Machine Translated by Google

COMPARATIVE NONLINEAR ANALYSIS OF PILE-SOIL INTERACTION AB 2D FRAME*

COMPÿRÿTIVE NONLINEÿR ÿNÿLYSIS OF A RC 2D FRÿME SOIL-PILE

INTERÿCTION*

ORIGINAL SCIENTIFIC PAPER

Boris FOLIÿ ORIGINAL SCIENTIFIC PAPER UDC:
Radomir FOLIÿ 624.154.072.332 doi:10.5937/
GRMK1801063F

1 INTRODUCTION 1 INTRODUCTION

During seismic analysis, pinching at the base is usually assumed, During seismic analysis of a structure, it is assumed that the building
and the flexibility of the soil and foundation is neglected. However, for is clamped at the base and the soil and foundation flexibility is ignored.
more accurate seismic analyses, it is necessary to include the foundation Yet, for more accurate seismic analyses, in addition to the building
and soil in the calculation in addition to the building structure, which structure it is necessary to introduce foundations and soil, which requires
requires the entry of the entire system as input data. There are special entering of the entire system as input data. In the process, special
difficulties when entering data on soil characteristics. In some works, difficulties arise when entering data of the soil characteristics. In some
special hysteresis dependencies and nonlinear response of a system papers, special hysteresis dependencies and non linear response of the
with one degree of freedom (SDOF) are used as a representative of the system with one degree of freedom (SDOF system) are used for
building structure, where the analysis is easier with the introduction of representing of the building structure, whereby an analysis which
foundation-soil flexibility and their influence on the response of the introduces the foundation-soil flexibility and their impact on the structural
structure. Generally, it is considered that the introduction of interaction response is easier to perform. It is generally considered that introduction
reduces the response of the structure, and thus the damage. However, in of interaction reduces the structural response, and thus damage. However,
some cases, there may be negative effects, which was discussed in the in some cases, negative effects may occur, which was discussed in the
paper [7]. Some studies [7] have shown that simplified soil models can paper [7]. Some research, e.g. [7], showed that in this seismic analysis,
be introduced in seismic analysis and significantly facilitate the calculation simplified soil models could be introduced thus making system design
of systems, especially regular buildings [4] and [9]. By comparing the considerably easier, especially design of regular buildings, see [4] and
results of drifts, on 2D and 3D frames, obtained by pushover analysis [9]. By comparing results of drifts of 2D and 3D frames, obtained by a
with the results obtained by applying time history analysis in [8], it was pushover analysis, with results obtained using the time history analysis,
shown that by introducing modification factors, conservative results see [8], it is showed that by introducing the modification factors, one can
applicable in design practice can be obtained. Also in the paper [22], the obtain conservative results applicable in designing practice. In the paper
approximate calculation of rigid frames with the introduction of soil- [22], an approximate design of rigid frames, applicable in designing
foundation-structure interaction, applicable in design practice, is presented. practice, with interaction of soil-
Analytical solutions are much rarer, although there have been attempts
[20].

Dr. Boris Foliÿ, University of Belgrade, Innovative Center of the Dr. Boris Folic, University of Belgrade, Innovation Center of Faculty of
Faculty of Mechanical Engineering Kraljice Marije 16, Mechanical Engineering, Kraljice Marije 16, Belgrade,
Belgrade boris.folic@gmail.com Radomir Folic,
boris.folic@gmail.com Radomir Foliÿ, University of Novi Sad, University of Novi Sad, Faculty of Technical Sciences, Trg Dositeja
Faculty of Technical Sciences, Trg Dositeja Obradoviÿa 6, Novi Obradovica 6, Novi Sad, folic@uns.ac.rs
Sad folic@uns.ac.rs

*
* We dedicate this paper, with respect, to Academician Dušan This paper is dedicated, with respect, to academician Dusan
Miloviÿ Milovic

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89) 63

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

The European regulation for seismic design EN 1998, Part 1 and foundation-structure, was presented. Analytical solutions are rare, even
Part 5, does not consider in detail the problem of introducing soil- though there were attempts, [20].
foundation-structure interaction (SFI) in numerical seismic analyses. In European regulation for seismic design, EN 1998, Part 1 and Part 5,
EN 1998-5 this is required where P-ÿ effects play a major role; does not consider in detail the problem of introducing the soil-foundation-
constructions with massive and deep foundations, and constructions on structure interaction (SFI) in the numerical seismic analyses. In EN 1998-5
very soft ground where the average speed of shear waves is less than it is required where P-ÿ effects have an important role: the structures with
100 m/s [5] and [10]. The differences in the seismic behavior of objects massive and deep foundations, and structures in a very soft soil where
with shallow foundations and on piles is described in detail in [4] and [5]. the average velocity of shear waves is less than 100 m/s, see [5] and
They describe in detail the method of analyzing kinematic and inertial [10]. The difference of seismic behaviour of the structures founded on
interaction during foundation on piles. The kinematic interaction originates shallow foundations and on the piles was described in detail in [4] and
from the difference in the movement of the soil and the foundation or [5]. In them, the method of analysis of kinematic and inertial interaction of
piles, during an earthquake, where mass is neglected. In inertial pile foundations was described in detail.
interaction, the influence of inertial forces from the structure to the
foundation is provided.

ÿ comparative non-linear static analysis (NSÿ), often called a

In this paper, a comparative nonlinear static (NSA), often called pushover analysis, as well as the dynamic non-linear analysis (NDÿ),
pushover analysis, and a dynamic NDA analysis, described in detail in described in detail in [3], and applied on 2D frames of RC skeletal
[3], were performed on a 2D frame AB skeletal building founded on piles. buildings founded on piles are presented in this paper. The model involves
The model also includes linear-nonlinear pile-soil interactions using link a linear-non-linear pile-soil interaction, using link elements. The soil is
elements. The soil is modeled with multiple (linear) plastic connecting modelled using multiple (linear) plastic link elements, as envelopes in the
elements, as envelopes in the form of py curves, on both sides of the pile. form of the p-y curves, on both sides of the pile. P-y curves are transferring
Py curves transmit (receive) pressure only. The py curves were modeled only compression and are modelled according to Cox, Reese and Matlock
according to Cox, Rees and Matlock [4] and [12] and [19] for submerged [4], [12] and [19] for submerged sand, and piles with a diameter 60 cm.
sand, and 60 cm diameter piles. The seismic performance of the structure-
foundation-soil system of a 2D frame founded on piles was analyzed. The
partially linear and somewhat more detailed nonlinear behavior of the roof
beam is shown, while the other frame beams behave nonlinearly. A The seismic performances of the structure-foundation-soil system of a
linear roof beam is one in which plastic joints are not introduced in the 2D frame founded on piles are analyzed.
nodes at the intersection with the internal columns, but also in the field of Detailed analysis of hierarchy of formation of plastic hinges in the frame
the roof beam. and piles is presented. A partial linear and a more detailed non-linear
behaviour of a roof beam are presented, while the remaining beams of
the frame behave non-linearly. Linear roof beam is the beam without
plastic hinges at intersections with inner columns or along the beam.
Kinematic interaction arises from different motions of the soil and
foundation, or piles, during earthquake, while the mass is neglected. In
inertial interaction, the effect of inertial forces from the structure upon the
foundation is considered.

2 DESCRIPTION OF THE CONSTRUCTION, FOUNDATIONS AND METHODS 2 DESCRIPTION OF THE STRUCTURE,

ANALYSIS FOUNDATIONS AND METHODS OF ANALYSIS

2.1 Analyzed structure 2.1 Analyzed structure

The facade frame, which has four columns, as well as the inner A façade frame with four columns and an inner frame are analyzed.
frame, were analyzed. There are corner posts and edge posts on the On the façade frame, there are corner columns and peripheral columns.
facade frame. The corner pillars of the facade frame are founded on a The corner columns of the facade frame are founded using a group of 3
group of 3 piles, and the inner pillars on a group of four piles each. piles, whereas the inner columns are founded on a group of four piles.
The facade frame is "condensed", so that all pile elements are inserted The façade frame is “condensed” by inserting all pile elements via
through a projection perpendicular to the middle plane of the frame. In projection along the direction perpendicular to the frame middle plane. In
this way, the frame model can be displayed in only one plane. A group of this way, it is possible to draw the frame model using only two dimensions.
3 circular piles has a part consisting of one pile (1D60), and a part The group of 3 circular piles consists of a part made of one pile (1D60),
consisting of two condensed piles (2D60) picture 1. So in the "condensed" and another part made of two condensed piles (2D60), figure 1. Hence,
model of three piles only two piles are entered, 1 is independent pile, and in this “condensed” model, only two out of three piles are introduced, one
the other is a double pile (ie, in the planar model, one pile is entered, of which is an individual pile, whereas the other is a double pile (i.e. a
where the cross-sections of the FRAME element, in the SAP2000 single pile was introduced to the model, whose Frame element cross-
program, in the Set Modifiers section, stiffness and mass are multiplied section, stiffness and mass were multiplied by 2), in SÿP 2000 software,
by 2). Accordingly, the py curves of the "double" pile have twice the within the Set Modifiers module. In accordance with this, the p-y curves
stiffness. The building has two relatively slim lower floors. The height of of the “double” pile
the first two floors is 5 meters each, but that's why the cross sections of
the columns are on these two

64 BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

floors 85/85 cm. also have the double value of stiffness. The building has two relatively
In the seismic analysis, nonlinear static analysis (NSA), often called slender lower stories. The height of the first two storeys is 5 meters each,
pushover, and nonlinear dynamic analysis (NDA), in the time domain but the cross-sections of columns in these two floors are 85/85 cm.
(Time History), were applied. Given that during NSA and NDA [3], plastic
joints in piles practically do not appear in stiff soil types (except in special Seismic analysis is performed using the non-linear static analysis
cases), the effects in the soil, through link elements in depth, under (NSA), often called the pushover analysis, and the non-linear dynamic
seismic action were also analyzed (TH El Centro 0 , 30g). analysis (NDA) in the time domain (time history). Regarding that during
NSA and NDA [3] there is practically no occurrence of plastic hinges in
the piles for stiffer types of soil (except in special cases) the effects in the
They are not sufficiently addressed in the available literature in this way. soil are analyzed too, via link elements along the depth, under a seismic
Extreme displacements, forces, total work, "current work" and their action (TH El Centro 0.30g). They are not sufficiently discussed in the
distribution by depth of pile link elements were analyzed. Only one pile available literature in this way. Extreme displacements, forces, total work,
from the group was analyzed, namely the independent pile (1D60) from “instantaneous work” in link elements and their distribution along depth of
the group of three piles (1+2, in Figure 1). It is the end pile on both sides the pile are analyzed. Only one pile from the group is analyzed, namely,
of the symmetrical frame (see Figure 1 right). Specifically, in this paper, a single pile (1D60) from the group of three piles (1+2, in figure1). These
only the leftmost pile was investigated. Also analyzed was the change in are the end piles on both sides of the symmetrical frame (see figure 1
the total shear force in the foundation with the increase in peak right).
acceleration (PGA), and the change in the condition of the plastic joints
in the structure and piles, and in accordance with this change in the
structural system, the change in the first and second characteristic tone
after the end of the seismic action (Time History). In this paper specifically, only the left endmost pile is studied. Also, the
variation of the total seismic base-shear force, with the increase of peak
acceleration (PGA) is analyzed, and also the variation of condition of
plastic hinges in the structure and the piles, and variation of the first and
second natural tones after seismic excitation (Time History)

Figure 1. The principle of "condensation" of a group of 3 piles into a group of 2 piles in a plane. 1D60 is an independent pile, and 2D60 is a
double pile. Figure 1. "Condensation" principle of a group of 3 piles (1D60 – individual pile, 2D60 – double pile).

