Beruflich Dokumente
Kultur Dokumente
August
1993
Birkhauser
Verlag
Basel
Boston
Berlin
A CIP
Library
catalogue record for this book is available from of Congress, Washington D.C, USA
the
Linde, Peter:
Numerical
modelling
and
Eidgenssische
Technische
:
(ETH),
1993
Zrich.
Birkhauser,
(Bericht 200)
material is subject to Copyright. All rights are reserved, whether the whole or part of the concerned, spedfieally the rights of translation, reprinting, re-use of illustrations, recitation, broadcasting, kind of use permission of reproduction on microfilms or in other ways, and storage in data banks. For any This work is
the
Copyright
owner must
be obtained.
1993 Birkhauser
Verlag Basel,
Printed
on
aeid-free paper
98765432 1
Peter linde
Eidgenssische Technische
Ju 1993
PREFACE
are
protecting buildings
against excessive early damage and against collapse under earthquake actions. Structural
walls
designed by the
world-wide
increasingly used capacity design method exhibit design ductility according to well known rules.
there is
a
Complementary
to the
design procedures
way that
analysis
of
be modelled in such
possible. Furthermore,
of the
the local
ductility
plastic hinge
design ductility
beginning
design process.
doctoral thesis, two kinds of model for in the time domain
are
on a
simulating
A
developed:
model and
Springs
overall
connected
In
by rigid beams.
the
macro
approach.
particular,
model
ductility demand,
using
the moment
and
compared
of time
to the
process. Hence,
technique
practical
been made.
Prof.
Hugo
Bachmann
ACKNOWLEDGEMENTS
This report
was
prepared
at
Engineering
Institute of Technology
rably
to the work
presented
Among them,
some
individuals
ought
to
be
presented
in this report
possible,
inspiring
Anderheggen for
constructive
especially owed
generously
to Mr. Thomas
patiently as
applications.
Gratefulness is also
matische
expressed
to
the foundation
Gipsfabrikanten (VSZKGF)
presented in
ABSTRACT
an
important
buildings against
quakes,
successfully design
it is therefore of interest to
develop
behaviour of these units. This report is concerned with numerical models intended
used in
be
analysis
of
complete buildings,
with focus
on
capacity designed
multi storey
buildings.
A
the
development of a new
of a
macro
model which
simulates the
relatively simple
kinematics and
As
a
element is
presented.
complement
treat
model,
attempted to
a
relatively
physical observations.
code and extensive tests
are
The models
implemented into
with
important part
Performance checks
are
capacity designed
demand in the
by
of the
newly
dynamic
curvature
plastic hinge
may be different
suggested
in the
when varied
over
different wall
aspect ratios. It is further shown that during nonlinear time history analysis flexural
yielding
may
frequently
place in
are
intended
to
that the
dynamic
assumptions.
An
improved
distribution of flexural
over
height
of the wall is
proposed,
which
clearly reduces
yielding
structural
analysis: earthquake-resistant
reinforced concrete: shear
strength; hysteresis;
structural
analysis;
structural
design; walls
TABLE OF CONTENTS
PREFACE
ACKNOWLEDGEMENTS
ABSTRACT
I JJ
1. INTRODUCTION General 1.1 1.2 Objectives and Limitations 1.3 Scope of Report
1
1
7 7 7
Macro Models
2.3 2.4
2.4
2.2.1 Beam element models 2.2.1 Truss element models 2.2.3 Multiple spring element models Micro Models Meso Models Choice of Models for Development
8
10
10
12
13 14
3. MACRO MODEL 3.1 Introduction 3.2 Model Configuration 3.3 Elastic Hexural Behaviour 3.4 Elastic Behaviour under Normal Force 3.5 Nonlinear Flexural Behaviour 3.6 Shear Behaviour 3.7 Element Formulation 3.8 Ductility Demand
17
17 18
21
25 26
44
51
63
65 65
67
77
80 83
5. NUMERICAL EXAMPLES 5.1 Implementation 5.2 Analysis Procedure 5.2.1 General 5.2.2 Time integration 5.2.3 Residual forces
89
89 90 90 90 92
rv
5.2.4 5.3
Damping
93
5.4
5.2.4 Ground motion Selectionof Numerical Examples 5.3.1 Test specimen 5.3.2 Capacity designed multi-storey wall Macro Model Results
5.4.1
building
Multi-storey wall
to
5.6
Comparison experimental Multi-storey wall Comparison between Macro Model and Micro Model
results 5.5.2
134 143
144
Energy Dissipation Flexural Strength 6.5.1 Implication of numerical results 6.5.2 Suggested flexural strength design 6.5.3 Numerical example
Shear Behaviour
6.6
149 149 150 155 160 163 163 167 170 177
AND RECOMMENDATIONS
ZUSAMMENFASSUNG, SCHLUSSFOLGERUNGEN
AUSBLICK
UND
187
187 188
Zusammenfassung
Schlussfolgerungen
Ausblick
190
REFERENCES
191
NOTATION
199
APPENDDCA
USER ELEMENT INPUT DESCRIPTION USER MATERIAL INPUT DESCRIPTION YIELD MOMENT FOR MACRO MODEL ELEMENTS FOR MICRO MODEL FREQUENCY STUDY OF DAMAGED STATES DESIGN DEFINITIONS
205
211
CHAPTER ONE
INTRODUCTION
1. General
In many
buildings,
provide
an
important part
of the
resistance
wind and
commonly used,
impression
of
misinterpretations
Should, despite
structural wall,
a
careful structural
a severe
flexural failure is
strongly desirable.
and
failures
occur
less
provide
better
means
of
avoiding a
in
structural
and
collapse,
well
as
better
flexuralry
failing.
In the
early days
as
were
mainly designed
and
analysed
"wide" columns.
During
the 1950's
increasing
[T055]
or on
lateral action
emerged
more
mainly developed
in California. The
design
essentially
still dominated
by
the
assumption of
emerging
at
this time,
increasingly feasible,
[Nils68].
analysis
CHAPTER ONE
design
on
gained
findings
reinforced concrete
area
published by
Method"
Park and
Paulay
of seismic
design
achieved
by
the introduction
of the
"Capacity Design
et
developed
in New Zealand
by Paulay
al,
see
given by
in
[BP90] and
concrete structures so as to
a
large number
mainly carried
out at the
University of Canterbury by
selection of
in
Christchurch, New
preferable
The
plastic hinge
zones,
careful structural
detailing
of these, and
zone
protection
against
yielding.
plastic hinge
of structural walls is
usually
become cracked.
Figure
1.1 shows
and
as
Structural wall
aaata
H
Sgs
--
im
Plan
L-_<
Section
Figure
1.1
resistance
of
horizontal actions
INTRODUCTION
Confinement hoops,
zone
Ly,
Z.
Figure
Figure
1.2
cross
confined
by hoops which
are
tighy spaced
in the
plastic hinge
Sometimes
protruding
order to resist
bending
moments.
as
economically
confining-reinforcement placement
process, where materials
are
provided
in
[PBM90].
In the modern structural
design
are
used
efficiently
and for
of the
Performance
of the
designed
structure.
For
rational
design is usually no longer possible without the aid of Computer programs. Also, design methods,
such
to
as
Capacity
Design Method,
means
are
verify
Performance by
of a Computer program.
as
simple
and
readily available
designed specifically
to
CHAPTER ONE
1.2
Objectives
and Limitations
The
seen as a
further attempt to
clarify
the
most
means
of numerical
modelling. Emphasis is placed on Performance control of capacity designed walls. accomplish this, the
the
work must be
organised
so as to
fulfil
number of
objectives
set out as
set out at
The
follows:
1. Based upon
an
art
of
major
numerical wall
models to select
model
as
complement
a
to
this model in
cases
irregulr
geometry etc.,
relatively detailed
The second
in the first
Performance.
3. To
perform
development
analysis. Emphasis
when
no
placed
on
simple solutions,
particularly
in
identify which
to compare
Further,
the
Performance
of the model
basic criteria
INTRODUCTION
6. If applicable, to suggest
some
general improvements
in the
capacity design
procedure
7.
Finally,
to
In order to
an
efficient
manner a
imposed on
follows:
major limitations
of this report
were set
out as
1.
Mainly
the
global behaviour
In-depth
on
effects, etc.,
are not
considered in this
smdy.
the nonlinear behaviour of walls
no
3. The
modelling
on
essentially
discussion of linear
smdy
limits itself
to
essentially
to
not
deal with
connection of walls
other
as
three-dimensional
of
configurations
such
cases
although
capability
modelling
should be
regarded in
an
beyond
it is known to have
some
an
important influence
the structural
CHAPTER ONE
1.3
The
as
After this
chapter is
devoted to
review of
different numerical models for the Simulation of nonlinear behaviour of structural walls.
The
existing
are
brief
background
is
given. Advantages
briefly
chapter concludes
Chapter
development
of
Simulation of multi-storey
as a macro
model is
developed
in stages,
as
including
well
as
phenomena
the model.
to
be
regarded
concrete
steel behaviour
well
as
in
chapter
on
comparison
experimental data,
In
as
well
as a
check
some
the
Performance
of
chapter six
are
design
chapter
The
as
containing
research.
were
implemented into
In the
respectively.
use
Appendix
is
describing the
of these models.
CHAPTER TWO
2.1
Introduction
many different models for the numerical
a
There
are
modelling
the literature. As
background to
advantages
the
subsequent chapters
of this report
of the
more
important numerical models are presented in this chapter. The different models are briefly
described and their
and limitations mentioned. For
a more
comprehensive
The of a
study
of any
particular
appropriate reference.
to the
mainly relation
as
analysis
global
building statically
structure,
or
subjected
to
gravity loading
well
as
dynamically. modelling
of structural walls may, from
The numerical
structural
engineering point
section
major model
levels:
macro a
structural wall
certain
height,
while the latter base the behaviour upon the constitutive laws of the
may is
models. A
good discussion
[VB87].
the two
provided by
It may be
especially
definitions used here should not be confused with similar definitions used in the field of
fracture mechanics.
