porous skeleton under an acceleration field varying with time are recalled first. Then they are combined in two differential equations that, reduced to their weak form, lead to a finite formulation of the problem in terms of discharge velocity. Due to the relatively large number of nodal variables, and to the iterative structure of the time integration algorithm, this approach requires a non negligible computational cost. Then a second approach is presented, based on some simplifying assumptions, where the pore pressure represents the nodal variable and that adopts a direct time integration scheme. The finite element programs implementing the two formulations are finally applied to the solution of a bench mark problem presented in the literature. The numerical results permit drawing some conclusions on the accuracy of the two approaches that will guide in the choice of the most convenient of them in extending the study towards the analysis of coupled two-phase problems.
Finite element approaches for the dynamic analysis of seepage
CIVIDINI, ANNAMARIA;GIODA, GIANCARLO
2010-01-01
Abstract
Two finite element approaches for the dynamic analysis of seepage flows are discussed. This represents the first step of a study on the effects of earthquakes on retaining or embedded structures in saturated granular soils. The equations governing the flow of a liquid within a porous skeleton under an acceleration field varying with time are recalled first. Then they are combined in two differential equations that, reduced to their weak form, lead to a finite formulation of the problem in terms of discharge velocity. Due to the relatively large number of nodal variables, and to the iterative structure of the time integration algorithm, this approach requires a non negligible computational cost. Then a second approach is presented, based on some simplifying assumptions, where the pore pressure represents the nodal variable and that adopts a direct time integration scheme. The finite element programs implementing the two formulations are finally applied to the solution of a bench mark problem presented in the literature. The numerical results permit drawing some conclusions on the accuracy of the two approaches that will guide in the choice of the most convenient of them in extending the study towards the analysis of coupled two-phase problems.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