The spatial (3D) frame is dimensioned for earthquake action in the The spatial (3D) frame is dimensioned with reference to earthquake
SAP 2000 v14 program. with the introduction of vertical direction and action, using SÿP 2000 v14 software, including the effects in the
torsion (with 5% eccentricity), for a behavior factor of 5.85. After that, the perpendicular direction and torsion (with 5% eccentricity), for a behaviour
previously described facade 2D frame with the associated load was factor of 5.85. The previously described façade 2D frame with its
separated from the dimensioned model. The span of the frames is 8m, corresponding loads is then taken out of a 3D model dimensioned in this
and this is also the axial distance of the pillars, in both directions. The way. The span between frames is 8m, which is also the distance between
object is biaxially symmetrical. The height of the first two floors is 5 m the pile axes, in both directions, since the structure in question is
each, and the other 6 floors are 3.1 m. The model is similar to the models symmetrical along two orthogonal axes. The height of the first two stories
given in [2] and [4]. The difference is in the py curves that are given in the is 5m, while for the remaining 6 storeys, it is 3.1 m. The model is similar
mentioned paper [4] for piles with a diameter of 1.2 m. Also, in the to models shown in [2] and [4]. However, the difference is in p-y curves,
mentioned paper, several different models are given, with and without which are, in [4], given for piles of diameter 1.2m. Also, more different
pile-soil interaction. The appearance of the frame adopted here, of only models with and without pile-soil interaction are given in [4]. The geometry
one model, is given in the continuation of the paper, when analyzing the of the single frame considered here is presented later in the section where
state of the plastic joints. In [2], for the constructed building, the pillars the state of plastic hinges is analysed.
are all 60/60 cm in cross-section, and the load is somewhat less.

Separating a 2D frame from a three-dimensional (3D) frame is

accompanied by specific problems [4] and [8]. The first parameter is Paper [2] is related to built structure, where all columns

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89) 65

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

the associated load, the second is the geometry of the cross-sections (first are 60/60cm, and the loading is slightly lower.
of all) of the beams, and for dynamic analysis, it is necessary to check the Extraction of a 2D frame from a three-dimensional (3D) frame is
difference in the characteristic periods of the separated 2D frame accompanied by specific problems, [4] and [8]. The first parameter is the
compared to the 3D model. It is desirable to also check the deformation corresponding load; the second is the geometry of cross sections
("bend") of the top of the 3D object in relation to the top of the 2D frame. (primarily) of the beams, while for a dynamic analysis, it is necessary to
A more realistic behavior of planar models (in which only one 2D frame is check the difference between the natural periods of the extracted 2D
selected), compared to a spatial frame, is given by models in which frame and the 3D model. It is desirable to check and compare the
several frames in the 2D model are connected by link elements or simple deformation (horizontal deflection) of the top of a 3D object in comparison
rods [11]. As for example coupled facade and internal 2D frame or 2 to the top of the 2D frame. A more realistic behaviour of the planar models
internal and 2 facade, with the condition that the coupled frames in the 3D (where only one 2D frame is extracted from 3D model), when compared
model must extend in the same direction. with the spatial frame, is obtained by 2D models where more 2D frames
are combined in one model, using link or truss elements, [11]. For
When extracting the 2D frame, it is necessary to check the acceptable instance, coupled façade and the interior 2D frame, or 2 interior and 2
matching of the values of the normal forces in the columns, and the façade frames, are examples of that. Of course, it is assumed that frames
reaction of the supports for both models. Also, at nodes where beams are aligned along the same direction.
from the other direction meet, the Set Modifiers command can be applied
to increase the stiffness of the cross-sections stressed to bending, due to
the participation of the torsional stiffness of the other direction.

In the case of (small) cross-sections of beams and thinner plates, where When extracting a 2D frame, it is necessary to check the acceptable
the torsional stiffness quickly decreases/degrades during an earthquake, agreement of values of normal forces in the columns, and the support
it is better to introduce a non-linear rotational spring, which simulates this reactions of both models.
stiffness, but also its decrease during an earthquake. Often in seismic Also, in a case of nodes where the beams from another direction are
(dynamic and quasi-static) analysis, due to the reduction of the number of connected, one might use the command Set Modifiers, in order to increase
unknowns, slabs are excluded, and gross sections of beams are used, the bending stiffness of sections, due to influence of the torsional stiffness
even without associated active slab widths, and without the Modifiers in another direction. In the case of (relatively small) cross-sections of
command. As it is on the side of safety, it was also applied in this work, beams and thin slabs, where the torsional stiffness quickly decreases/
for facade beams. The influence of floor slabs, on equalizing the horizontal degrades during earth-quakes, it is better to introduce a non-linear
displacements of floors, in 2D and in 3D frames, can be improved with the rotational spring which simulates this stiffness and its decline during
Joint - Constraints (Equal) command. earthquakes. Very often, in the seismic (dynamic and quasi static)
analysis, due to reduction of a number of unknowns, the slabs are
excluded, and gross cross-sections of beams are used, even without the
The upper nodes (in the roof plane) of the internal columns do not belonging active widths of the slabs, and without the Modifiers command.
meet the requirements of the stiffness ratio of beams and columns, so in Since it is on the safety side, it is imple-mented in this paper, too, for the
the first part of the work, that part of the beam was linearized, i.e. no façade beams. The effects of the floor slabs to unify the horizontal storey
plastic joints were inserted in that part. This will be looked at, later, in the displacements, in 2D and 3D frames, may be improved by using the
part of the analysis of the results. (move comma) command Joint – Constraints (Equal).

If the recommended option of SAP, the automatic division of plastic

joints in line elements of 0.02, is not used, beam breakage occurs on the
second floor quite early, and later in other locations, so the calculation
time is extended. This example is not shown, although this way of breaking Upper nodes (in the roof) of the inner columns do not meet the
is also possible, but less likely, if the construction is well executed. That's conditions of stiffness ratio of beams and columns, so in the first part of
why in all the models shown here, this option is implemented. the paper this upper beam is linearized, i.e. no plastic hinges are
introduced in this beam. It will be considered later, in the section of the
result analysis.
The ratio of the total seismic force in the base and the total force of
the weight of the object, in the (regular, square in base) 3D model, is: Unless the recommended option of SAP, Assign/ Frame/ Frame
5316 kN/71093 kN= 7.48%. The characteristic periods of the 3D model Signed Overwrights/ Auto Subdi-vide Line Objects at Hinge of 0,02 is
are: T1=T2=2.04 sec, T3=1.47 sec. used, the failure of the beams occurs on the second storey quite early,
The first two tones are horizontal-lateral, and the third is torsional-rotational and later in other locations as well, so the time of calculation is prolonged.
[4]. This example is not presented, even though such failure mode is quite
possible, but unlikely, if the structure is well constructed. For this reason
in all the presented models, this option is implemented.

The ratio of the total base-shear force and the total force of the
building weight (of a regular, square layout)
3D model, is: 5316kN/71093kN= 7.48%. Natural periods of the 3D model
are: T1=T2=2.04 sec, T3=1.47 sec. The first two tones are horizontal-
lateral, and the third one is torsional-rotational [4].

66 BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

2.2 Earthquake behavior and modeling 2.2 Soil behaviour and its modelling during earthquakes

The answer to the question of what happens in the soil during an The answer to the question what occurs in the soil during earthquakes
earthquake depends, first of all, on the way the soil is modeled in the depends, primarily on the method of soil modelling in the structure-
foundation-soil construction system, then which earthquake and what foundation-soil system, and then what earthquake and what effect are
impact is specifically being investigated. Probably the best answer to this specifically investigated. The best answer to this question is probably
question is provided by wave mechanics, and its application to numerical provided by the wave mechanics, and its implementation in the numerical
analyzes using solid soil elements. This is a special issue, because in the analyses using Solid elements of soil. It is a specific problem, because in
modeling process, it requires finding the optimum while satisfying often the process of modelling, it requires finding an optimum for satisfying often
conflicting conditions. confronted conditions. The parameters which need to be determined in the
model definition process are: dimensions of the structural model, system
The parameters that should be determined in the process of defining model size, FE mesh size, frequency domain and types of propagating
the model are: the dimensions of the construction model, then the waves, boundary issues, etc. [17]. In this method, it is important to assess
size of the system model, the size of the KE network, the frequency a minimum size of the model for the structure-foundation-soil system, in
range and types of propagating waves, boundary problems, etc. [17]. order to keep the calculation time as short as possible, but still to include
With this method, it is important to estimate the minimum size of the model all the necessary wave phenomena. Also, the maximum size of finite
for the structure-foundation-soil system in relation to the size of the elements of soil must be determined, which must not be exceeded so as
structure, so that the calculation time is as short as possible, while to avoid the undesirable wave effects in the actual finite elements of the
adequately covering all the necessary wave phenomena. It is also soil, etc. We are discussing here the use of link elements in soil (LES), as
necessary to determine the maximum size of the final soil elements, which Takeda envelopes of experimentally dete-rmined p-y curves (ATPY),
must not be exceeded in order to avoid unwanted wave effects in the final because it is a simpler model for practical use, so the response will be
soil elements themselves, etc. Here we will stick to the use of link soil based on it.
elements (LES), as Takeda's envelope of experimentally determined py
curves (ATPY), because it is a simpler model to apply, and the answer will
be based on the same.

2.3 PY curves for piles in sand 2.3 P-Y curves for piles in sand

In the analysis of the dynamic interaction of the structure-foundation- Models of different levels of sophistication can be used for presentation
soil system, the soil can be represented using models of varying degrees of soil in an analysis of the dynamic interaction of a structure-foundation-
of complexity (sophistication). soil system. The usual methods of seismic analysis of non-linear behaviour
Common methods of seismic analysis of nonlinear behavior of structures of structures are the quasi-static pushover NSA analysis and the non-
are quasi-static pushover NSA and nonlinear dynamic analysis in the time linear dynamic NDA analysis in the time domain, as a numerical integration
domain NDA, as numerical integration of accelerograms, so-called. step of accelerogram, a so-called step-by-step time history method. For the
by step method (time history, step by step). For seismic effects, El Centro seismic action, accelerogram of El Centro, for PGA of 0.20; 0.25 and 0.30
accelerograms were used, for PGA 0.20; 0.25 and 0.30 g. g are used.

Soil can be modeled through various support conditions, structures or Soil can be modelled using various conditions of support, structure or
piles, such as: ÿ linear springs with one nodal point piles, such as: ÿ Linear single-node springs,
which equally resist
(spring), which suffer both tension and pressure, tension and compression, ÿ Linear
ÿ linear link elements link elements

ÿ multi-linear plastic link elements, which can be set to transmit only ÿ Multi-linear plastic link elements which can be set so that only
pressure. transfer compression.
The ground is modeled through connection elements, the so-called Soil is modelled using the connection elements, so called link
link elements, according to the py model for sand developed by Ris and elements, according to the p-y sand model developed by Reese et al,
Dr. Reese, Cox, Koop, 1974, and Reese, Sullivan, 1980, cited by [15]. Reese, Cox, Koop, 1974, and Reese, Sullivan, 1980, cited according to
[15] .
According to [13], probably the first model of py curves was introduced According to [13] probably the first model of p-y curves was introduced
by McClelland and Focht (1958), recommending a procedure for correlating by [10], which recommends the procedure for correlation of data of triaxial
triaxial stress-strain test data with pile force-displacement curves for stress-strain test with the force-displacement curves of the piles for certain
certain depths, via the expected soil reaction modulus, for each soil layer, depths, via the expected soil reaction modulus, for every layer of soil, along
by depth. Reese was the first to present his concept of wedge-shaped soil the depth. Reese was the first to present his concept of wedge-like soil
failure, which occurs near the soil surface [19]. The influence of varying failure which occurs close to the soil surface [19]. The influence of variation
the input parameters of the py curves on the pile response can be seen in of input parameters of p-y curves on the pile response can be observed in
[12]. [12].