After
description
applicability to
various
structural
analysis situations
development in
this
chapter concludes
with
for further
this report.
as a
incorporate
height
or
major region
of
as a
storey
reinforcing steel,
CHAPTER TWO
'/////////////ZA
Y///Y/////A
O Node with translational
and rotational d.o.f.
Jk~^
Figure 2.1
Figure 2.2
macro
models
are
appears
possible
important
and
frequently
following.
The
simplest numerical
model for
degrees of freedom
is
a
is
regarded
as a
deep column.
a
This
very
model
which is
are
due to flexure
one
considered
single beam
seen
a
in
figure
modelling
as
is shown in
case
figure 2.2.
considerable horizontal
length,
as
well
in the
of
an
interaction with
edge
deformations. A
simple Solution including this effect has been suggested by adding hori-
a1-
Stiff Beams
h
r-
--
It
9//////////TA
-w
Figure
Figure
zontal
as
shown in
figure 2.3,
and
Few
and
The
uncomplicated modelling,
sometimes
possibilities
to
dynamic analysis.
along
its
are
mainly
due to the
inability to
at
each
edge
considered if
there
horizontal
rigid beams.
and
rigid
will
not
be
typical
for
subsequent yielding
reproduced. This
especially noticeable
considered
not
by
the model.
10
CHAPTER TWO
The next
macro
are
presented.
elements,
Typically,
a truss
shown in
figure 2.4,
are
consists of connected
two
a
vertical
truss
diagonal
truss
element. These
by
al
by Vallenas et
[VBP79],
by Hiraishi
strut"
[ACISP84].
The
diagonal
truss
is
opposite diagonal
applications
seems
problematic,
well
as
the realistic
modelling
and combined.
application, and
for
small
its
gravity load,
use
provide
the
useful results, if
appears
be limited to
compression
strut
of this nature
to
versatility
compared
2.2.3
The third
model is the
was
analytic prediction of
out at
experimental
tests on a
full-scale
seven
the
see
Building Research
figures
specimen,
was
used for
sections
one
in
and is shown in
one
figure
2.6. It
rotational
horizontal
seven
spring,
test
which
by rigid beams.
storey
structure could
Generally, important
large
can
tensile strains,
well
as
signifieant
shear deformations,
be simulated
adequaxely by models
on
this
approach.
11
I
s
i
i
I
i
VO
-ir*
VO
II"
8"
CS
(.
6.00
1.
5.00
J
6.00
b)Plan
I
yYv////Y^YfYyYr//////y/
----4
I
6.00
$1
I
I
5.00
6.00
\
test structure
I
I
a) Elevation
c) Macro Model
with structural
wall, tested in
m VMMMMM^^^^^
Vsssss/wss^^^
&
model
Figure 2.6
Initial
macro
major
are:
the
cross
sections remain
plane which is
poor
objective
and
here.
even
12
CHAPTER TWO
of
show that in
mainly
cross
plane
even
account a
localised
damage, such
as
crack direction.
some
experimental
Refinements of this
with
more
original model
have been
attempted by
some
The category of Micro Models is based upon the mechanics of solids, and wide field of the
comprises the
in
modelling
implementation
case
approximations,
this
performed by applying
plane
as
stress
by
implementation in
approaches
smeared for
membrane elements,
shown in
figure 2.7
During the early research of nonlinear concrete behaviour in the late 1960's two major
modelling cracking
approach
and the
approach.
to
efficiently
between reinforcement and concrete and well distributed cracks of moderate crack width. On the other hand, the discrete crack and Nilson
concrete
and Scordelis
structures
[NS67]
such
as
[Nils68]
was
a
found
to
be well-suited
unreinforced
a
dams, where
openings play
the These
signifieant
role in the
changed
approach
problem
of mesh
updaxing
still hold
by Skrikerud [Skri82].
on
findings essentially
today, although
the discussion
this
topic continues.
introduced
approach
was
analysis
of structural walls
An
by
application given by
the
was
ten-storey
concrete
building
seventies
13
i
/
a) Discrete approach
b) Smeared approach
Figure 2.7
Micro
modelling of cracking
zone
in
into
modelling
the behaviour of cracked concrete and the interaction between steel and
to
concrete.
develop
to
simple
micro
Clear
on
its
give information
elaborate model
compared to macro models however, is a clear drawback. For multi-storey buildings with
several structural walls
a
presented
use
in many reports
well
as
the lack of
global results,
resulting in
described
limited
in
practice.
element codes
Some of the
as
commercially-available finite
be that this is
in detailed
provide
material behaviour
A
problem
seems to
may not be of
use
dynamic
user
modify
this
concrete
behaviour
by
the
is
usually limited.
placed between
the
macro
presented by
e.g. Meskouris et al
as a
[MKHH91], and
to
will be denoted
complement
simplified
two
14
CHAPTER TWO
are
presented.
The
justification for
to
these models
was
mainly
bilinear
to
the
more
detailed models.
simple
Explicit
by orthotropic
expressions,
and
but instead
are
cracking
Thus,
yielding.
models, although they
are
meso
in continuous elements,
macro
use
simplified
slightly
and
belongs
In
the
varying quality.
only
computational
Development
advantages
and
disadvantages
for
given analysis
The
only or mainly
advantages.
opportunity
applied to the
same
analysis of the global behaviour of tail slender structural walls under primary interest. Thus,
they
function
to a
seismic excitation is of
the models
on
large degree in
the
global manner.
Of the
three types of
macro
models described
previously,
of
promising
and
capable
well
simulating
as
as
cross
sectional
Output
quantities, make it suitable for the analysis of capacity designed walls. multiple spring model
is referred
to as
was
Consequeny,
the macro-level
investigation
in this
on
now on
are
The
against
wall
experimental
to
a
dynamic
multi-storey
complement the multiple spring model by a clifferent numerical model, in order to have
15
Macro
bmparison of
dynamic behaviou
model
Performance tests
>
Capacity designed
structure
(Chapter 3)
(Chapter 5)
Experimental data
(Chapter 5)
Figure
Furthermore, for
some cases
of
with
irregularities
analysis
of of
preferable.
When
a more
detailed
particular region
of a wall is of interest,
an
seems to
be favourable. In
attempt
to
develop
relatively simple
and transparent
micro model for structural walls will be within the framework of this report. This micro model should be based
suitable for the
on
approach,
modelling
of
uniformly reinforced
concrete structures
structural
walls.
Although clearly
the
meso
some
cases,
especially as a complement to
non
same
analysis,
e.g. for
linearly,
the
defined
macro
the essential
points
of view
Therefore,
report.
The
of a micro model in
well
as
comparison of their
respective dynamic
of
a
behaviour is
The actual
Performance check
use
performed mainly by
the
of the
macro
results in
chapter five,
but discussed in
more
detail in
chapter six.
Based upon
some
findings
the
macro
model,
also
modifications in the
in
are
presented placed on
chapter six.
general
analysis objectives
will be
well
as
presented
graphical form
testing
in
of Performance
of capacity
the discussion
on
16
17
CHAPTER THREE
MACRO MODEL
3.1
Introduction
This
Simulation of the
seismic action.
macro
subjected
to
Springs
and
belongs
to
the category of
macro
models.
belonging
nonlinear
to
the category of
models discussed in
beams
a was
chapter
two
on
found
to
chapter,
model of this
functioning numerical
use
subsequent chapters of
was
al
[KSOA82], and
shown in
figure
used
during
the
US-Japan cooperative
to
earthquake research
a
program
during
provide
to
simple
analysis
multi-storey
and
subjected
earthquake
sections.
developments
recently
by
briefly
discussed in the
following
In this
chapter
the
geometrie considerations
are
short
review of
previous
work
by
simplified
steel, combined with observations of physical behaviour, the nonlinear flexural properties
are
developed.
are
treated
separately. Simple
and efficient
hysteretic
rules
developed largely
based upon
ofthe
stiffness matrix of a
macro
element is
18
CHAPTER THREE
3.2 Model
Configuration
model
The
original
by Kabeyasawa
et
nonlinear
beams. The
Springs
were
made up
as
follows:
two
one
vertical outer
boundary columns,
one
central vertical
spring, representing
central horizontal
one
spring representing
rotational
finally
central
spring,
central
the base.
Each
one
seven
test
specimen
an
in Tsukuba,
as a
presented in
chapter two,
as
modelled
by
a set
of Springs to form
element used
storey model
shown in
figure 3.1.
In later
developments
was
examples
aecuracy,
discretised into
than
one
e.g.
[BWL92a].
VtfSSmWjWjWjM/mW^^^
svWMMMMMWMm)m/Mmmi/mM//nuM &
Figure
The
the
3.1
was
essentially used in
numerical
analyses for
prediction
cyclic
and
pseudo dynamic
Tsukuba, and the scale models in the US-Japan cooperative research program.
The attempt to
manner
led to
compatibility problems, discussed by Vulcano and Bertero [VB87] and Linde [Lind88],
[Lind89]. These difficulties arise mainly when flexural and axial properties
to
are
assigned
Springs, respectively,
base
on
as
since these
assigned properties
the
MACRO MODEL
19
were
made in
[VB87]
to correct this
problem by assigning
this Solution
was
properties
to
the rotational
seems
spring. However,
not
reliable
efficient. It also
difficult to
explain
the
softening stiffness
physically.
Furthermore, it
was
attempted
in
[VB87]
to
Springs by
spring assembly
was
yielding.
