Seismic risk analysis of wind turbine support structures - Seismic Risk Analysis and Management of Civil Infrastructure Systems - page 722

the soil mass which is in contact with the pile is considered to be the 'slave
surface'. The 'small sliding' tracking approach is employed for the contact
of the two bodies assuming that even if the two bodies undergo large
motions, there is relatively little sliding of one surface along the other. An
elastic-plastic Coulomb model is also used to describe the nonlinear behav-
ior of the soil-pile contact.
A model using the p-y curves is implemented for estimating the pile
lateral stiffness, to be then used in correction terms. The p-y method uses
an elastic beam column member to model the pile and nonlinear horizontal
springs to represent the soil reactions. The p-y curves describe the nonlinear
behavior of the soil springs. They were originally proposed by Matlock
(1970) for soft clays under the water table and models for hard clays and
sands were shortly after introduced by Reese et al. (1975). In this work the
sand, soft clay, and hard clay models are used. The clay model requires the
specifi cation of a parameter ε

50 that is assumed to be 0.005, as recommended

by Reese et al. (1975).

Equality and lower bound data

Following Gardoni et al. (2002) and Ramamoorthy et al. (2006), the data
from the virtual experiments are divided into equality and lower bound
data. An equality datum is such that the quantity of interest (deformation,
shear force, or bending moment) is recorded in the 3D fi nite element analy-
sis and the recorded value is believed to be accurate. A lower bound datum
is such that an accurate record of the quantity of interest is not available
and only a value that is a lower bound to the true value is available.
Specifi cally, when the displacements in the fi nite element analyses are
large, the results might not be accurate and it is found that they are sensitive
to how the solution method handles large displacements and second order
effects. Therefore, we consider a threshold for drift of 5%, such that if the
maximum drift during one time history analysis is less than 5%, then the
deformation, shear, and moment data are considered as equality data. If an
analysis produces a drift that exceeds 5%, then we consider the 5% as a
lower bound datum for the deformation, and the maximum shear and
moment that occurred prior to reaching the 5% drift as the lower bound
data for the shear and moment, respectively. With this approach, we include
the data from analyses that lead to large deformations without letting inac-
curate values wrongfully infl uence the model parameters.

26.3.4 Probabilistic demand models

A virtual experiment database is generated (as described in Section 26.3.3)
to estimate the unknown model parameters of the probabilistic models