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89) 67

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

2.4 Pushover NSA – non-linear static analysis 2.4 Pushover NSA - non-lineÿr stÿtic ÿnÿlysis

In the paper [3], modern methods for nonlinear seismic analysis of The paper [3] is analyzing the contemporary methods for non-linear
structures and the way of introducing damping when using some of the seismic analysis of structures, and the ways how damping is introduced
methods are analyzed in detail. when using some of the methods. The pushover (PO) analysis is here
Here is a brief presentation of the pushover (PO) analysis in which the briefly presented, which involves determination of curves which show
dependence curves of the displacement of the control node umax the dependence of control node displacement umax (typically at the top
(usually at the top of the frame) are determined in relation to the seismic of the frame) with the seismic base shear (BS) force, for assumed shape
shear forces at the base (BS-Base Shear), and for the adopted form of of lateral load distribution along the height. It is assumed that the
load distribution along the height of the object. adopted form of load remains unchanged for all intensity levels, along
It is assumed that the adopted load form remains unchanged for all with the structure’s deformed shape. Gradual increase of the load
levels of intensity, and thus the deformed shape of the structure. A intensity is perfor-med in steps, along with the opening of plastic hinges
gradual increase in the intensity of the load is done in steps with the up to a point where the structure becomes a mechanism.
opening of the plastic joints until the structure turns into a mechanism.
In the construction of pushover curves, apart from those required by
regulations, given in EC8, it is desirable to apply several different forms
of load distribution. When constructing pushover curves, the use of several different shapes
of load distributions is recommended, along with the ones prescribed by
The following forms of load distribution by frame height (or 2D building the regulations given in EC8. In this paper, the following shapes of load
model) are applied here: ÿ Constant distribution distributions along the frame height were applied: ÿ Constant distribution
(const). ÿ Linear variable (lin). ÿ (const). ÿ Linear variable (lin). ÿ Proportional to the shape of
Proportional to the shape of the first the first natural mode (1 mode) and ÿ
characteristic tone (1 mode) and ÿ Proportional to the distribution Proportional to the distribution
of of (corresponding) masses (acc).
(belonging) masses (acc).

Different PO display types can also be applied

guilty, and in SAP2000 are available for this: basically (BS) In addition, different types of PO curve displays can be applied, and
1. Resultant force acc in the case of SÿP 2000, the following ones are available: 1. Resulting
observed displacement (MD), 2. ATC base shear force (BS)
40 capacity spectrum method, 3. FEMA 356 according to the observed displacement (MD), 2. ÿTC 40 spectrum
coefficient method, 4. FEMA 440 equivalent capacity method, 3. FEMÿ 356
linearization method, and 5. FEMA 440 displacement coefficients method, 4. FEMÿ 440 equivalent
modification method. linearization method, and 5. FEMÿ 440
displacement modification method.

3 CALCULATION RESULTS AND THEIR ANALYSIS 3 CALCULATION RESULTS AND THEIR ANALYSIS

3.1 NSA results 3.1 NSA results

Here, the PO curves are determined in the SAP2000 v14 program, Here, the PO curves are determined using SÿP 2000 v14 software,
but not via the Display/ Show Static Pushover Curve option, because but not with the Display/Show Static Pushover Curve option, since in
then the diagram is not viewed enough, the value readings are this case the diagram is not visible enough, reading of values from it is
insufficiently precise and appropriate manipulations cannot be performed, insufficiently accurate and appropriate manipulations cannot be
but that is why it was done via the Display path /Show Plot Function, i.e. performed. Thus, the above process is performed using the path Display/
via the diagram of the Umax/ BS function. The PO curve was also Show Plot Function, i.e. by using the function diagram Umax/BS. In
determined according to the FEMA356 procedure. addition, the PO curve is also determined according to the FEMÿ 356
procedure.
On the summary diagram, for this determination of the PO curve, a
significant difference in the maximum displacements of the control node
is visible, depending on the form of the load, as well as the difference in In the summary diagram, for PO curves compared in this way,
the initial initial stiffness. A more detailed analysis is given in Table 1. there is a noticeable difference of maximum control node displacement,
depending on the load shape, along with a difference in initial stiffness.
ÿ more detailed analysis data are given in table 1.

68 BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

PUSHOVER const PUSHOVER lin

1200 900

800
1000
700

800 600

500
600
400

400 300

200
200
100

0 0

0 0,02 0,04 0,06 0,08 0,1 0,12 0,14 0,16 0 0,02 0,04 0,06 0,08 0,1 0,12

Figure 2a. Pushover fault. Constant load distribution Figure 2b. Pushover curve. Linear load distribution
by height. Force at base BS=1069 kN, maximum by height BS=793.1 kN, umax=10.73 cm.
displacement umax=14.97 cm.

Figure 2a. Pushover curve. Constant load shape along Figure 2b. Pushover curve. Linear distributed load shape
height BS=1069 kN, umax=14.97 cm. along height BS=793.1 kN, umax =10.73 cm.

Figure 3. Summary diagram of pushover curves for 4 forms of load distribution: linear, 1 mode, constant (const) and
proportional to the masses acc.

Figure 3. Summary diagram Pushover curves, for 4 shapes load distribution: linear, 1 mode, const. and acc.

Table 1. Comparative presentation of the maximum displacements of the node in the top and forces in the base depending on the form of the load.
In the time analysis depending on the PGA. Linearized roof beam [6].
Table 1. Comparative analysis of max top node displacements and Base Shear, with respect to load shape. In TH with
respect to PGA. Linear roof beam. [6].

Method of vertical load distribution.

PGA (g) El Centro Distribution of vertical load
0.20g 0.25g 0.30g* PO lin PO const PO acc 1312 1615 1899 793.10 1068.65 PO 1 mode
BS (kN) 1492.66 8,56 11,29 14,47 10.73 14.97 23.54 893.87
Umax (cm) 12.83
FEMA 356 C
BS (kN) 798.67 1076.10 1504.40 900.60
Umax (cm) 27.3 26.9 24.3 28.4
*
cut off at 7. 2 sec; FEMA 356 C - Site class C; Pushover= PO

The considered frame is comprehensively treated, considering that The considered frame is comprehensively treated,
before the TH (NDA) analysis, the voltage state was taken and since before the TH (NDA) analysis, the stress state
self-weight constructions, linearized roofs of the structure under its self weight is taken, the
the beams of the considered frame [6] remain straight (picture 4). Code linearized roof beams of the observed frame [6] remain
non-linear ones, on the other hand, are clearly visible straight (figure 4). In the non-linear ones, on the
will be shown in the results analysis section in Figures 27, contrary, there are clearly observable deflections which
28 and 29. Kod analize link elements, nelinearni su i ovi will be presented in the section analysis of results in
nodes, but plastic joints are not included here either figures 27, 28 and 29. In the analysis of link elements,
in the middle of the beams. Frame model with plastic joints the nodes are non-linear, but the plastic hinges in the
and in the middle of the beams, it is shown at the end of the work, only on middle of the beams are not assumed. The model of the

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89) 69

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

Figure 31, with a brief overview of the problem of the same. frame with plastic hinges in the middle of beams as well, is presented at
Nonlinear dynamic analysis was performed for the El Centro the end of the paper, in figure 31 only, with a short comment on the issue.
accelerogram for peak PGA values of 0.20, 0.25 and 0.30 g. The
displacement of the node at the top and the total seismic force at the base Non-linear dynamic analysis is performed for the El Centro ground
were considered. The condition of the plastic joints (breakage) was motion record, for peak PGÿ values 0.20, 0.25 and 0.30 g. Node
checked at the end of each earthquake. displacements at the top and the seismic base shear are considered. The
states of plastic hinges (failure) are checked at the end of each earthquake.

Figure 4. Condition of PHS plastic joints at the site of the El Centro earthquake, left PGA 0.20g PHS: 79Y, 19 IO, right PGA 0.25g PHS:
71Y, 25 IO and 2 LS. Linear roof beam Figure 4. State of plastic hinges (PHS) at the
end earthquake ElCentro, left PGA 0.20g PHS: 79Y, 19 IO, right PGA 0.25g PHS: 71Y, 25 IO and 2 LS. Linear roof beams

Figure 5. Diagram of the displacement of the node at the top of the building during the accelerogram El Centro left PGA 0.20g, Umax=8.56cm, right
PGA 0.25g, Umax =11.29 cm. Linear roof beam.
Figure 5. Displacement plot of the joint at the top of the building, due earthquake acc. El Centro: left PGA 0.20g, Umax =8.56cm, right PGA 0.25g, Umax
=11.29 cm Linear roof beams.

The coefficient of proportionality for both forces and displacements, The proportionality coefficient, both for the forces and for
when crossing a system with several degrees of freedom of movement, displacements, during transition from a mullti degree of freedom system
(MDOF) to a system with one degree of freedom (SDOF), is: (MDOF) to a single degree of freedom system (SDOF) is:

ÿÿÿ
ÿ
T
m 1
ÿ
ÿmi ÿ
i
ÿ
m
*
(1)
T
ÿÿ
2 2
ÿm mi ÿ mi ÿ
i i

70 BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

3.2 Column drifts for different PGA values 3.2 Column drifts for different values of PGA

With non-linear (NL) roof beam and entire NL 2D For the non-linear roof beam, and entire 2D frame,
shoulders, the extreme values of column drift are calculated the extreme values of column drift and their variation
and their change during the action of the accelerogram El during action of accelerogram of El Centro, for PGA of
Centro, for PGA 0.20; 0.25 and 0.30 g. 0.20; 0.25 and 0.30 g, are calculated.

extreme Drift Floor ElCentro 0,20g

10

4
Floor
2

1 1,5 2 2,5 3 3,5 4 4,5

Local Drift (‰)

Figure 6a. Extreme floor drift (pillar) El Centro 0.20g. Exceeds allowed values
Figure 6a. El Centro 0.20g. Extreme Local Drift (column) exceeds permissible values

extreme Drift Floor ElCentro 0,25g

10

4
Floor

1 1,5 2 2,5 3 3,5 4 4,5 5

Local Drift (‰)

Figure 6b Extreme floor drift (pillar) El Centro 0.25g. Exceeds allowed values
Figure 6b El Centro 0.25g. Extreme Local Drift (column) exceeds permissible values

extreme Drift Floor ElCentro 0,30g

10

4
Floor
2

1 1,5 2 2,5 3 3,5 4 4,5 5 5,5 6

Local Drift (‰)

Figure 6c. Extreme floor drift (pillar) El Centro 0.30g Exceeds the permitted values
Figure 6c. El Centro 0.30g. Extreme Local Drift (column) exceeds permissible values

Figure 6c differs in shape compared to 6a and 6b, Figure 6c, is different in shape from Figs. 6a and 6b,
for PGA 0.20 and 0.25g. which are given for PGA 0.20 and 0.25g.

3.3 Py curves 3.3 P-y curves

The Py curve (Figure 7) consists of 4 parts, the first linear from P-y curve (Fig. 7) consists of 4 parts: the first is linear
coordinate origin to point k, the second exponential from the coordinate beginning till the point u, the second
part from k to m, and the third part is another linear function of m is exponential part from k to m, the third is the second
to u, and after the point u, the py curve is a constant straight line. linear function from m to u, while after the point u, p-y
Reduction coefficients (Figure 8) A and B depend on the species line remains constant.
loads, and curves are used for dynamic analysis Reduction coefficients A and B (Fig. 8) depend on
cyclic load. The coefficients A and B are given at the type of the load, and for dynamic analysis, the

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89) 71

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

diagram in the interval from 0 to 6 z/b. After 5 pile diameters, coefficients curves of cyclic load are used. The coefficients A and B are given in
A and B have a constant value. zcr is the depth after which the wedge the graph in the interval 0 to z/b. After 5 pile diameters, coefficients
fracture form changes to a block fracture form. A and B have a constant value. zcr is the depth after which the
wedge-like failure transforms into the block-like failure.