This
The
represented by
on
parallel
and in-series
figure 3.2a.
single spring
while the
parallel Springs
a
below
model cracked
concrete
respectively.
The steel
spring follows
or
bilinear curve,
(cracked State)
takes up action
(closing
physical behaviour rather than employing hysteretic models for the composite
reinforced
concrete.
^^^^^^^^^^^^^^^^jj^.y^^^^^A
mm
WMM/SM. >jS,}s>;//sa zzzzzz
[VBC88Jand[FF91]
Fischinger [FF91]
to
then
replaced
the rotational
spring by
Springs
This method
able
to
simulate the
more
complicated
Generally,
20
CHAPTER THREE
to
develop simple
model. In
suggested
static
loading.
In this
and
model will be
developed,
which is
based
on a
shown in
figure 3.3.
arrangement is
spring
and to
perform
derivation of the
axial and flexural
properties
for the
remaining
three vertical
Springs
so as to
satisfy both
spring, modelling
making
in
figure
3.3.
spring arrangement
in order to
figure
simplieity by using as
Springs
as
possible.
next
behaviour in the
y^/////////////////mV///j/jj///^
bMMMWMWWMWM>m>/JM/MWM/Vm/f/.
Figure
3.3
quite accurately
central vertical
with
Springs,
spring. Since
connecting
the nonlinear
same as
Springs
are
flexurally
more
rigid,
the kinematic
possibilities
essentially the
provide
an
efficient
MACRO MODEL
21
was
shown in
figure 3.3,
we
properties for
the elastic flexural behaviour. Axial behaviour, inelastic flexural behaviour and shear behaviour will all be treated in
we
Springs
at our
disposal,
and
let them simulate the flexural behaviour of the entire wall section, i.e. the web of the wall,
and
boundary
columns
area.
preferably figure
for
accomplished by comparing
model. For this purpose,
we
shown in
3.4a, which
we
refer
to as
obeys simple
elastic
theory
figure
////////YY///////
V
Figure 3.4
By expressing
1
b) Model
a) Elastic theory
Kinematics
ofwall rotation for uniform moment distribution displacement at the top of the wall for the two
some
walls in
figures. 3.4a
and b,
derive
height h, length /,
cross
sectional
area
A,
of inertia /, and
Young's modulus E.
22
CHAPTER THREE
we
may then,
by prescribing
the
same
rotation and
to
displacement
area
as
the
assuming
as
outer
Springs
be /, obtain the
of the
Springs, As,
well
hc.
spring
obtain
only
flexural deformations.
For
figure
3.4a
we
obtain
R-Mh
e-~kT
(3-D
8h=^*
2EI
(3.2)
and for
=,
-,
and
*=
LA
we
get
Mhhc
Sk-ehc=^tl2A.E
(3.4)
to
(3.3),
and
(3.2)
to
(3.4)
we
properties
K--w l2
<3-5>
K~\
representing
the outer
(3-6)
spring
area
respectively.
Ku
of the outer
spring
is then
given by
*=*By replacing
the moment M with
a
<3-7)
shear force V,
as seen
acting
at
in
figure 3.5
gradient.
we
check
on
the
MACRO MODEL
23
/YYYYYYYY////////
\
Figure 3.5
'
a) Elastic theory
b) Model
Kinematics
figure
3.5
we
by elastic theory
Vh2
=
2EI
(3.8)
Vh3
=
3EI
(3.9)
2/
(3.10)
=
v
Fh
V*1*1
=
AEI
(3.11)
e=2A=Yh?_
l 2EI
(3.12)
8h
dhc
}2L
AEI
(3.13)
24
CHAPTER THREE
Comparison
This is due
account
of
equation 3.8
to
(equation 3.13).
to
the fact that the model derived for pure flexure is not able
accurately
be reduced
for a
moment
This
by discretization using
shear force V only,
[Lind88], [Lind89].
derive the
quantities A.
in
hc using
the
case
with
acting on top
of the wall,
as
figure 3.5.
according
to
for elastic
theory.
obtain
F.
^=
(3.14)
*=TF
^!
IA.E
(3'15)
e=2A
l
2VhhL ?A
Sh
Bhe
-^ir
PA
IVhh]
(3-17)
Equating
4Ht 3/2
the
expression
3.8
to
to
3.17,
we
(3-18)
K=2J
This shows that for the
case
(3.19)
of shear force
only
with moment
gradient
the centre of
case
relative rotation would be located lower (hl3 from the base) than for the flexure (h/2 from the base). This
moment
means
of pure
have
that in
general
and the
case area
where
we
always
hc,
of the
outer
Springs As
reality
we
always
acting
at a
wall section,
The
possibility would be
would
to
cases.
simplest
which
two cases,
give
MACRO MODEL
25
A<
[f f]/2 f|
+
=
(3.21)
more
sophisticated
properties according
to
to
expected
possibly according
the
instantaneous ratio.
A*
al2+b%
4/
l-
3/2
(3.22)
hc
a^+b^
4
(3.23)
a +
(3.24)
is
not
to
be described.
assumptions
are
only valid
we are
to
are
acting
the top,
as
reduee the
shear action,
more
considerations will,
discussed above.
will see, be
important,
analytical differences
The
two outer
spring stiffness,
spring,
each
is needed to
In
figure 3.6,
cause
are
subjected to
theory we obtain
*.~ EAmm,
(3.25)
26
CHAPTER THREE
E, As J
V///////////////
a) Elastic theory
b) Model
Behaviour under normal force
Figure 3.6
and for the model
we
correspondingly obtain
N
N
-
"'
IKst + Kcs
~2EAS
(3.26)
_
spring. By
between
one outer
spring
and the
As I Am,,
the elastic
spring
spring may
be written
Ka
EAw
l-2a
(3.27)
Together with
the outer
Springs
this
spring
accounts
the
complete elastic
properties beyond
cross
region
it is necessary to
we
smdy
the
physical
behaviour of the
section of
establish
cross
a moment-curvature
a
relation, by
means
see
of
wall
and
section into
figure
concrete
obey their
shown in
figures
3.8 and
MACRO MODEL
27
Sk_ Centroid of
cross
section
Fibre
no.
0.012
Ec
Figure 3.8
For the concrete
we
use
the relation
as
given by
compressive behaviour,
stated
follows
0.002
U.002J
(3.28)
/c'[l-Z(ec -0.002)]
(3.29)
0.5
50-50*-0.002
(3.30)
28
CHAPTER THREE
where eSOu and 50A represent the strain the unconfined and confined
at
50%
strength
seen
on
the
descending
branches of
(hooped)
concrete,
in
Figure
3.7. The
following
50u
3+0.002/; /;-iooo
(3.31)
0.75^-4(3.32)
bh
hoops, and sh
is the
spacing
of
the
hoops,
and
psi.
No difference in
strength
is assumed
was
although
an
found
by
some
researchers.
Cracking
steel
was
having
0.0001.
For the
reinforcing
bilinear elastic
was
used, shown in figure 3.9. The hardening ratio may be chosen by the
In the fibre
a
area
steel
area
proportional
to
For the
boundary
elements
a more
refined
modelling is possible
fibre.
so
that
essentially
each
single
fu
'
fy
Experiment
0.05
0.10
0.13
MACRO MODEL
29
cross
employs
at
an
incremental
as a
iterative
procedure
described uniform
as
follows.
Any gravity
load is
applied
first
normal
force, resulting in
state
the
incremental
compressive
edge
is
set to a
compressive increment
edges,
produce
force, calculated
as
their respective
on
particular fibre
normal
section
giving a resulting
acting on
the
cross
section is
force and
follows
equilibrium
%V+ %%-**%
i
(3-33)
where the first and second terms represent the concrete and steel fibre forces
respectively,
obtained
F=Ae<t
F.s
(3.34)
Alal
larger than a preselected value,
(3.35)
If this residual is
adjusted,
a
and
a new
iteration is
performed, with
the
compressive multiplied
a
kept fixed,
until
sufficiently
In addition to
summing
are
also
resulting
bending
moment which is
acting
the
cross
section due
to
/r*i + 2X*.
i
i
(3-36)
as seen
in
figure 3.7.
30
CHAPTER THREE
Theoretical
curve
Simplified trilinear
curve, fitted to theo
retical
curve
8C
Sy
cross
at
procedure
length.
edge strain
in
is increased
by
new
manner we
obtain
cross
moment curvature
flexural behaviour
is
transformed into
between the
the moment
by the distance
Springs.
have
significantly
length
dynamic
behaviour. As
stated in
section 3.3,
F-M
(3.37)
It is then
possible
to
fit
bilinear
see
or
trilinear
curve
for the
spring behaviour
to the
resulting
Figure
loading,
unloading
break
and
points
[Lind89].
MACRO MODEL
31
*,A
around O.Ol K.
Elastic
stiffness in compression, Ke
curve
forflexural Springs
simplify the approach further by assuming that the wall is to some degree
at
cracked
have
a
(flexurally)
the
Thus due to
zero
normal force
we
(uncracked) stiffness.
0.8
ranging
from 0.5
to
give reasonable
to
agreement with
same
ratio of cracked
than
as
attempting to obtain
[Lind89], and may
the
direcy
from
relation
in
be
justified due
a
to
behaviour at
In
flexural crack
flexural cracks
reality,
only
a
occur at some
stiffening
kept
until
yielding
oceurs.
The
yield
moment curvature
from the
procedure
by using
preselected
steel and
read the
yield
curvature
from the
strain
hardening of the
is
set to some
yield stiffness
chosen
by the user.
skeleton
Values
a
value of around
percent
was
found
to
be reasonable. The
as
suggested
3.11.
curve
shown in
figure
32
CHAPTER THREE
10~KCS
Sa
Elastic
stiffness in compression,
K.
es
Figure 3.12
The central vertical
moment
Behaviour
spring is active
cross
in
compression only,
to
consequeny
a zero
when the
acting
on
the
section is
sufficientiy large
produce
spring is
set to zero.
A small value is
assigned to
stiffness in order
to
spring is
shown in
figure 3.12.
large
through
that outer
a
by a cross
section with
zero
plus the
spring
and centroid,
i.e.
M"
M+
(3.38)
a cross zero
M denotes
which
for
a cross
section with
approximation
to one
implies aetually
hence
compressive
force is close
edge,
which is
be the
case
approximation
on
the
position
Generally
expression "low
to
understood in such
macro
way that
no
model is to
MACRO MODEL
33
500-T(ton)
element elongation
over
in order
to
limit
com
pressive nonlinearities,
With the above the monotonic behaviour of the wall model is established. We have treated the skeleton
curve
yield stiffness,
as seen
figure
and
3.11. We will
now
determine the
loading.
attempt
explain
the
unloading
as
the
outer
vertical
specimen
versus
Tsukuba
[ACISP84],
in the form of
boundary
elongation
base shear. If
to
we assume
cyclic
the
testing,
test
proportional
serve
the
spring
spring,
in the
figure 3.13
the
elongation
over
versus
seven-storey wall
example is shown.
by Kabeyasawa Springs,
al
The
original
model
et
[KSOA82]
used
relatively complicated
hysteretic
which is shown in
figure 3.14.
34
CHAPTER THREE
<D,.F_>
0,
Dve
=(D,-3yc)
Figure 3.14 Original hysteretic model for outer Springs by Kabeyasawa et al [KSOA82]
[Yane82]
yielding
in
tensien
t
(l-)h
"r0 /.in compression
TE A
c
element
E A
ss
t'NiEA/EA
c c
(s)
s
L
1
L (c)
'
.element
1 h
<c
A1
I
|eac
t"c
ui[1<y"'-i]
& Bertero
[VB87J
MACRO MODEL
35
Foive
Fi,
(k.
Figure 3.17 Modified hysteretic model for outer Springs by Linde [Lind89]
f
(MX
uy).
hysteretic
3.15
Springs
and
proposals for
on
modifications
these models
are
shown in
figures
to
see
the
appropriate references.
were
prescribed deformations
wall tested
were
program, such
or
as
the
structure
at
tested in Tsukuba
[ACISP84]
tested
displacement
histories
36
CHAPTER THREE
The
macro
model
developed
general
the
finite element
code, the dynamic analysis being performed by direct integration methods using general
ground
motion histories
macro
as
input Therefore,
are
the
requirements
on
hysteretic
model
model
Simplieity is needed mainly due to the already heavy computational demands by the
method of
analysis.
very
complicated hysteretic
method.
model every
having
analysis
models
Generally,
residual forces
time. The
requiring equilibrium
iteration which
some
costiy computational
hysteretic
by
suggested by
same
authors,
display several
points
are
determined
expressions.
[Lind89] it
when it may be
is
clearly
necessary from
experimental
data.
Generally, the fewer the stiffness changes of the model the better. Despite the simplieity requirement discussed above
be reliable
more
good hysteretic
simplify it
than to
certain
degree. Otherwise,
impaired.
the outer
We propose here
Springs,
seen
in
figure 3.19,
which It
major characteristics are simplieity and reasonable agreement with experimental results.
is based break
on
the model in
and is made
more
efficient
as
point
in the tensile
of the skeleton
curve
is moved to the
origin,
i.e. it is
already slighy
assumption
a
is
quite
realistic for
existing buildings,
construction due induced
or
since
slight cracking
usually exist
as
to
environment and
possibly traffic-
seismic
ground motion.
some
tensile crack
point
more
model
displays
modified
reloading behaviour as
seen
from
reloading
a
oceurs
direcy
displacement
including
as
compressive branch
3.17
are
the
one
figure
displayed
reloading branches
lot The
cyclic
behaviour
shown in
chapter five to be
MACRO MODEL
37
quite satisfactory
mental data. The
with
regard
to
compared
to
experi
proposed
versus
model in
figure
3.19
displays
the force in
one
Springs
spring.
to
The
an
loading
"skeleton"
curve
elastic
compressive stiffness Ke
terms
subjected
gravity
load in
of normal force,
into
a
during cyclic
certain
behaviour when
one
edge (one
outer
cross
spring) gets
compressive State.
to
In tension the
simplification
prior to
the skeleton
section is considered
be cracked to
extent
analysis
curve.
giving
fraction of around
Behaviour
In tension
prior to yielding is
unloading always
branch
oceurs
point
on
the elastic
compressive
The
flexural
reinforcement steel.
figure on
the
the
The
unloading
at
from
yielding
oceurs
the
unloading
From
yielding point
section
as seen
in the
cross
unloading
moves toa
bythe
38
CHAPTER THREE
Figure
3.20
user,
point
where cracks
are
closing.
This
point
was a
found
to
be
dependent
larger normal
to
require
obtain realistic
experimental
data. A force
For
high
normal forces,
found in the
case
-AFy
for the
case
eight storeys
(normal force
a
on
equal
to
large part of
(passing through
more
the
compressive
internal
at
zone
located
near
the
edge).
moment
in
multi-storey
the tensile
edge) only
contribute
Reloading
maximum and
oceurs
towards
was
point
on
the
yield
curve
where the
displacement
reached earlier,
as seen
figure.
For
unloading
the stiffness is
proposed
a
shown in the
figure,
with stiffness
hysteretic rules,
it is useful to
are
assign
or
shown in
figure 3.20
briefly
described
follows:
MACRO MODEL
39
compression.
the element is
rule 2.
prior yielding.
Unloading leads to rule 4. Further tensile force leads to rule 3 representing yielding.
Rule 3.
Yielding.
prior yielding.
If prior
Unloading
leads to rule 5.
Rule 4.
-aciFy it
leads
to
Rule 5.
to
Rule 6.
Is entered from rule 6. If reloading has occurred which did not lead to further it may be entered from rale 9 at
to
zero
-aciFy it leads
rule 7.
Rule 7.
Reloading.
or
from rule 7. Takes direction towards maximum rule 3. If unloading oceurs, it leads to rule 9.
displacement
to
Rule 8.
Compressive elastic
stiffness.
Rule 9.
40
CHAPTER THREE
4.0
.
"
""
-*
.
""""""
r-
3.0
/'r
2.0
fi
// /''
E
o
fi
li
1.0
aer aer
0.8, ay 0.5, ay
0.02 0.01
fi
Tf
0.0025
0.005
0.0075
0.01
displacement [m]
Figure 321 Influence of outer vertical spring stiffness on global flexural behaviour
The model
spring
was
found to result in
model which
appeared
reasonable
hysteretic loops
such
spring
is in itself of
limited value,
although
some
force
spring
is otherwise
essentially artificial
the
in order to
reproduce
Concerning
Partly
resulting
to
overall flexural in
hysteretic behaviour,
was
original
vertical
seen
figure 3.14,
found
to
give
thin
loops
sometimes.
on
be due to the
unloading stiffness
which is
a
dependent degree
the
maximum
small
from
some
The
proposed
in this report
was
deliberately
chosen to be
partly
partly for
simplieity.
model
The
was
variable location of
points
original
deliberately abandoned,
rules
were
also
deliberately chosen
a
simple
as
possible,
but
as
comprehensive general.
as
necessary to obtain
Before
turning
to
some
remarks
are
made
"global"
MACRO MODEL
41
-imiiiitttimmtttmnRm
Ke(Kcr> fy)
Ke
*?&*.%./>>
Hat.
tt
Y*\YY7,^afr
,
Ka
rg
***>tm.*t-i*ittiti*rr*m*-
\-
IrSection
Figure
3.22 Moment
of inertia made
up
flexural behaviour. We limit the discussion here to the skeleton curve, i.e. to the rules 1
to
3 of figure 3.20,
comprising
yield stiffness.
relation,
simple example is
section with
a
shown in
by
the
to
macro
model. A wall As
a
cross
yield
subjected
pure
bending.
reference,
by the
the
solid line. For this reference the tense stiffness of the outer
uncracked
was set
equal to
compressive stiffness,
obtained, each with
and
a
no
yielding
was
comparison
of the outer
solutions
vertical
were
yielded stiffness
as
Springs.
spring
stiffnesses
follows: cracked: 80 % of
uncracked, yielded: 2 % of uncracked. The second Solution had the stiffhesses: cracked:
50 % of uncracked,
The
yielded:
1 % of uncracked.
figure
moment
divided
by
curvature.
Normalising
the first
unity,
we
obtain for
comparison
Solution
global
global
flexural
yield
obtain
corresponding
by solving
this relation
analytically.
In
figure
3.22
from
distance /. We
of two active
spring
has
"yielded",
or,
we
contribute to the flexural stiffness. For the discussion of the wall section moment of
42
CHAPTER THREE
inertia
we
areas
of the
outer
Springs,
area
spring
spring
is
an
is
proportional to
an
its stiffness
(height h,
unchanged).
We define
idealised
spring area
and
area
the elastic
spring
stiffness
see
changed,
the
spring
A.
changed,
expression
are constant.