Figure 7. Construction of characteristic forms of py curves, Ris, Koks, Kup et al. 1974, cited according to [4]
Figure 7. Construction of characteristic shapes of p-y curves Reese, Cox, Coop at all. 1974, after [4]

Figure 8. left Reduction coefficients A and B; Image on the right Factors for calculating the limit resistance of the soil for a horizontally
loaded pile in sand C1, C2, C3 and zcr, in relation to the angle of internal friction [15].

Figure 8. left Reduction coefficient A and B; right Factor for calculation of ultimate soil resistance for horizontal loaded pile in section C1,C2,C3
and zcr, related to the friction angle, after [15].

Table 2. Horizontal soil reaction coefficient for sand. The initial slope of the py curve, as a function of relative stiffness and groundwater
level, submerged and dry sand.

Table 2. Coefficient of horizontal reaction for sand. Initial inclination p-y curve vs. relative density and level below ground water (submerged
and dry sand).

Soil modulus k parameter k for the relative degree of sand compaction

Realistic compactness: Rastresit Medium compact Crutch

Relative density: Loose Medium dense Dense

Submerged sand 5.430 kPa/m 16.300 kPa/m 33.900 kPa/m
Submerged sand
Suv, above NPV
6.790 kPa/m 24.430 kPa/m 61.000 kPa/m
Dry sand, above water level

72 BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

Figure 9. Initial modulus of horizontal soil reaction in Figure 10. Characteristic curves for sand with influence
depending on compaction and angle of internal friction, API cohesive part, Rees et al. [4]
(American Petroleum Institute), according to [4]

Figure 9. Initial modul of lateral reaction of soil in function Slika 10. Caracteristic curve for sand with
of compaction and internal angle of friction, API (American influences of cohesive part, Rees at all [4]
Petroleum Institute), after [4]

Marchinson and Oneill, 1984 [15], simplified su Marchinson and Oneill, 1984 [15] simplified the
the above procedure and three parts of the right replaced with one above process, and replaced the three part p-y curve
equation, as follows: with a single analytical equation, as follows:

p ÿ k ÿ
ÿÿ
nA ÿ
fishy
H (2)
and
ÿ ÿ

ÿ ÿ
ÿ ÿ

p in ÿ nAp in ÿ

pu – limit horizontal resistance at depth H of Where:

floor surface, pu – ultimate lateral soil resistance at a depth H
kH – soil stiffness, initial modulus of horizontal reaction below ground surface,
(according to table 2, for sand), kH – soil stiffness, initial modulus of lateral reaction
y – horizontal displacement of the pile, (according to table 2, for sand)
n – geometric factor, =1.0 for prismatic piles, y – lateral displacement of pile
A = 0.9 for cyclic loading, (3-0.8(z/b)) ÿ 0.90 for n – geometric factor, =1.0 for prismatic piles
static opt., z depth for which the p/y curve is determined. A = 0.9 for cyclic load (3-0.8(z/b)) ÿ 0.90; for the
py curves were experimentally derived for static i static load, for the depth z applies the p-y curve.
cyclic load, so when we use cyclic p-y curves are experimentally derived for static and
due to dynamic load, we still use it cyclic load, so when the cyclic curves are used for
a relatively calm load, where they can only be caught dynamic loading, it is still a relatively calm loading, so
effects of load repetition, but not completely only the effects of loading repetition can be assessed,
dynamic influences. but not the complete dynamic effects.

Tabela 3. py criva: ÿ= 35°; D=0.60 m; ÿ= 10 kN/ m3 ; k= 33900 kPa/ m.

Table 3. p-y curve: ÿ= 35°; D=0.60 m; ÿ= 10 kN/ m3 ; k= 33900 kPa/ m.

i
is ya pa=pk pb=pm pc=pu
1 from
33900 8.15E-04 27-64 42.86 52.23
2 1
67800 7.00E-04 47.47 105.34 144.80
3 23
101700 3.53E-04 35.90 178.72 285.95
4455 135600 5.63E-04 76.29 303.64 485.82
6677 169500 8.18E-04 138.73 461.23 737.98
8899 203400 1.12E-03 227.78 651.51 1042.41
10 10 11 237300 1.47E-03 347.96 874.45 1399.13
11 271200 1.86E-03 503.70 1130.07 1808.12
305100 2.29E-03 699.40 1418.37 2269.39
339000 2.77E-03 939.43 1739.34 2782.95
372900 2.88E-03 1074.71 1952.70 3124.33

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89) 73

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

Tabela 4. py criva za ÿ= 35°; D=0.60 m; ÿ= 10 kN/ m3; k= 33900 kPa/ m

Table 4. p-y curves for ÿ= 35°; D=0.60 m; ÿ= 10 kN/ m3; k= 33900 kPa/ m

is= 33900 is= 1 z= 67800 is= 2 z= 101700 is= 3 z= 135600

z= 4

pc= 52.23 pc= y 144.80 pc= y 285.95 pc= y 485.82

(m) p (kPa/m) (m) p(kPa/m) (m) p(kPa/m) y (m) p(kPa/m)

0.000001 0.00001 0.000001 0.00001 0.000001 0.00001 0.000001 0.00001
0000 00 00
-0.001 -29.82 -0.001 -63.24 -0.001 -97.62 -0.001 -132.19
-0.002 -44.98 -0.002 -106.22 -0.002 -174.86 -0.002 -246.15
-0.003 -50.15 -0.003 -128.35 -0.003 -225.42 -0.003 -332.50
-0.0045 -51.93 -0.0045 -140.58 -0.0045 -263.57 -0.0045 -412.93
-0.005 -52.07 -0.005 -142.14 -0.005 -270.08 -0.005 -429.66
-0.007 -52.22 -0.007 -144.39 -0.007 -282.04 -0.007 -466.69
-0.009 -52.23 -0.009 -144.74 -0.009 -285.00 -0.009 -479.47
-0.01 -52.23 -0.01 -144.78 -0.01 -285.48 -0.01 -482.18
-0.015 -52.23 -0.015 -144.80 -0.015 -285.94 -0.015 -485.60
-0.02 -52.23 -0.02 -144.80 -0.02 -285.95 -0.02 -485.81
-0.0229 -52.23 -0.0229 -144.80 -0.0229 -285.95 -0.0229 -485.82
-0.025 -52.23 -0.025 -144.80 -0.025 -285.95 -0.025 -485.82
-0.03 -52.23 -0.03 -144.80 -0.03 -285.95 -0.03 -485.82
-0.035 -52.23 -0.035 -144.80 -0.035 -285.95 -0.035 -485.82
-0.18 -52.23 -0.18 -144.80 -0.18 -285.95 -0.18 -485.82

In an attempt to include them dynamically as well as possible In order to include dynamic effects as much as
influences, link elements are modeled over multiple linear possible, the link elements are modelled using multiple
of plastic elements of the hysteresis type, where it is cyclic linear plastic elements of hysteretic type, where the
py curve served as a kind of envelope. They are guilty cyclic p-y curve is used as kind of an envelope. The
calculated by a combination of both methods, it is for the first one curves are calculated by combining both methods, and
used program that calculates curve parameters for for the former the software is used which calculates the
every meter of depth, the results are given in table 3, a curve parameters for each meter of depth, the results
were introduced in SAP2000 as follows in table 4. being provided in table 3. Also, they are introduced in
SAP2000 as given in table 4.

3.4 Forces in the foundation due to the considered seismic 3.4 Base shear force due to the considered seismic
facts action

Figure 11a. Force at base X. ElCentro PGA 0.20 g. BS max Figure 11b. Force at base X. ElCentro PGA 0.25 g. BS max
1608 kN (5,460 sec). BS min 1304 kN (1,540 sec). 1834 kN (5,460 sec). BS min 1582 kN (1,540 sec).

Figure 11a. Base Shear X. ElCentro PGA 0.20 g. BS max Figure 11b. Base Shear X. ElCentro PGA 0.25 g. BS max
1608 kN (5.460 sec). BS min 1304 kN (1.540 sec). 1834 kN (5.460 sec). BS min 1582 kN (1.540 sec).

74 BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

Figure 11c. Force at base X. ElCentro PGA 0.30 g. BS max 2114 kN (2.720 sec). BS min 1854 kN (1,540 sec)
Figure11c. Base Shear X. ElCentro PGA 0.30 g. BS max 2114 kN (2.720 sec). BS min 1854 kN (1.540 sec)

All three force graphs are basically very similar. However All three graphs of the BS forces are basically very
it is observed that the peak value is the maximum for 0.30g, similar. However it can be observed that the peak value
moved from 5.46 sec to 2.72 sec. According to research for 0,30g, changes from 5.46 sec to 2.72 sec. According
[18], when analyzing the impact of the accelerogram, it is not only important to the research [18], during the accelerogram (i.e. time
peak ground acceleration PGA, it is already necessary to observe i history) analysis, not only peak ground acceleration
the immediate environment, and see how it is the maximum (PGA) is important, but it is necessary to observe the
spregnut sa sudesnim extremima. To se ovde immediate surroundings of the peak, and find out in
also applies to the force in the base. which way the peak is related to the adjacent peaks. It is
implemented here for the Base Shear force.

Table 5. Dependence of base force and PGA(g). Moment max and min.
Table 5. Variation of base shear force with respect to PGA (g). Instances of max and min

From Center Base Shear Base Shear t max t min

PGA (g) max (kN) min (kN) (sec) (sec)

0.20 1608 1304 5.460 1.540
0.25 1834 1582 5.460 1.540
0.30 2114 1854 2.720 1.540

3.5 Effects in link elements from NDA (TH) for effect 3.5 Effects in the link elements from NDA (TH)
„El Centro“ and preko rada action of „El Centro“ and via the work

As a result of this analysis, Fig. 12 shows As a result of this analysis Fig. 12 presents
displacement and force diagrams of coupled pairs link displacement and force diagrams of coupled link
elements 1 and 2 (depth 1 m), for PGA 0.20g. These paired elements 1 and 2 (depth 1 m), for PGA 0,20g. Coupled
the elements are connected in the same pile node. link elements are related to the same node.

Slika 12a. Link 1 i 2, nivo -1,0 m. PGA 0,20g El Centro Figure 12b. Link 1 in 2, level -1.0 m. PGA 0.20g El Centro
NDA. Displacement: max 0.201cm. min 0.194 cm NDA. Force. max 45.88 kN
Figure 12a. Link1 and 2, level -1,0 m. PGA 0,20g El Figure 12b. Link 1 and 2, level -1.0 m. PGA 0.20g El
Centro NDA. Displacement: max 0.201cm. Min NDA Center. Force. max 45.88 kN
0.194 cm 0.201cm. min 0.194 cm.

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89) 75

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

In addition to nonlinear analysis, absolute work was considered, In addition to the non-linear analysis, the Absolute
of paired link elements, as a positive effect value work of coupled link elements is determined as a posi-tive value of
forces along the road. Diagram of cumulative Absolute work, force action along the path. The diagram of
of paired link elements is shown in Fig. 13. Absolute cumulative Absolute work of the coupled link elements is
work is necessary to avoid cancellation presented in Fig. 13. The absolute work is necessary to
of positive and negative work during the summation. Negative avoid cancelation of the positive and negative works
work is a consequence of multiplying the force in the link element by during addition. The negative work is a consequence of
with a negative sign, with a positive sign multiplication of a force in the link element with a
moving it. In Figure 13, steeper parts can be seen negative sign, with positive displacement. In Figure 13
cumulative curves, which represent places where steep sections of the cumulative curve are noticeable,
"heap" intensity of the accelerogram that affects the work of the link which represent the locations where the intensity of the
elements. This is actually about seismic dissipation accelerogram which affects the work of link elements is
energy in the soil, at a depth of about 1 m from the soil surface. “piling up”. It is actually the case of dissipation of seismic
energy in soil, at around 1 m bellow the surface.

Kumulativ. SUM |ABS Medium force x dif. Displacement| Series1

6,00E-01
5,00E-01
4,00E-01
3,00E-01
2,00E-01
kum. 1,00E-01
|x
0,00E+00
0 2 4 6 8 10 12 14

t (sec)

Figure 13. Cumulative Absolute rad link elements during the El Centro event. Link 1 and 2, level -1.0 m ground depth. PGA
0.20g EL Center NDA.
Figure 13. Cumulative Absolute work of link elements under action of El Centro. Link 1 and 2, level 1,0 m depth below
ground surface. PGA 0.20g ELCentro NDA.

Furthermore, the characteristic ones are listed in the tables Further, the tables 6 to 9 show the characteristic
values of individual influences in link elements. values of individual parameters of the link elements.

Table 6. Link 1 and 2 Current Work = Force * Displacement Fi x U1i (kNm) El Centro 0.20g.
“ *
Table 6. Link 1 and 2 Instantenous work“ = Force Displacement Fi x U1i (kNm) The Center 0.20g.

Link 2 Link 1 ElCe 0.20g

min 4.98E-02 8.90E-02
max -7.04E-08 -4.79E-07
extr 0.04982 0.089016
Addition 2.945 4.522 7.467
% 39.44 60.56

Pure work is not considered here as current work, as Here, instantaneous work is not considered as an
the effect of the force along the path, but only the current values of the force effective work, as an action of the force along the path,
(Fi), as reactions of the link element at the moment (ti) multiplied but only as the instantaneous value of the force (Fi), as a
with the current (extreme) value of the corresponding reaction of the link element at a time (ti) multiplied by the
displacements of the node (U1i) in which link element i meets instantaneous (extreme) value of the corresponding
the final elements of the pile, at a given moment of action node displacement (U1i) at which the link element and
accelerogram. That is why the product of force i is written the finite elements of the pile join together, at a given
movements, i.e. force x displacement. Both of these sizes are moment of ground motion action. That is why it is written
linear functions, in the observed time as a product of the force and displacement, i.e. Force x
intervals, and in the general sense (as an integral) is multiplied Displacement. Both parameters are linear functions, at
triangle with a trapezoid (here the trapezoid is replaced the observed time intervals, so in the general sense (as
rectangle) whose ordinate is the mean value of the force. an integral) a triangle is multiplied by a trapezoid (here a
Strictly speaking, therefore, the average should be observed trapezoid is replaced by a rectangle) whose ordinate is a
force value (Fsr) and displacement difference (ÿU1), in each mean force value. Strictly speaking, one should observe
individual time interval, as pure work. Clean work the mean force value (Fsr) and displacement difference
the link of the elements is given in table 7, and these are the values of the smaller (ÿU1), at each individual interval of time, as effective
of an order of magnitude from the aforementioned current work. WITH work. Effective work of elements is provided in table 7,

76 BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

given that the output data, unlike the input data, are given in and those are the values of a lower order of magnitude
equal time intervals of 0.02sec, from than the previously mentioned instantaneous work.
values of pure work can be obtained directly and power po Regarding that the output data, as opposed to the input
to the link element by multiplying by 1/0.02=50 Hz (J/sec=Watt). data, are given at equal time intervals of 0.02sec, from
the value of the effective work one can directly calculate
the power by a link element, by multiplying with
1/0.02=50 Hz (J/sec=What).

Tabela 7. Link 1 i 2. Apsolutni rad, ABS |A = Fsr x ÿU1| (kNm); El Centro 0,20g
Table 7. Link 1 i 2. Abs. work, ABS |A = Fsr x ÿU1| (kNm); El Centro 0.20g

Link 2 Link 1
min 3.96E-03 6.77E-03
max 0.00E+00 0.00E+00
extr 0.00396 0.006773 Addition
Addition 0.197484 0.308631 0.506115
39.02 60.98 %

Extreme values of the sum Abs of the work of forces in pairs The extreme values of sums of Absolute work of the
link elements are different. For the above case it is a relationship forces in the coupled link elements are different. In this
39/61=0.64. This is a consequence of the introduced nonlinearity case, it is the ratio 39/61=0.64. It is a consequence of
behavior of link elements and accelerogram asymmetry. the introduced non-linear behaviour of link elements and
asymmetry of accelerogram.

Table 8. Link 1 and 2 Ratio (m); Force - reactions (kN); ElCentro 0.20 g
Table 8. Link 1 and 2 Displacement (m); Force – reaction (kN); ElCentro 0.20 g.
Extr Link 2 Link 1

Displac.(m) 0.00201 0.00194

Force (kN) 37.18 45.88

Extreme values of forces and displacements in a paired link The extreme values of forces and displacements in
elements are different. For the above case it is the relation for the coupled link elements are different. In this case, the
forces 37/46=0.81 which is not negligible, and for displacements force ratio is 37/46=0.81, which is not negligible, but
201/194=1.036 which are close values. obtained displacement ratio is 201/194=1.036 which are
close values.

Table 9. Change in extreme displacement (cm) and force (kN) in link elements with increasing PGA El Centro.
Table 9. Variation of extreme displacement in (cm) and force in (kN) in link elements under PGA El Centro

movements movements behind Behind

PGA (g) Link 1 i 2 Link 3 i 4 Link 1 i 2 Link 3 i 4

0.20 0.201 0.100 45.88 65.26
0.25 0.231 0.114 47.65 72.71
0.30 0.281 0.145 49.71 85.81

There is a strong linear dependence between peak There is a strong linear dependence between the
acceleration of the PGA and displacement of the Link element. Approximately for peak ground acceleration PGA and Link element
Link 1 i 2: U1ÿ0.95 · PGA(g); a za Link 3 i 4: U1ÿ0,48 · displacement. Approximately, for the Links 1 and 2 it is:
PGA(g). As for Link element forces (y), linear U1ÿ0.95 * PGA (g); and for the Links 3 and 4:
the dependence between PGA (x=ah/ g) and the same is determined U1ÿ0.48*PGA (g). As for the forces of the Link element
more precisely by the least squares technique, both for Link 1 and 2: y (y), the linear dependence between PGA (x=ah/ g) and
= 38.3·x + 38.172, (R2 = 0,9981); a za Link 3 i 4: y = forces is more accurately determined using the least
205,5·x + 23.218, (R2 = 0,9754). Kod Link elemenata 1 i square technique. For the Links 1 and 2 it is: y = 38.3·x +
2, due to the increase in PGA from 0.20g to 0.30g , displacement 38.172, (R2 = 0.9981), while for the Links 3 and 4 it is: y
increase by about 40%, while at Link 3 and 4, for the same change = 205.5·x + 23.218, (R2 = 0.9754). In the case of Link
PGA extreme displacement increases by 45%. elements 1 and 2, due to the increase of PGA from
Furthermore, diagrams are shown in Figures 14 to 24 0.20g to 0.30 g, the displacements increase for around
forces and displacements of Link elements for El Centro PGA 0.3g. 40%, while for the Links 3 and 4, for the same change of
PGA the extreme displacement increases for 45%.
Further in the text, Figures 14 to 24 are presenting
the force and displacement diagrams of the Link
elements for El Centro PGA 0.30 g.

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89) 77

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

Slika 14a. Link 1 i 2, nivo -1,0 m. PGA 0,30g ElCentro NDA. Pomeranje: Slika 14b. Link 1 i 2, nivo -1,0 m. PGA 0,30g ElCentro NDA. Sila max
max 0,281cm. min 0,272 cm Figure 14a. Link 1 and 2, level 49,71 kN Figure 14b. Link 1
-1.0 m. PGA 0.30g El Centro NDA. Displacement. max 0.281cm. and 2, level -1.0 m. PGA 0.30g El Centro NDA. Force. max 49.71
min 0.272 cm kN

There are noticeable gaps in the reaction forces of Link elements 1 There are noticeable gaps in the reaction forces of Link elements 1
and 2. At about 3.9 sec, then 4.6 sec, and at 9 sec there is the largest and 2. They are at around 3.9 sec, then at 4.6sec, and at 9 sec there is
pause in the reaction forces of Link elements 1 and 2. This could be a the largest gap in the force reactions of Link elements 1 and 2. It could be
sign that separation has occurred (gap) at the contact between the pile a sign that there is a gap between the pile and the soil.
and the ground.

Figure 15a. Link 3 and 4, level -2 m. PGA 0.30g El Centro NDA. Slika 15b. Links 3 and 4, snow -2.0 m. PGA 0.30g El Centro.
Displacement: max 0.281cm, min 0.272cm. NDA. Force max 85.81 kN.
Figure 15a. Link 3 and 4, level -2 m. PGA 0.30g EL Centro NDA. Figure 15b. Links 3 and 4, level -2.0 m. PGA 0.30g El Centro. NDA.
Displacement: max 0.281cm, min 0.272 cm. Force max 85.81 kN

Figure 16a. Link 5 and 6, level -3.0 m. PGA 0.30g El Centro NDA. Slika 16b. Links 5 and 6, snow -3.0 m. PGA 0.30g El Centro.
Displacement: max 0.0505 cm, min 0.0452 cm. NDA. Force max 44.10 kN.
Figure 16a. Link 5 and 6, level -3.0 m. PGA 0.30g El Centro NDA. Figure 16b. Links 5 and 6, level -3.0 m. PGA 0.30g El Centro. NDA.
Displacem.: max 0.0505 cm, min 0.045 cm. Force max 44.10 kN

78 BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

Although the distribution of forces is almost uniform in terms of Even though in terms of time, the force distribution is almost uniform:
duration Link 5 / Link 6 ÿ 60% / 40%, Link element 5 has higher reaction Link 5 / Link 6 ÿ 60% / 40%, the Link element 5 exhibits higher reaction
forces (almost twice as large as L6). forces (almost twice as that of L6). More accurate data are presented in
More precise data is shown in the summary table of Link elements. the summary table of the Link elements.

Figure 17a. Link 7 and 8, level -4.0 m. PGA 0.30g El Centro NDA. Slika 17b. Links 7 and 8, snow -4.0 m. PGA 0.30g El Centro.
Displacement: max 6,376*10-5 m, min 6,048*10-5 m. NDA. Force max 15.99 kN.

Figure 17a. Link 7 and 8, level -4 m. PGA 0.30g El Centro NDA. Displace. Figure 17b. Links 7 and 8, level -4.0 m. PGA 0.30g El Centro. NDA.
max 6.376*10-5m, min 6.048*10-5m. Force max 15.99 kN

Considerable asymmetry of pile reaction forces is noticeable at a There is a considerable asymmetry of reaction forces of the pile at a
depth of 4m from the ground surface. Practically only link element 7 reacts depth of 4m below the surface. Practical-ly, only the link element 7 is
and in relation to link element 8, it is over 90% of the reactive force during reacting, and in relation to the link element 8, it is over 90% of the reactive
the total duration of the seismic response at El Centro of 0.30g. force during the total duration of the seismic response to El Centro of
0,30g. It is assumed that such behaviour is related to the absence of
It is assumed that this behavior is related to the gap of reactions in link reaction of the link elements 1 and 2, and to force intensity asymmetry of
elements 1 and 2, and asymmetries in the force intensity of link elements the Link elements 3 to 6.
from 3 to 6.