The
moment
may then,
with
use
Modifying one of the two idealised spring areas to the cracked value Km.,
moved centroid, located from the
we
obtain
lc
Kerl
(3.40)
and the
L^K^l-lmf + Kmilmf
It is
now
(3.41)
possible
global
moment
of inertia divided
by
elastic
area, and
stiffness
ot
(3.42)
and
a'=-f
i.
t
(3.43)
the latter
basing
on
expressions (3.40)
and
(3.41)
spring
stiffness
Ka
substituted
MACRO MODEL
A3
1
1
a
1
0 0.2
Local
0.4
0.6
0.8
1.0
(spring)
stiffness
ieduction,a<r ,a
Figure 323
reduetion
ofthe numerical
example
shown in
figure 3.21,
to
as
fraction of the
to
3.43
we
0.8:
le =0.444/ le =0.0196/
and and
c= 0.889
yield stiffness, Oy
and for the second
0.02:
a'y= 0.039
comparison Solution
=
0.5:
lc= 0.333/
lc= 0.0099/
and
=0.667
yield stiffness, Oy
all of which show
0.01:
and =0.020
good agreement
for local
be
rule of thumb it is
possible to
the the local
to
State that
As for
be
roughly
10
a^
25 percent
higher than
the
spring stiffness,
with the
(spring) and
global (flexural)
stiffness is
shown in stiffness
spring
of the
figure typically
rest
of the
44
CHAPTER THREE
completely
simulated
by
spring.
In accordance
we
properties
theory,
spring by comparing
in
model
behaviour to that of
real wall
obeying
elastic
as seen
figure
3.24.
'/////////////////;
a) Elastic theory
b) Model
we
wall
as
follows
8*.=
KVh
(3.44)
dependent
empirical value
of
x*
K_
0.75(1 + u)[l-u2(1-v)
[l-3(l-v)]
geometrical parameters
for
a
(3.45)
where the
and
v are
shown in
shear
stiffness
Kve
given by
-GK
K...=
hK
(3.46)
MACRO MODEL
45
l'=uL
l=(l+u)L/2
shear
cracking
oceurs,
whereupon
which
cracked
stiffness is stiffness
employed.
An
empirically-based
Otani et al
was
suggested by
[ACISP84]
follows
,
'
0^ + 0.14
fc
(3.47)
Equation
3.47
gives
for
typical
reinforce
ratios
to
was
b)
to walls with
cracking
force in shear,
cross
sectional
area
degrees,
a
and / is
strength
(3.48) gives
very
simplified
more
as
W.
pJs
(3.49)
where
Kt is
form factor
given by
46
CHAPTER THREE
1.05t/b
(350
where shear
lp
is the
length
an
equivalent
stress
given
as
/*asCfV1+c"
where cr0 is axial
stress
(3"51)
and Q is
given by
C,
=2.2>/J-0.00546/;1-456pw007S
3.49
to
(3.52)
Equations
For
a
3.52
are
taken
source
and
use
units
kg
and
cm.
simplest possible estimation of the ultimate shear capacity we may add the concrete contribution of (3.48) to the horizontal steel contribution of the bars with area A^ crossing the inclined surface,
as
suggested in [Lind88]
(3.53)
Vm
V + V,=AmJ, + AJ,
realistic estimation is
more
provided by
the
[Hiro75]
0.0679a023C+17.6)
"
,
v*
_.
r-
+
m
VM/VX
+ 0.12
0.845^/^0^MJrryH^k + 0.lo"0o
A.
bJ
(3.54)
given in MPa,
are
explained as follows
p,
"=W-tii)
A.
=
(3-55)
area
boundary element,
MI(VL)
f
=
depth ratio,
MACRO MODEL
AI
p^
P~+=TL
bts
s
=
(3.56)
vertical
=
o.
entire wall
cross
section
area
b.
HH)
and forL and h,
see
(3.57)
figure
3.24
Since the
in shear
are
the
done in
[Lind89], followed by
yield
relatively
in
a
occur
region of behaviour
here.
clearly
to
developed
in the
Design
provided
capacity design
method,
see.
e.g.
[PBM90]
or
rPP92].
Concerning
behaviour for
the
is put here
on
modelling
the
unloading
An
large
reason.
the
University
versus
of California,
Berkeley. during
unloading, figure
eycles
it may be
seen
from the
that the
pinching is
not
mueh less
pronouneed.
specimen was
capacity designed and the test programme did not correspond to actions
original
model
by Kabeyasawa
et
al
[KSOA82] employed
was
trilinear
origin-
oriented
modelled
by the shear
spring
alone. The
48
CHAPTER THREE
Figure 326
wall
Shear deformation
tested
at
at
three-storey
specimen,
the
yield part.
The
unloading
3.27.
curves are
always directed
origin.
The
latest unloading
figure
As may be
from
not
give
particularly
realistic Simulation of
typical
large deformations.
The
loops
are too
thin in the
vicinity
of the
origin compared
be
not
direcy origin-oriented.
some
and
possibly,
in addition, exhibit
pinching
design according
to
the
method
(nonlinear
predominany
we
regions
pinching disappears
importance to model.
are not
For the
analysis of general buildings, i.e. including buildings which regard the possibility of strong
and exhibits
capacity designed,
seen
in
right
In
different
means
MACRO MODEL
49
et
al
[KSOA82J
Q
F
i
/
/
'
/
/
/
/
"!
J
Qy-
/^77
rr
^
A
^^
improved hysteretic
model
was
proposed
in
50
CHAPTER THREE
Force
f,
f~
Yr~"
F-
-rfjfjfmjr
j /i.
Dcfocn
mainly
predomiserve
nantiy in
as
developed in
mainly
numerical tools
are
together
with the
deformations
Based
on
avoided.
these
assumptions
use a
a
horizontal with
an
obtain
hysteretic
relatively
designed
model is shown in
few stiffness
its
changes,
walls is modelled
quite realistically.
MACRO MODEL
51
Figure 330
For the
we use
equation (3.49)
and used
or
alternatively
limit on the
as was
readily calculated
not
cracked
be exceeded
during
numerical
analysis.
The
macro
model
developed
in the
previous
and
sections of this
chapter
is to be used in
see
[Abaq89]
general
introduced into
over a
completely
typical
wall behaviour
to a macro
chosen
height
of
referred
element
The
generally formulated
element
equation
of motion
terms on
representing
the
macro
letters for
spring
and
properties
by bold
52
CHAPTER THREE
scalar
properties by
as
follows
f,+fd+f.=fg
where the terms
are
(3.58)
given by
(3.58a)
(3.58b)
f,=ku fd=c
/.
(3.58c) (3.58d)
tn
ft=ag(t)m
where k, c, and
displacement velocity
the
right
of
hand side
dynamic force acting on the element's degrees freedom, where ag(t) represents the general ground acceleration.
In this section
we concentrate on
expression represents
Kt.
In
chapter five the modelling of the other two internal forces is discussed.
The nonlinear stiffness the
are
adopted
to
Springs
a
of the
macro
macro
model into
performed in
assembly
of the
Springs
which
modelled
as
2-degree
of freedom
(d.o.f.) elements,
and further
was
by
the
assembly of
as
mainly
adopted is to numerically
Springs
Option
number of undesirable
degrees of freedom
static condensation.
and allows for
appeared,
Apart
of which may be
macro
from
that, the
relatively versatile,
derivation of
a macro
element based
on
rigid beams
may be
an
essentially
acts as a
membrane element,
figure 3.31,
and
MACRO MODEL
53
Ut
Uk9
vv"vWWWl
M6
I ~~Miimwvi'.w
lort^fs
u.
*35&&asa;m3i^^
^2
t*
I Jh
b) Macro element
Figure
which be
degrees offreedom
can
readily used for the numerical modelling of multi-storey structural walls. figure 3.31
ten
On the left in
is shown the
macro
element which
are
was
programmed
and
girders.
The stiffness
to
properties of this
of freedom,
element
are
expressed
corresponding according to
the left in
the
degrees
Springs arranged
shown
on
the
previously described
nonlinear
figure 3.31.
macro
The
the
resulting
essentially
as a
membrane element,
seen on
top. The
d.o.f.
macro
element
as
few
pertaining
internal
to
the wall
cross
section
moment.
spring properties
need to be
given
input
spring
model
By expressing rigid
(e.g.
at
Springs)
d.o.f.'s which
are
displayed
figure 3.31,
we
normally
not
be
we
displacement
displacements
are
of each d.o.f.
to zero.
equal
to
unity,
time
keeping
a State
of
equilibrium
to
the
54
CHAPTER THREE
displacement
was set to
unity. unity.
We thus have
ky,
the column of
displacement uj,
set to
begin the
treated
separately later.
are
change
the
final d.o.f.
d.o.f.'s
begirining.
spring by Ks,
spring by K\,
the
right
outer
spring by K%
spring by 3,
and
by Kapreviously described
"rigid",
are
The beams
which
as
were
as
are
for
flexurally
rigid.
frame-wall interaction
application
macro
corner
of the
horizontal
degree of freedom
chosen
at
each end,
to
are
numerically is
wall.
so as
behaviour of the
an
one
equal
to
unity,
i.e. u\ of
1,
as
shown in
figure 3.32.
macro
spring forces
the disassembled
Figure
3.32
MACRO MODEL
55
*A
lower assembly
-wmm
KAD
KAD
Axial forces
Figure 333b
acting
on
lower
assembly,
u\
-&?) \
rl
-bi*?:
K.-
K.
ui
1,
we
terms
*h! *a
rr
"-r
^f(
*n=0
.
%
=
_Ksc
2 l
rr
*S1
k41
K.
~Ka+ A
2 /
k5l
*io,i=0
(3.59)
56
CHAPTER THREE
case
when 2
are
obtained, since
and the
Springs.
figure 3.34,
spring
when 2
are
shown in
figures 3.35a-h.