Figure 18a. Link 9 and 10, level -5.0 m. PGA 0.30g El Centro NDA. Slika 18b. Link 9 i 10, nivo -5,0 m. PGA 0,30g El Centro. NDA.
Displacement: max 1,330*10-4m, min. 9,831*10-5m. Sila max 16,38 kN.

Figure 18a. Displacement Link 9 and 10, level -5.0 m. Figure 18b. Links 9 and 10, level -5.0 m. PGA 0.30g El Centro. NDA.
PGA 0.30g El Centro NDA. max 1,330*10-4m, min. Force max 16.38 kN
9.831*10-5m.

The distribution of reaction forces becomes uniform again at Links 9 Distribution of reaction forces becomes even again in the Links 9 and
and 10 (5 m from ground level). This could be the point of pinching of the 10 (5 m below the ground level). It could be location where the pile is
pile, in the model of the replacement console, (5/0.6=8.3 D), with the fact clamped, in the substitute cantilever model (5/0.6=8.3 D), provided that
that the pinching can also be elastic. the restraint may be elastic as well.

Figure 21b shows the transitional form of the force diagram, in relation In Fig. 21 one may notice a transition form of the force diagram with
to higher (and lower) ground levels than the -7m level. Longer intervals of respect to higher (and lower) soil levels then the level -7m. The longer
reaction forces of individual link elements are still observed, but the lines intervals of reaction forces of individual link elements can still be observed,
are no longer as smooth as for the upper soil layers, and resemble less but the lines are not as smooth as for the upper layers.
tires and more

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89) 79

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

to impulse structures (needle-shaped). Also, the lines resemble envelopes less, but rather resemble the impulsive
structures (having a pointed form).

Figure 19a. Link 11 and 12, level -6.0 m. PGA 0.30g El Centro Slika 19b. Links 11 and 12, snow -6.0 m. PGA 0.30g El Centro.
NDA. Displacement: max 1.025*10-4 m, min. 7,628*10-5m. NDA. Force max 15.32 kN.

Figure 19a. Displacement Link 11 and 12, level -6.0 m. PGA 0.30g Figure 19b. Links 11 and 12, level -6.0 m. PGA 0.30g El Centro. NDA.
ELCentro NDA. max 1.025*10-4m, min. 7.628*10-5m. Force max 15.32 kN

Figure 20a. Link 13 and 14, level -7.0 m. PGA 0.30g El Centro Slika 20b. Link 13 i 14, level -7.0 m. PGA 0.30g El Centro.
NDA. Displacement: max 3,440*10-5 m, min. 5,259*10-5m. NDA. Force max 8,086 kN.

Figure 20a. Displacement Link 13 and 14, level -7.0 m. Figure 20b. Links 13 and 14, level -7.0 m. PGA 0.30g ELCentro.
PGA 0.30g El Centro NDA. max 3,440*10-5 m, min. 5.259*10-5m. NDA. Force max 8.086 kN

Figure 21a. Link 15 and 16, level -8.0 m. PGA 0.30g El Centro NDA. Slika 21b. Link 15 i 16, nivo -8,0 m. PGA 0,30g El Centro. NDA.
Displacement: max 2,197x10-5 m, min. 1,640x10-5m. Sila max 4,413 kN.

Figure 21a. Link 15 and 16, level -8.0 m. PGA 0.30g El Centro NDA. Figure 21b. Links 15 and 16, level -8.0 m. PGA 0.30g El Centro. NDA.
Displacement max 2.197*10-5 m, min 1.640*10-5m. Force max 4.413 kN

80 BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

The diagram of reaction forces becomes sharp with alternately The reaction force diagram becomes pointed, with alternatively
arranged elements Link 15 and 16, level 8.0 m from the ground surface arranged elements Links 15 and 16, level 8.0 m below surface (8/0,6=13.3
(8/0.6=13.3 D). D).
The movement diagram of Link elements 21 and 22 is not visible, There is no noticeable displacement diagram of Link elements 21
because they are small sizes, but the numerical values, which are given and 22, because those are small values.
in the description of picture 24a, can be clearly read. (11/0.6=18.3 D). However, the numerical values may be clearly seen, as they are provided
in the description of Figure 24a.
(11/0.6=18.3 D).

Slika 22a. Link 17 i 18, nivo -9,0 m. 0,30g El Centro. Slika 22b. Link 17 i 18, level -9.0 m; 0.30g The Center. NDA. max Sila 3,881
NDA. Displacement max 1,536x10-5 m. kN.
Figure 22a. Link 17 and 18, level -9.0 m. 0.30g El Centro. NDA. Figure 22b. Links 17 and 18, level -9.0 m 0.30g El Centro.
Displacem. max 1.536x10-5 m NDA. max Force 3.881 kN.

Slika 23a. Link 19 i 20, nivo -10, m. 0.30g El Centro. Slika. 23b. Link 19 i 20, snow -10, m. 0.30g El Centro. NDA. max Sila 3,468
NDA. Displacement max 1.131*10-5 m. kN.

Figure 23a. Link 19 and 20, level -10. m. 0.30g El Centro. NDA. Figure 23b. Links 19 and 20, level -10.0 m 0.30g El Centro.
Displace. max 1.131*10-5 m. NDA. max Force 3.468 kN.

Slika 24a. Link 21 i 22, nivo -11,0 m. 0,30gEl Centro. Slika. 24b. Link 21 i 22, level -11.0 m. 0.30g The Center.
NDA. Displacement max 4.949*10-6 m. NDA. max Force 1,747 kN

Figure 24a. Link 21 and 22, level -11.0 m. 0.30g El Centro. NDA. Figure 24b. Links 21 and 22, level -11.0 m 0.30g El Centro. NDA.
Displacem. max 4.949*10-6 m. max Force 1.747 kN.

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89) 81

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

Table 10. Link elements by depth, for the left end pile. Extreme displacement, extreme forces for El Centro 0.30 g
Table 10. Link elements with depth, for left edge pile. Extreme displacement, extr. forces, for El Centro PGA 0.30 g.

Z Link U extr (m) 1 F extr (Fi*Ui) extr ÿ (Fi*Ui) (kN) (kNm) (Fsr*ÿUi)extr ÿ ABS(A) A/ÿA
(m) (kNm) 49.71 0.1352167 6.1454125 (kNm) (kNm) %
1 0.00272 0.01424095 0.4188097 59.89
2 0.00281 46.74 0.0939416 4.3699499 0.00716251 0.2804514 40.11
2 3 0.00138 85.81 0.1184187 6.5062411 0.01073076 4 0.00145 62.63 0.0566888 0.4461167 66.98
2.8040334 0.00529569 5 0.0004518 44.10 0.0199249 1.2593537 0.00203323 6 0.2199727 33.02
3 0.0005053 23.77 0.0057889 0.2804816 0.00062265 7 6.048E-05 15.99 0.0009670 0.0915624 77.69
0.0612844 0.00016362 8 6.376E-05 5.46 0.0001129 0.0012892 2.4366E-05 0.0262916 22.31
4 0.0076096 94.26
0.0004630 5.74
5 9 0.000133 13.65 0.0011192 0.0590964 8.3331E-05 0.0045659 43.80
10 9.831E-05 16.38 0.0016099 0.0642328 0.00015859 11 0.0001025 10.48 0.0058584 56.20
6 0.0005471 0.0289231 4.3299E-05 12 7.628E-05 13 5.259E -05 14 0.0000344 0.0025059 32.07
15.32 0.0011687 0.0609466 0.0001198 0.0053071 67.93
7 4.93 0.0001033 0.0048663 1.8819E-05 8.09 0.0002782 0.0006463 24.67
0.0153709 5.56E-05 0.0019731 75.33
8 15 2.197E-05 2.94 3204E-05 0.0007730 1.2754E-05 0.0002461 29.33
16 0.0000164 4.41 7.238E-05 0.0021958 2.1798E-05 0.0005931 70.67
9 17 1.536E-05 18 3.88 4.967E-05 0.0010876 1.6543E-05 3.79 4.731E-05 0.0003283 47.73
1.496E-05 19 0.0013961 1.0419E-05 3.47 3.565E-05 0.0010162 0.0003595 52.27
10 1.051E-05 20 1.2222E-05 3.13 2.896E-05 0.0008312 6.186E-06 1.75 0.0002711 55.46
1.131E-05 21 8.22E-06 2.34E-04 2.76 E-06 0.0002177 44.54
11 4.76E-06 6.98E-05 56.37
22 4.95E-06 1.64 7.29E-06 1.75E-04 1.54E-06 5.40E-05 43.63

Maximum displacements of link elements in the first three The maximum displacement of link elements at the
meters of depth are from 2.8 mm to 0.4 mm. Despite first three meters of depth range between 2.8mm and
with such small displacements over 95% of the seismic 0.4 mm. In spite of such small displacements, over 95 %
the energy of the link elements of this pile is spent precisely on of seismic energy of link elements of this pile is
that depth. It is (3m/0.60m=5D) a depth of five diameters dissipated exactly at that depth. It is (3m/0,60m=5D) a
pile. This is consistent with the largest influences on the coefficients depth of five diameters of a pile. It is in agreement with
A and B (for limit load in displacement and force) code the highest effect on coefficients A and B (for limit loads
theories of py curves for static and repeated loading. of displacements and forces) of the theory of p-y curves
Table 11 refers to the leftmost standing pile for the static and cyclic loads.
diameter D60cm, founded at a depth of 12m, whereby Table 11 refers to the left-end standing pile, with
are foundation pillows 100 cm thick, and the length of the pile is 11 m, diameter D60cm, founded at a depth of 12m, whereby
from the bottom edge of the cushion to the pinching in the base. Py curves the foundation cap is 100cm thick, so the pile length is
they were made for each meter of pile depth, with the fact that 11m, from the lower side of the cap to the base. The p-y
the influence of the foundation cushions is neglected. (Note: code curves are calculated for each meter of pile depth, while
of making the foundations of machines, it is necessary to perform well the influence of the foundation cap is ignored. (Note:
compaction of the soil around the foundation, because this impact of contact when constructing foundations for machinery, it is
foundation and soil becomes significant). According to table 11, for levi necessary to compact the soil around the foundations
end standing pile D60cm, 90% energy dissipation link well, because the effect of the foundation and soil
soil elements is carried out in the upper two meters of depth contact becomes influential). According to table 11, in
(2m/0.60m=3.33), and 99% of energy dissipation link case of the left-end standing pile D60cm, 90% of energy
elements of the soil, is done in the upper four-five meters dissipation of the link elements of the soil is performed in
depth (5m/0.60m=8.33D). Total work of link elements the top 2 meters of depth (2m/0,60m=3.33). And 99% of
of this standing pile, during the action of the El Centro earthquake energy dissipation of link elements of soil is performed in
of PGA 0.30 g, is relatively small and amounts to only 1513 Nm. the top four-five meters of depth (5m/0.60m=8,33D). The
Although seemingly small, this pressure is well distributed throughout total work of the link elements of this standing pile,
soil depth and very significant for seismic resistance during the action of El Centro earthquake with PGA 0.30
constructions. Figuratively, it would be the work that a man would do g, is relatively small and amounts to mere 1513 Nm.
which would lift a 155kg load from the surface with a winch Even though it is seemingly small, this pressure is well
of the ground at a height of 1 meter. distributed along the soil depth, and very important for
seismic resistance of the structure. In descriptive terms,
it would be the work performed by a man who would lift a
155kg weight using a pulley to a height of 1 meter.

82 BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

Table 11 Link elements by depth for the left end pile. Cumulative ABS operation of paired link elements
(for every meter of depth).
Table 11 Link elements with depth for left edge pile. Cumulative ABS work of coupled link elements.

z(m) Link cumulative ÿABS Ai (kNm) % -1 L1+L2 S%

0.6989213980 46.183 46.183
-2 L3+L4 0.6655929120 43.981 90.163
-3 L5+L6 0.1178280390 7.786 97.949
-4 L7+L8 0.0080650920 0.533 98.482
-5 L9+10 0.010409607 0.688 99.170
-6 L11+12 0.007806610 0.516 99.686
-7 L13+14 0.002618564 0.173 99.859
-8 L15+16 0.000839051 0.055 99.914
-9 L17+18 0.000687646 0.045 99.960
-10 L19+20 0.000488583 0.032 99.992
-11 L21+22 0.000123804 0.008 100.000
S 1.513381300

Energy disipation link element by depth of pile 1

0 10 20 30 40 50

46,183
-2 43,981
7,786
-4 0,533
0,688
z
depth
-6 0,516
0,173
-8 0,055
0,045
-10 0,032
0,008
-12

(%)

Figure 25. Percentage (%) of energy dissipation for link elements by depth for pile 1
Figure 25. Percent (%) energy disipation of link elements with depth from soil surface for pile 1

% energy disipation of pile 1 with dept

-30,00 -20,00 -10,00 -2 0,00 10,00 20,00 30,00

-4

-6

-8

-10

-12

Left Side of Pile 1 Right Side of Pile 1

Figure 26. Percentage (%) of seismic energy dissipation, for link elements, by pile depth 1 (for left and right sides).
Figure 26. Percent (%) seismic energy disipation, of link elements with depth for pile 1 (left and right sides)

It is assumed that the seismic shock occurs in It is assumed that the seismic impact occurs in one
one plane, the plane of the 2D frame. So it can be observed plane, the 2D frame plane. Thus, the left and the right
left and right side of the frame. This is not entirely true, but sides of the frame may be distinguished. This is not
can be accepted in the step-by-step impact analysis procedure entirely true, but it can be accepted in the analysis pro-cedure of
in the soil and the soil-pile system. the effects in the soil and the soil-pile system.

3.6 Development of plastic joints and the first characteristic tone 3.6 Development of plastic hinges and the first
(El Centro sa PGA 0.20; 0.25; 0.30g) natural mode (The Center of PGA 0.20; 0.25;
0.30g)
The change in the state of plastic joints was also studied -
state of plastic hinge (SPH) and due to a change in static The variation and change of condition of plastic
system change of the first characteristic shape, 2D frame hinges and the first natural mode of a 2D frame founded
founded on piles, with an increase in PGA. on piles, with the increase of PGA is studied.

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89) 83

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

Figure 27. El Centor 0.20g condition at the earthquake site. Left) layer. joints: 90 Y + 6 IO; right) form 1 vibration
Figure 27. El Centro 0.20 g. State at the end of an earthquake. Left, Pl Hinge state: 90 Y + 6 IO. Right, Mode 1.

Figure 28. El Centor 0.25g state at the end of the earthquake. Left) layer. joints: : 92 Y +7 IO; right) form 1 vibration
Figure 28. El Centro 0.25 g. State at the end of an earthquake. Pl Hinge state: 92 Y +7 IO. Right, Mode 1.

Figure 29. El Centor 0.30g condition at the earthquake site. Left) layer. Joints: 86 Y +10 IO+3 LS; right) form 1 vibration
Figure 29. El Centro 0.30 g. State at the end of an earthquake. Pl Hinge state: 86 Y +10 IO+3 LS. Right, Mode 1.

84 BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

Table 12 shows the change of the first and second Table 12 is presenting variation of the first and the
characteristic tone, after the effect of the accelerogram El second natural modes, after action of accelerogram El
Centro, from 0.20; 0.25 and 0.30 g and corresponding changes Centro, of 0.20; 0.25 and 0.30 g and the corresponding
static system due to the appearance of plastic joints. change of the statical system due to appearance of
plastic hinges.

Table 12. The first two eigenperiods after El Centro of different PGAs. 2D frame.
Table 12. The first two natural periods after El Centro with different PGA. 2D Frame.

PGA (g) T1 (sec) T2 (sec) T1 % T2 %

start 1.37255 0.44269 0 0

0.20 1.73011 0.86837 26.05 96.16

0.25 2.36767 1.00557 72.50 127.15
0.30 2.39338 1.03398 74.37 133.57

Figure 30. Displacement of node 9, at the top of the frame, during the action of the El Centro earthquake, for PGA 0.20; 0.25 and 0.30 g. Top left for
0.20 g, so that 0.25 g is 0.30 g
Figure 30. Displacement of node 9, at the top of the frame, during action of earthquake El Centro, for PGA 0.20; 0.25 and
0.30g. Upper left for 0.20 g, upper right 0.25 g and down 0.30 g

Tables. 13. Displacement of the node at the top of the column, for different PGA, for non-linear and linear roof beam.
Table. 13. Displacement of the node at the column top, for different PGA, for nonlinear and linear roof girder.

PGA (g) min U1 Joint 9 0.20 max U1 Joint 9 extr U1 Joint 9 Lin. Roof Beam L/NL RB%
-0.0731 0.0485 0.0731 0.0856 117.10
0.25 -0.0894 0.0636 0.0894 0.1129 126.29
0.30 -0.1078 0.0802 0.1078 0.1447 134.23

The extreme displacement of the node at the top is the code Extreme displacement of node at the top in the case
linearized roof beams larger at PGA 0.20g for of linearized roof beam is 17% higher for PGA 0,20g,
17%, with PGA 0.25g for 26%, and with PGA 0.30g for 34%. 26% for PGA 0,25g and 34%for PGA 0,30g.
The presentation of the impact of the introduction of plastic is also of interest It is also of interest to present the effect of

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89) 85

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

joints in the middle of the beam span, for different PGA and introduction of plastic hinges at the mid-span of beams, for different
arrangement of other plastic joints, including those occurring in PGA and the arrangement of other plastic hinges, including those
piles. occurring in the piles.
In Figure 31, the system becomes sensitive to the introduction In Figure 31, the system becomes sensitive to introduction of
of plastic joints in the middle of the beam spans. Roof beams plastic hinges at the mid-spans of the beams. The roof beams
experience collapse, for all 3 PGA (0.20; 0.25; 0.30g). Other beams collapse, in case of all 3 PGA (0,20; 0,25; 0,30g). The other beams
and columns behave more favorably, however, this is at the and columns behave in a more favourable way; however, this
expense of the piles, because the plastic joints move into the piles, comes at the expanse of the piles, because the plastic hinges
already at PGA 0.20g. All other plastic joints, except the roof joints, migrate to the piles, as early as at PGA 0,20g. All other plastic
enter state B, i.e. the beginning of yielding Y(yield). A similar hinges, except the roof ones, acquire the B state, which is the onset
phenomenon occurs [4] in the middle frame of the bridge, but it is of yield Y(yield). A similar pheno-menon took place in [4] at the
a logical phenomenon of testing the change in soil stiffness. middle frame of a bridge, but it is a logical consequence of testing
Namely, in [4], the stiffnesses of the linear springs of the soil on the variation of soil rigidity. Namely in [4], the stiffness of linear
the piles were varied, when the phenomenon of migration of plastic springs of the soil on piles was varied, and the phenomenon of
joints into the piles was often observed, during the reduction of soil migration of plastic hinges into piles was often observed, during the
stiffness. This phenomenon does not always happen, and it also reduction of soil density. This phenomenon does not occur always,
depends on the first (sometimes second) own tone of the and it depends on the first (and sometimes on the second) natural
construction, and the response spectrum of the applied mode of the structure, and the response spectrum of the applied
accelerogram. [21] studied the depth of appearance of plastic joints accelerogram.
in the middle frame of the bridge for several types of soil. In [14] it
is stated that the results of py curves can differ several times in
some cases. The depth of the onset of plastic hinges, for the middle frame of a bridge,
for several types of soil was studied in [21]. In [14] it was mentioned, that
the results of p-y curves in some cases can be different several times.

Figure 31. Distribution of plastic hinges for different PGA when plastic hinges occur in the middle of the roof beam span Figure 31. Plastic
hinge in middle of beam span, for different PGA, and distribution of plastic hinges

Here at the buildings, the stiffness of the soil is not varied. For the buildings here, the soil stiffness is not varied.
Namely, only one type of soil was used (stiff sand, submerged), Namely, only one type of soil is used (dense sand, submerged),
always the same length of pile and pinching conditions in the base. always the same length of piles and clamped conditions at the
Only one type of accelerogram was used, namely El Centro, base. Only one type of accelerogram is used: it is El Centro, only
horizontal component only, for PGA 0.20; 0.25 and 0.30 g. This the horizontal component, for PGA 0.20; 0.25 and 0,30g. This
phenomenon in buildings requires further research. Among other phenomenon related to buildings requires further research. Among
things, a more precise application of the models given in [2]. For other things, a more accurate use of models provided in [2]. For
certain types of soil, accelerograms, peak accelerations and certain types of soils, accelerograms, peak accelerations and pile
characteristics of piles, results can be obtained, which model the charac-teristics, results may be determined, where one could model
secant stiffness of soil springs instead of py curves [16]. the secant stiffness of the soil springs instead of using the p-y
curves [16].

4 CONCLUSION 4 CONCLUSIONS

The soil under the foundation is often abstracted in seismic Soil beneath the foundations is often ignored in seismic
analyses, and the structure is considered wedged into the foundations. analyses, and structures are considered as clamped in the
However, with tall buildings, bridges of larger spans and foundations. However, tall buildings,

86 BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

with some engineering structures, the structure-foundation-soil interaction large-span bridges and some engineering structures require inclusion of the
should also be included in the seismic analysis. structure-foundation-soil interaction. Introduction of the analysis based on
The introduction of analysis on 3D models is very complex, so in this paper 3D models is very complex, so in this paper it is shown that by replacing the
it is shown that by replacing the spatial skeletal structure of the building with spatial frame structure of a building, with a 2D frame, the problem is
a 2D frame, the problem is greatly simplified. considerably simplified.

When determining the seismic performance of a structure using NSA When determining the seismic performance of a structure using the
(pushover analysis), it is important to determine the point when the structure NSA (pushover analysis), it is important to determine the point at which the
transitions into a mechanism. The change in the number and condition of structure becomes a mechanism. The change in the number and states of
plastic joints with increasing displacement, in the NSA steps when plastic hinges resulting from the increase in displacements, (in steps) of PO
determining the PO curves in the SAP2000 v14 program, is not easy to curves using SÿP2000 v14 software cannot be easily determined. ETÿBS
determine directly. The ETABS program has a better display of PO curves software has better displays of PO curves, although this can be achieved in
for buildings, although even with the SAP2000 program package, good SÿP2000 as well (especially in the case of engineering structures) if
displays can be obtained, especially for engineering buildings, but alternative alternative procedures are taken into account. A PO analysis is applied
procedures should be taken into account. Analysis using the N2 PO method within the N2 method in order to determine the target structure displacement,
is applied to determine the target displacement of the structure, as a point of as an intersection point of the seismic requirements (through spectrum
intersection of the seismic demand (over the response spectrum) and the response) and of the seismic capacity of structures. The presented relatively
seismic capacity of the structure. A relatively simplified procedure for simplified procedure for determining the effects of NSÿ and dynamic NDÿ
determining the impact of NSA and NDA dynamic soil-pile-structure soil-pile-structure interaction is provided in this paper. In order to obtain a
interaction is presented. In order to obtain a more comprehensive picture of more comprehensive insight about the structure’s performance, it is
the construction's performance, apart from several different models, with necessary to apply several different models, load shapes, types and scales
and without interaction, it is necessary to apply several different methods, of accelerograms, procedures and software packages, with and without
load forms, several different types and scaling of accelerograms, then interactions.
procedures and software packages.