2=1
Figure
3.34
Deformed State
at 2
I(1)
Figure
3.35a Internal
lower
assembly, 2=1
lower
assembly, 2=1
Figure
spring
3.35c
lower
assembly,, 2=1
MACRO MODEL
57
lower assembly,
2=1
I* (1)
K S-
K
Figure 335e
upper
Internal springforces
acting on
assembly, 2=1
JC (1)
f (1)
upper assembly,
2=1
58
CHAPTER THREE
^-4
m
*4 f
K
Figure 335g Spring force from central vertical spring
and its reactions,
on
upper
assembly, 2=1
iii
K.
c
h-c\
l
)
1
( cY h-c
H7A/JJ
K.
c
JWMH1H1MUUIBB
2 /
2 /
upper assembly,
2=1
corresponding
to 2
as
follows,
cases
created
by superposition
of die
figures
3.35a-h.
^~
fr
Kj.9.
2 /
I-
K'
**"
2 /
*52=-^y
(3.60)
MACRO MODEL
59
degrees of freedom
obtained in
similar way
as
for the
two cases
was
by
kij
(3.61)
which
applies due to symmetry of the stiffness matrix, was used rather as control. degrees of freedom,
seen
in
be treated. We
employ
7 and
an
only
degrees
of freedom 6,
only
joimng girders
calibrated
to
analysis
properties
may be
by girders.
In this report,
only
studied and
no
such calibration
was
performed.
of
possibities
modelling the
girder
or
general formulation
macro
properties
element
figure
as
they
refer
the
macro
element of
in
figure 3.31.
book.
kn,, for
equation 3.62.
text
given here,
7^
Figure 336
"io/'Tn
12EI
6EI
Y2EI
6EI
/3
l2
AEI
/3
6EI
l2
2EI
kib
l
=
l2
\2EI
6EI
(3.62)
/3
l2
AEI
Sym
60
CHAPTER THREE
The blown up stiffness matrix of the internal beam, i.e. (3.62) assembled into the
macro
as
follows
12EI
67
0 AEI
127
6EI
/3
~~7~
6EI
~-T
im
~T
~T~
0 6EI
0
127
Sym
~7~
~~F
ah
~T
Blown up
macro
element with
is
shown
on
the
next
page, followed
Ks
K.
h.1
K,(h-c) K,(h-c)
'
K.
2
/
2
2 l
2
/ l
-KA+0
~^K3
'
5
s
o
*i
2 /
*+*.[j
4f
^(c)(h-c)
-~K-{i)
2 l
~
7
Ks(h-c)
2 l
^-<tJ
-4--(^
0
~
'
Ks
5l
5l
4
2
l
**
+
K'[l
2
l
0
2
l
-^--r
A-j
(cV/i-c)
K'\T\
KAh-c)
/
l
t-k{t)
K,(h-c)
2 /
'K2~T
K+-K*
+
Ks
i-
Ki
0
-
h-c\2
2
K, (h
c)
l
K,
rrfh-c)2
Ks
Ka+ A
Ks(h-c)
4
Sym
K*
A-3
(h-c\2
T+K{)
Blown up stiffness matrix for lower and upper assemblies of macro element
of freedom
OS
K,
K'C
K,
A
.
m,K'
K.
4 l
4
K,(h-c)
2 2
l
TT
K,
c
0
0
K,
K. (h-c)
as
Ksc K.c
~
K3
(c\2
"TT
2
K>+T+K{jj
-.-''*<7)V
0
T
KAh-c)
2
l
Ka
Kt
K.c
K.
4 / 4 2
JC, (A-c)
0
Ks
"TT
*3
c
K2
-2-rK,
(tJ
K, K3
4
"TT
TT
K, K,
a
(cMh-c) K'{l) l
0
A-,
c
-<>-'; *<W'
KAh-c)
2 /
KAh-c)
2 l
K,
(h-cY
\2EI 6 EI
KAh-c)
2 /
12EI
6EI
(h-c}2 { l )
6EI
1
2E1
4 EI
~T~
~T
K,
KAh-c)
2
/
Sym
Ki
+ K.
f/i-cV
Kj-r-f
12EI
6EI
J+
f
AEI
1
~T
.
1
including rotational degrees of freedom
Complete
MACRO MODEL
63
3.8
Ductility
demand
The
of certain
preselected
critical
zones
of a structure is The
one
of is
the most
important
in
observations to be made
during
nonlinear
on
analysis.
designer
assumptions
which the
design was
to take
based.
Especially
plastic hinge
detailing
yielding is intended
place, the
this
ductility
must
readily
computation of
nonlinear time
history analyses.
definitions
as
ductility
is
make
some
follows. Rotation 0, in
general,
Q=Svl~ Svi
l where
(3.63)
8vl
and
8^
are
the vertical
displacements of the
in
one
outer
Springs. We
as
rotation at which
yielding begins
of the outer
Springs
the
yield rotation,
figure
yielding, is
8^-8
e>=~i~
(3.64)
l^j^j^^^^^^lfe^h?*lft'ifc^>?1li!^.>>jl-l
d) Rotational ductility
b) Curvature ductility
64
CHAPTER THREE
larger
than the
yield rotation
the ratio of 6 to
dy
is defined
as
the
ductility.
(3.65)
the vertical
displacements
curvature.
For
an
element
we
then obtain
(3.66)
and
^=-f
whereby the equation 3.66 may also be obtained as
(3.67)
l^
ductility according
to
(3.68)
Curvature
expression 3.67
is
implemented
stored.
to
65
CHAPTER FOUR
MICRO MODEL
4.1
Introduction
mainly
act
in their
own
to
the
stress relation
implemented in
as
modelling
concrete,
well
as
interaction,
so
composite
more more
of continuum
a macro
separate
treatment
of the constituents is
The nonlinear
global mainly
behaviour of reinforced
concrete
as
structural walls is
yielding
of
the reinforcement. A
simple
phenomena.
among micro models two
chapter two,
there
are
modelling
the
cracking
approach,
the
promising
is based
for the
modelling
The smeared
approach
element
on
assumption
single physical
an
crack is simulated
area
by
cracking
within
area, e.g.
as
an
associated with
with
an
integration point,
is simulated
fictitious
early
commonly
when
to zero
balancing
on
the
e.g.
[Cerv70], and
[Chen82].
The
Performance
of
tension
strength. The
al
improvements, such
or
as
(following
principal stress)
making
approach
structures.
competitive
modelling
of reinforced concrete
66
CHAPTER FOUR
Reinforcement bars
modelled
as a
layer
Concrete
element
chapter is devoted
to
the
development of a micro
employing
model,
on a
as a
complement to
the
macro
composite element
see
figure 4.1,
with separate treatment ofthe concrete and the interaction effects. used with any
developed
the
here could
essentially be
plane
stress
smdy,
were
isoparametric plane
from the
Software tis
[Abaq91]
chapter.
In the numerical
including
the
seen
in
figure 4.2b
briefly introduced
to
Appendix
D.
The numerical
integration is
carried out
according
Standard
integration procedures
JBTDBdV
strain-displacement matrix,
(4.1)
see
e.g.
[Zien77]
or
provided by
see
the Software, and will not be discussed in any further detail in this and
study,
[Abaq89]
[Abaq91].
the
We wl rather concentrate
on
coded
and is
given
foUows
MICRO MODEL
67
da = Dde
(4.2)
where d denotes incremental stress and strain. The different constituents contribute to the
as
follows:
(4.3)
concrete,
where
Dc represents
Ds
finally Du represents
physical phenomena
the inter
which
are
development
of
approach
the
global
behaviour of walls,
limit the
modelling
the
as
micro level to
some
of the most
important phenomena
softening,
ie
unloading
For the
compressive behaviour of concrete many material models have been proposed, mainly
be divided into linear and nonlinear elastic models, and models based
which may
68
CHAPTER FOUR
on
plasticity.
given
e.g. in
the
compressive region
structural wall
and there is
acceptable
for many
stresses,
designs,
during
an
analysis
control is
reassurance
that
they ideally
is
compressive
excess
strength,
the
to
which level
linearity
oceurs.
commonly observed,
of
compressive strength
the
compressive nonlinearity
mostiy affects
In the
following presentation
will
now
develop
fully isotropic,
is
given by
da.
1
c
dE.
doA=
der. ~y)
1-VJ
0
1-v
dy
>
(4.4)
~2~. dexy
where
Ec represents Young's
concrete.
is the
Although
strength,
displays
well.
small
noniinearity
in tension
reaching
the tense
the relation
cracking
study
Upon reaching
Solution to this
assumed that upon
certain tense
was
a
strength
the relation
(4.4)
must be
changed.
simple
It
was
problem
presented
in the
reaching
a
direction
to
perpendicular to
the
principal
cracking
was
assumed
parallel layers,
and furthermore
assumption
was
made that
layers.
The
physical interpretation
figure 4.3.
direction
in
one
was
given by ie
EJCC
(4.5)
MICRO MODEL
69
T=0
Figure 4.3 Cracked concrete without shear transfer across crack planes [Cerv70]
where
Ec
as
is the uncracked
Young's
transformation matrix
written
COS
0 (4.6)
sin20
cos
0 sin 0
where
<f>
denotes the
angle
between the
global
plane,
as seen
in
figure 4.3.
results for
simple problems.
was
For cracks in
second direction,
usually simply
set
equal
to zero
[Chen82].
simplification
planes.
crack
problems
due to
evenly
deficiency
however
neighbouring
orthogonally
chapter.
(4.3)
Another
was
simplification
State as
in relation
as
in
originating
to
using
some
procedure.
70
CHAPTER FOUR
ExperimentaUy, however,
the concrete
reaching
certain
peak stress, by
among
displays
studied
al
softening
zone.
Various behaviour.
simplified
suggested
to
numericaUy
simulate this
more
softening
elaborate
linear
or
bilinear models,
multilinear models
also
[RR92]
proposed
nonlinear
we use a
assumes
linear elastic
as
strength
is reached, followed
by
linear
softening branch,
in the
figure
largest principal
stress
strength of
orthotropic.
still be carried
the tension
a
be transferred. The
the tension
figure
4.5.