Numerical investigations of the impact in the soil have determined that

the reaction forces of the link element are small compared to the total
seismic force in the foundation. Although the reaction intensities of the link Numerical research of effects in the soil, determined that reaction forces
elements, during the action of the earthquake, in relation to the value of the of link elements are small in relation to the total base force. Even though the
forces in the base, are relatively small, they are very significant for the overall intensities of link elements reactions during earthquakes are relatively small
seismic resistance of the building. Certain regularities in the change of the in comparison to the value of base forces, they are very important for the
force diagram according to the depth of the link elements were observed, total seismic resistance of the structure. Certain regularities in the variation
but it is necessary to interpret them on the diagrams, which is done in this of the force diagram, along the depth of the link elements are observed, but
paper. they need to be interpreted on the diagrams, which has been done in this
paper.
By analyzing the seismic behavior of the input data (accelerograms for
PGA 0.20, 0.25 and 0.30g), the considered system is very sensitive to the
introduction of plastic joints in the middle of the beam span. The roof beams The analysis of the seismic behaviour using the input data (accelerograms
broke, for all 3 PGAs (0.20; 0.25; 0.30g), the other beams and columns with PGA 0.20; 0.25 and 0.30g) showed that the considered system is very
behave more favorably because the plastic joints "move" into the piles, sensitive at early formation (introduction) of plastic hinges at mid-spans of
already at PGA 0.20g. That is why it is very important to avoid the formation the beams. There was a failure of the roof beams, for all 3 PGA (0.20; 0.25;
of plastic joints in the beam fields by adequate dimensioning and details 0.30g), while the remaining beams and columns behave more favourably,
when designing AB frame constructions. because the plastic hinges migrate to piles, as early as at PGA 0,20g. For
that reason, it is very important to avoid formation of plastic hinges in the
beam spans using adequate design and details of RC frame structures.

It can be concluded that the introduction of SSI achieves a positive

effect, especially in the case of stiffer building structures, in order to avoid
greater ceiling deformations and potential collisions with neighboring
buildings in denser urban areas, which was confirmed in [1] and [9]. . It may be concluded that by introduction of SSI a positive effect could
be achieved especially if stiff building structures are in question, in order to
Further research should be extended to all frame piles, and for py avoid severe ceiling deformations and potential collision with adjacent
curves for different relative sand densities. It is also necessary to introduce structures in densely populated urban environments, which is confirmed in
the vertical interaction with the soil, which is neglected in this paper. Also, [1] and [9] as well.
with pushover curves, it is necessary to determine whether there is a strong
regularity of the form of the vertical load from the upper structure, with the The following research must be extended to all the frame piles, and to
form of the response of the py curves by the depth of the piles (the form of p-y curves for different relative sand densities. It is also necessary to
the response of the movement of nodes and drift piles). introduce a vertical interaction with the soil which is ignored in this paper.

This response can also be observed in TH analysis, and whether there is a Also, in pushover curves, it is necessary to determine whether there is a
clear legality has yet to be determined. strong regularity of the vertical load upon the superstructure, with the form
of response of p-y curves, along the depth of the piles (shape of the nodal

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89) 87

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

displacement response and the pile drift). This response may be analysed
through the TH analysis as well, and it still needs to be determined if there
is a clear regularity.

GRATITUDE ÿCKNOWLEDGEMENT

This work was done with the financial support of the Ministry of This paper was done with the financial assistance of the Ministry of
Science, Education and Technological Development of the Republic of Science, Education and Technological Development of the Republic of
Serbia, within the project for technological development TR36043. Serbia, within the project for technological development TR 36043.

5 LITERATURE
REFERENCES

1. Carbonari, S., Dezi, E., Graziano, L. (2012): Nonlinear seismic behaviour 11. MacLeod I.A.(1990): Analytical modelling of structural system. Ellis
of wall-frame dual systems accounting for soil-structure interaction, Horwood limited. London.
Earthquake Eng. and Structural Dynamic, 41, pp. 12. Mayer, B.J., Reese,L.C. (1979): Analysis of single piles under lateral
loading, Res. St. 3-5-78-244, Texas Sdof Highways PT
1651-1672. DOI: 10.1002/eqe.1195. 2. ÿauševiÿ
M.: Dynamics of constructions. Golden Marketing. Technical book. Zagreb, 13. Maymond P. J. (1998): Shaking table scale model test of nonlinear soli-
2010. 3. ÿosiÿ, M., Foliÿ, R., Brÿiÿ, S. (2017): An pile –superstructure interaction in soft clay, University of California,
overview of modern seismic analyzes with different ways of damping Berkley. Ph.D.
introduction, Building Materials and Structures (Building materials
and structures) (60), no. 1, pp. 3 – 30. 14. Miloviÿ, D., ÿogo. M.: Soil-foundation-structure interaction problems.
Serbian Academy of Sciences and Arts branch in Novi Sad. Novi
Sad 2009
15. Mosher R., Dawkins W.: Theoretical Manual for Pile Foundations, U.S.
4. Foliÿ B.: Seismic analysis of concrete structures founded on piles. Army Corps of Engineers, Report ERDC/ITL TR-00-5, Washington,
Doctoral thesis. FTN. USA, 2000.
University of Novi Sad. Novi Sad 2017
5. Foliÿ, B., Foliÿ, R. (2009): Design methods analysis of seismic 16. Pando, M. (2013): Analyses of Lateral Loaded Piles with p-y Curves -
interaction soil-foundation-bridge structures for different foundations, Observations on the Effect of Pile Flexural Stiffness and Cyclic
in: Coupled Site and Soil-Structure Interaction Effects with Loading.
Application to Seismic Risk Mitigation, Ed. T. NCDOT 7thGeo3T2, Raleigh, NC, Thursday, April 04, 2013. Paper:
3B-1_A49
Schanz and R. Jankov, Springer Sciences+ Business Media, pp. 17. Petronijeviÿ, M. (1993): Analysis of the dynamic combination of soil
179-191. and building using the finite element method. Doctoral thesis.
6. Foliÿ B., Ladjinoviÿ ÿ., Sedmak S., Liolios A.:Comparative nonlinear Faculty of Civil Engineering, University of Belgrade.
analysis soil-pile interaction 2D frame. 7th international conference:
“Geotehnicis in Civil Engineering”, ACE of Serbia, Proc. Ed. R. 18. Prakash, S. Ed, (1992): Soil under Dynamic Loads.
Foliÿ, Šabac, November 14-17. 2017. pp. 473-484. Geotechnical Special Publication. Geotechnical Engineering Division
of ASCE. 1992. No. 34 19. Reese L., Van Impe W. (2001):
Single pile and pile groups under lateral loading, Balkema, Rotterdam,
7. Foliÿ, R., Liolios, A.: Application inclined piles in seismic prone area, 2001
useful or not? 7th international conference: “Geotehnicis in Civil
Engineering”, ACE of Serbia, Proc. Ed. R. Foliÿ, Šabac, November 20. Stewart,J.P., Fenves, G.L., Seed, R.B. (1999): Seismic soil-structure
14-17. 2017. pp. 461-472. interaction in buildings.1: Analytical methods, Journal Environm.

8. Forootan, F., Moghadam, A.S. (2006): Comparison of 2D and 3D Engineering, ASCE, V. 125:1, pp. 26-37 21. Suarez,
pushover analysis with time history analysis in asymertic building, V.(2005): Implementation of Direct Displacement Based Design for Pile
First European Conf. on Earthq. Eng. and Seismology, Geneva, 3-8 and Drilled Shaft Bents. NCSU. North Caroliona State
September, Paper Number: 447 9. Jawad Arefi, M. (2008): Effects
of soil-structure interaction of the seismic University. October. 2005.
response of existing RC frame buildings, A dissert. of master degree, 22. Tabatabaiefara,H.R., Massumi, A. (2010): A simplified method to
Universita degli studi di Pavia, Italy 10. Kraus, I., Džakiÿ, D. (2013): determine seismic response of reinforced concrete moment resisting
Soil-Structure interaction effects on seismic behaviour of RC frames, building frames under influence of soil-structure interaction, Soil
5th Intern. Conf. CE EEE 1963. IZIS, Skopje, Dynamics and Earthquake Eng. 30, 11, pp. 1250-1267.
p. 8.

88 BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)
Machine Translated by Google

SUMMARY SUMMÿRY

COMPARATIVE NONLINEAR ANALYSIS COMPÿRÿTIVE NONLINEÿR ÿNÿLYSIS OF A RC 2D

PILE-SOIL INTERACTIONS AB 2D FRAME FRÿME SOIL-PILE INTERÿCTION

Boris FOLIÿ Boris FOLIC

Radomir FOLIÿ Radomir FOLIC

In the paper, a comparative nonlinear static (NSA) and nonlinear Comparative non-linear static (NSÿ) and non-linear dynamic analyses
dynamic analysis (NDA) of the seismic behavior of the frame as part of the (NDÿ) of 2D frames (as parts of skeletal 3D structures) of RC buildings
skeletal structure of the AB building founded on piles was conducted. In founded on piles are presented in this paper. In order to produce a more
order to get a more realistic picture of the behavior of the frame structure, realistic presentation of behaviour of a frame structure, the analysis
the analysis includes the interaction of the structure - the foundation - the involves a structure-foundation-soil interaction. Also, the model involves a
soil. At the same time, linear-nonlinear dynamic pile-soil interaction using linear-non-linear dynamic pile-soil interaction, using link elements. The
link elements is also included in the calculation model. foundation consists of drilled piles having 60 cm in diameter. The soil is
modelled using Multi-linear plastic link elements, as well as with p-y curves,
on both sides of the pile, assuming that p-y curves transfer only
The construction of the foundation consists of drilled piles with a compression (p-y curves are experimentally determined non-linear
diameter of 60 cm. The soil is modeled with multiple (linear) plastic tie relationships of displacement/pressure in soil, along the depth of a pile).
elements, as py curves, on both sides of the pile, for submerged stiff sand, The analysis shows the problems which accompany extraction of a 2D
and assuming that the py curves (experimentally determined nonlinear frame, as a representative of a regular 3D space frame. The impact of
dependence curves: displacement/pressure, in the soil per pile depth) onset and location of individual plastic hinges on seismic performances of
receive just pressure. The analysis pointed out the problems that the analyzed structural system are investigated, and relative floor drifts are
accompany the extraction of 2D frames as a representation of a regular analyzed. It was concluded that the analysis of 2D frame, in the interaction
spatial 3D construction. The influence of the appearance and location of with the foundation and soil, may provide sufficiently accurate results of
individual plastic joints on the seismic performance of the analyzed behaviour and assessments of seismic performances of skeletal RC multi-
structural system was studied, and relative floor movements (drifts) were storey building. It is important, because introduction of a spatial structure
analyzed. It was concluded that by analyzing the 2D frame in interaction in such analyses is very complex and challenging.
with the foundation and soil, sufficiently accurate results of the behavior
and evaluation of the seismic performance of the skeletal AB multi-story
building can be obtained. This is significant because the introduction of
spatial construction in such analyzes is very complex and demanding.

Key words: Dynamic soil-pile interaction, nonlinear dynamic analysis Key words: Dynamic soil-pile interaction (DSPI), non-linear dynamic
(NDA), nonlinear static (push-over) analysis (NSA), soil-structure analysis (NDÿ), non-linear static (pushover) analysis (NSÿ), soil-structure
interaction (SSI), multi-line plastic link element MPLE, py curves, influence interaction
distribution by ground depth link elements (SSI), multiline plastic link elements (MPLE), p-y curves, after-shock,
distribution of influence with depth of soil

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89) 89

BUILDING MATERIALS AND STRUCTURES 61 (2018) 1 (63-89)