MICRO MODEL
71
'""
*'
;#S
'*
L+Le
Figure 4.5
the
Linear tension
length L
we use a
a
Here
to
tense
strength ft
calculated
as
0.0001,
value
commonly observed
on
weU
as
Ec, may
to some
not
degree be dependent
treated here.
see
The
softening
curve
Gf
an
as
foows. The
area
under the
representing
we
always
an area
associated with
integration point,
to
we
make
sure
correspond
the
same
in
real structure.
According
to
softening
follows
-kc
Ec{
LJ
a
(4.7)
where L represents
and X is
given by
>.-**&f.
where
zone
(4.8)
Gf
representing
see
the
area
versus
process
fictitious crack,
figure 4.6,
corresponding to
observed
experimental data.
72
CHAPTER FOUR
.iS'V&H
gc^j$s^wftg
if
Figure 4.6
Fracture energy
as area
under stress
versus
[CPE90J
Gf
jcrdw
L is here assumed to be
one
(4.9)
The
length
of the sides of
areas
area
which is close to
quadratic.
For other
integration
and element
equivalent
and Ottosen in
[DO90]
strength
as
by the
on
transverse stress,
experimental
presented
behaviour,
see
e.g.
[Kupf69].
involved
by Ottosen,
with the
e.g. review in
[Chen82]. These
and how it is
affected
by the
we
Since
phenomena,
severe
designed
in such
crushing
or
other
role,
we are
able to
essentially
focus
interest
on
the tension/tension
region
and to
some extent
e.g.
Kupfer
[Kupf69].
We will
use a
simplified cracking
criterion shown in
Kupfer's test
For each
is assumed to
when die
MICRO MODEL
73
tensile
strength
is reached. A
in the tension/
da=D,d
da'
'-KK
.=
0"
0
dEx de2
(4.11)
da2 dan
where
0
0
Ec
0
G. den
D,
Dn
D2l Dn
have been
and
suggested by
authors
as a
small value
involving
is not
Dn.
However since
zero.
the
fuy
clear
we set
these terms
equal to global
The
equation (4.10)
coordinate
system
as
Dg=QD.QT
(4.12)
74
CHAPTER FOUR
where
Dg
and
is
a trans
formation matrix, shown in (4.13), axis and the normal to the crack
containing 0
the
angle
between the
global
x-
plane.
-sin 20
COS20
sin20
cos20
-sin 20
V2
sin 20
sin20
sin 20
V2
cos
(4.13)
L V2
V2
20
see
figure 4.5,
the tension
softening
does
not act
any
opened
case we
obtain the
relation
0"
0
dex
d2
da,
da.
12
0 0
Ec
0
(4.14)
G. .dn.
compared
to
Unloading behaviour
uniaxial
number of
experimental
such
results from
stresses.
out
with
alternating
that the
seen
tensile and
a
compressive
a
Figure
stress-displacement
[RC80]. It is
behaviour of
test
carried
a
by
seen
unloading foows
that
somewhat
stiffness
path
than the
high compressive
during unloading,
i.e. the closure of cracks, is reached further out in the tensile for
displacement region
for eycles with
eycles which
place
have reached
at a
large
maximum
displacements.
This
compressive
In
stress
a
figure 4.8b
by
Curbach
[Curb87]
shown in
figure 4.8a.
stress at
is modeed weU in
figure 4.8b
as
weU
as
the
compressive
modelled.
crack closure, but the residual tense strain at crack closure is not this
Admittedly,
proposal
has the
advantage
of
eycles.
for numerical models of cyclic behaviour may be found in the
Some other
proposals
unloading
softening
branch back
to
the pro-
origin,
again
MICRO MODEL
75
IN/tam'.
O.N/mm'1
.A
2.0
1.0
4
/J/7K
^"
..l:
2.0
1.0
K
^
US
-1.0
-lOl
Wl
-zo
-2.0
3.01
C
-10 23
40
63
BI0
V10
12
20
40
60
80
100
120
l/jml
unloading
behaviour.
Relatively complex
are
focal
point
models, attempting
closely
follow
experimental data,
further evaluated in
[GyU89],
residual
We here propose
account
strain
at
figure
only,
for
simplieity,
simpler
softening region
linearly
and
a
in
direction towards
at a stress
level -acc
-kccft,
peak
unloading. By setting
data
was
factor kcc
experimental
Variation of the
result from
as seen
figure 4.8a,
displacement (strain)
crack
simple
model
presented
regard.
The fact that
we
already
to
at zero stress
level,
and
as
is often
assumed in
theory,
physically
be
explained
be due to
crushing
spaUing during
of
aggregate
closure
on
of
compressive
stress
the
procedure.
oceurs
elastically. Reloading
the energy
foows the
same
path
back
to
p. Thereby,
we
deliberately neglect
during cyclic
action between crack closure and the strain from which the
we
mainly
deal with
76
CHAPTER FOUR
Unloading
Once the
will
larger
than Eo,
oceurs
as
with
zero stress
described above.
no
opened
in that
integration point
tensile stress
occur
region during subsequent eycles. principal direction exceed the tense strength, a tension
0*
0
dx
d2
(4.15)
da2 dan
-KEC
0
G. dn
are
the first
opened crack
cause a
orthogonal
to
the first
one.
proposed by
modelling
[deBN85].
in the
An alternative to
crack
is to aow for
cracking
orthogonal
degree
correct
acting,
begin
the
cracking
would be ie
case
based
only
upon
orthogonal
stress. A
further
possibility
MICRO MODEL
77
keep
orthogonal system,
was
continuously
principal
does
directions
shown
by
Rots
change its direction at the crack tip, the concept of continuously rotating the entire existing
cracks
seems
problematic
from
physical point
of view,
although
it has sometimes
here
proven to
will be
with
experimental
data. The
modelling presented
kept
realistically
4.3
Aggregate
Interlock
In cracks
undergoing opening,
to as
as
well
as
in
opened cracks,
there is
some
shear trans
several contributing parts; cohesion, friction between the misfit due to offset surfaces. Dowel action
phenomena associated
and Walraven
presented
in work
by Pruijssers [Prui88]
rWalr80] among
Numerically, by reducing
factor. In the
between
the combined effects of aggregate interlock, have been taken into account
by
an amount
referred
to as
early
was set to an
zero
in
equal
to zero,
i.e.
no
cracks,
as
Glemberg
and Samuelsson
[GS83] proposed
shear
proportional
so as to
to
softening
Young's modulus,
Kolmar retention. A
obtain
a
moderating effect
of different
on
equilibrium equations.
representing
the shear
[Kolm85] presents
variety
proposals
across
for
physical interpretation
with
of shear transfer
cracks is shown in
figure
4.10.
Generally,
with
good agreement
for the shear
opening strain,
retention factor upon
[Rots84]
derived
expressions
function of the
experimental
work
by Paulay
78
CHAPTER FOUR
yG
Figure 4.10
1
=
l + 4447e.
(4.16)
The second
as
expression is
by
foows
1
=
4762
1346^
curves
(4.17)
where is the strain in the direction normal to the crack The and
of equations (4.16)
(4.17)
are
shown in
figure 4.11.
experimental study
on a
mixed mode
(Mode I
opening mode
and
see
e.g. Hassanzadeh
[Hass91], [Hass92],
Mohamed
[Noor92].
of path
dependency
between
generally concluded
taking into
was
expression
ratio I y
shear strain
well
as
where the
3 gave
a curve
lower
retention for
given
expression
with
P
1
=
PE.+1
(4.18)
2500
=
0.14
0.76-0.16-^- 1-exp
7
H>^
\Y\
(4.19)
'"/
MICRO MODEL
79
10
3
i
1
1
Rots
0.8
1
1
1 i
0.6
1
0.4
tt
\\
0.2
\\ \
>
^^ZZZmmT
i
4
<
10
10"3
1.0
Ornax=l9,T m
I
0.9
Enr
Y
1
0.8 0.7
0.6 0.5 O.C 0.3
3
in
li
II
ll
0.2
Q1
k
>.
Vv
0
1
Enn-10
Dmax
largest aggregate.
to
For
some constant
ratios of
7%
the
expression
Here
is
plotted
for
Dma equal
19 mm, in
figure
4.12.
we
mainly
on
use a
[Rots84], based
the work
by Paulay
proposals presented
the
global
briefly
chapter
five.
80
CHAPTER FOUR
such
as
is the
case
during earthquake
action, the shear retention factor should gradually decrease with the number of eycles,
due to
grinding effects.
was
It
is, however, unclear how this decrease would affect the relation
4.16, which
unloading
path
shear
experiments
one
to
be
higher
but
than the
loading
modulus. back
as
This would
indicated take
imply
the
that for
unloading,
seen
same curve
by
curve
by Rots,
in
figure 4.11,
different
which would
in his
to the
some convex
shape.
This effect
was
by Skrikerud [Skri82]
may not be
direcy applicable
an area
problem
more
where
research is needed.
use
We
equation
4.16 therefore in
an
as
we
as
for
cyclic
we
behaviour. For the closed crack branch with elastic almost fll shear retention,
or
employ here
rather,
factor of 90 % of the
on
isotropic drop
value. The
slight
grinding damage
behaviour. A
cycle.
according
see
to two
by
the the
by
figure
modeUing
correspond
to
give
an
reinforcement within
element
lumped
in such
a manner
that
reinforcing
are
bars
are
connected between
adjacent nodes.
assembled
reinforcing bars
early
modelled
as truss
elements which
are
as
shown in
figure 4.13b.
The discrete
modelling
of
used mueh in
lumping
reinforcement within
an
figure 4.13c,
in which the
as
reinforcing bars
create a
and enter
Smeared
modelling is
more
convenient from
user's
the
properties
maybe
MICRO MODEL
81
Reinforcement
lumped
v
-I
a) Prototype reinforcement
Vertical reinforce
ment
layer.
Concrete
Horizontal reinforce
ment
layer
Figure 4.13
Basic types
of reinforcement models
reinforcement model. Since
us reason to
assume
bars
only horizontally
and
global
proved
to
experimental data,
large
as
long
as
occurrence
of few bars
Drr
[Drr80] suggested
arguing
uniformly
rein
by
single heavier
some
by
82
CHAPTER FOUR
nonlinear bond
slip Springs
to
be
applied,
as
Since in this
study
we
on
global behaviour
vertically
and
of well-detailed and
uniformly reinforced walls, the smeared reinforcement model is used for the modelling
all reinforcement. Different reinforcement ratios
of
generay applied,
mesh.
Ds
to
vertical direction
only,
obtained
as
PA D.=
o
PA
0
(4.20)
where p denotes the reinforcement ratio. It is noted that and the smeared reinforcement
acts as trusses
Ds
in
contains
no
coupling
terms,
elastic-linearly
hardening
figure 4.14.
-i
Of E
r
7
k
yC-S
A
i
hardening
ratio ay may be
given by
the
user.
For
compressive yield
stress and
hardening
are
assumed to the
a
be the
region.
hardening branch,
to
wl be foowed
eycles,
which
more or
less
corresponds
kinematic
hardening.
MICRO MODEL
83
experimental
subjected
where
to
cyclic
action,
at
as
yielding
eycles
The interaction between reinforcement steel and concrete may be divided into two main
categories:
stifferiing
involving
bond
opened cracks.
the "tension
cracking
thus
treated
oceurs, the concrete located between the cracks still acts under tense stress and
phenomenon was
to treat
by
Bachmann
tension and
stiffening mainly
steel,
a
were
by Cervenka
al [CPE90],
the stress-strain
or
or
by modifying description
lasy, by modelling
most
a
complex
and
involves
of the effect
along
transformation to the
stiffening
reinforcing bars
is shown in
figure
As
Upon
formation of cracks
begins.
tension
softening
yield, whereupon
the tension
stiffening
effect
will
cease to act
drops
effect may be
the stress-
modify
shown in
figure 4.16,
stepwise
given
stiffening
In order to
simplify
shown in with
a
our
modeing
for
the
figure
a
4.16 may be
smoothed,
as
figure
attempted
al
nonlinear tension
et
stiffening path
long tail,
example Hayami
et
[HMM91], Ohomori
al
acts
mainly in
the direction of
the reinforcement, and the best way would be to add the effect
to
treat
the effect
numericaUy or
the reinforcement modulus matrix. However, it is also found that the ten-
84
CHAPTER FOUR
er
Figure 4.15
Tension
Figure 4.16
sion
Tension
stiffening transformed
to
contribute it is
significany
stiffness any
more.
In
quite important to realisticaUy model the behaviour after tension stiffening no longer acts.
For the purposes in this report it appears
to
have the
Option
of
including
the effect in
and it wl be
performed by
modifying
the additional
carrying capacity
of
figures 4.15,
as
shown in
MICRO MODEL
85
model with
use a
point
strength
tension
stiffening
would in fact
By superimposing advantage
the of
a
ie tension
stiffening effect on
the tension
softening,
we
less
abrupt
largely
avoid the
previously
dependence, compared
problems
are now
stiffening to
smoothed
some extent as a
some care
concerning the
reinforcement
a
that for
tension
Eots taken
10^. Based
on
the
assumption
we
ots
ranging
from
Dinges' Suggestion
of
up to about
20ecr, which is
in the
vicinity
use
regarded
simplieity,
for the
we
also
the
unloading behaviour
stiffening
zone as
softening
zone
discussed earlier.
for
an
in
figure 4.18.
reinforcement is
the onset of
figure as
well
as
yielding.
This
figure
is
only
correct
as
if the
reinforcing
orthogonal
to
stiffening
implemented
86
CHAPTER FOUR
Composite
modulus
Yielding reinforcement
Elastic reinforcement
Unreinforced concrete
Tension
stiffening
Strain
exacy
at
the onset of
yielding.
resulting modulus at
become
the tension
stiffening phase
wl also
depend
mueh
on
ie
area
of reinforcement If the
even
resulting
negative. It may
the
elements have
supporting action,
the
stiffening
on
foUowing.
The
reinforcing
bars which
stresses not
only
of the steel
area
resulting
reinforcing
bar to be
crack
see
smeared rein
stress to
forcing
further
by adding tense
the
smeared
reinforcing bars,
yield
increase of the
Young's modulus
for the
reinforcing
qualitatively
in
figure 4.19.
across
<*uK
f.
MICRO MODEL
87
**-
b) Simplified modification of
steel
/.=
L.-BA-3.
(4.22)
which after
bud up
in
phase
should be added
parallel
to the
relation
of steel,
in
as seen
figure 4.19a.
For numerical
modelling,
die
figure 4.19b, may be adequate. Here, the effect of the tension stiffening,
an
in
figure
4.19a, is replaced by
may be
increased
Young's modulus
approximated as
*,
_/,+/
=
(4.23)
concrete
tensile
strength
stiffening
ratio of
taken
as
assuming
a
reinforcement
of
one
percent
an
original Young's
yield strength
500 MPa
J*
=-3s.
p
120 MPa
0.01
=fy=
y
500
=
0.0025
E.
200000
88
CHAPTER FOUR
500 + 100
g
SM
240000 MPa
0.0025
From
expression (4.22)
the cracks
it is
seen
larger
for smaller
physicay
means
feasible in
a
so
far
as
by
of
smaller reinforcement
concrete
At smaller bar
simplification
described here.
In
open up and
close
repeatedly,
and after
few eycles
essentiaUy no tense
effect
as an no
through
stiffening
given
longer
Since the
Young's
is rather
problem-related question:
experiment
an
monotonic
probably be preferably
advanced
degree.
to better
However, for
model ie
earthquake
with
Simulation it should
concrete. An
unchanged
eycles
damaged
Option
lastiy
turn
briefly
to
opened
being researched
it is
some extent
direction. Although it is assumed that the dowel action also affects the modulus stiffness
terms
^33
in this model. In
some
examples
corresponding to
between 2% and 5% G,
simplieity the
micro model in this report is added into the local system, and
together
contribution rotated
to
the
global system.
to
Hereby,
the
the
micro model is
subjected
to a
89
CHAPTER FIVE
NUMERICAL
EXAMPLES
5.1
Implementation developed
in
chapters
a
were
implemented
in the
Abaqus
for the
[Abaq91],
which is
are
general purpose
reasons
discussed
[Wenk90].
The
macro
model of
an
chapter
three
was
coded
by
means
Option.
own
With tiiis
Option
arbitrary
a
by
the user's
to the other
parts of the
hand side (force) vector for each increment. Each of the nonlinear
user
Springs was
coded
wall section
as
shown in
figure 3.3,
was
the also
element
developed
in section 3.7
element
was
coded
by
means
Option.
This
Option
is similar to the
user
element
Option, except
behaviour
library
various
plane
stress
elements with
arbitrary
user
integration points.
subroutine
supplies
user
right hand
vector.
element and
material
options
State
variables
used to stre
quantities
formulation in
subroutine. For
subsequent increments.
are
updated
in the
user
description of the
State
variables,
[WLB93].
For
dynamic problems
the structural
mass was
a
modelled
by point
mass
elements
at
mass
acting only
in
preselected direction,
An introduction to the
use
of the models
and four is
given
macro
Appendix A,
Appendix B.
90
CHAPTER FIVE
5.2
Analysis
Procedure
5.2.1 General
separate
static monotonic,
eigenvalue,
dynamic analysis.
loads followed
a
may be executed in
sequence, such
e.g.
static
by
preceding step
The
were
gravity loads
user
applied
floor levels
during
static monotonic
this step,
nonlinear
analysis is carried
by applying
lateral action
This nonlinear
or a
in
a
one
step,
dynamic analysis,
ground motion
as
5.2.2 Time
integration
procedures
foUow Standard methods described in
The Solution
[Abaq89].
of nonlinear
which
n.
integration
the
Hber-Hughes-Taylor
an
ct-scheme is used,
see
e.g.
extension ofthe
[Hilb76]. Only
brief introduction
on
to
the
integration scheme
family of equations,
given
here. The
integration
scheme is based
as
the Newmark
may be written
H&mti+Cmtl + KUltl
U^=Un + 6tH +
FlM
(5.1)
^[(l-2)n
2,.+1]
(5.2)
t/.+1=l/
Af[(l-y)lU+1]
<5-3>
NUMERICAL EXAMPLES
91
global
mass-,
damping-,
respectively, displacement
and
, ,
vectors,
respectively. Lastly,
F represents the
generaUy varying
with time.
on a
are
defined
as
n+1
=UH
AtH+^-{l- 2)_
(5.4)
.+1=n+(l-y)to.
With known values for
(5.5)
displacement
and
velocity
at
time
started
as
M0=F-C0-KU0
or
(5.6)
,.+i is determined as
(5.7)
(M+yAfC+Mr2*)^ =F.+1-CUH+l-Kn+l
where the factors
and
=
y are chosen 0
by
the user,
see
further e.g.
[Hugh87].
From
(5.7)
of
equations
out.
provided
corrected
that C and M
are
impcit integration
is carried
The
displacements
and velocities
obtained as
UK+1
n+1+At2H+l
+
(5.8)
Vx= 6^
7*0^
some
(5.9)
aecuracy
by
and
foows
aKU.
F(tH+tt)
(5.10)
where
(tH+a)
tH+l
following choices
and y
=
are
recommended:
ae[-^,0], (l-a)2/A
(l-2a)/2.
92
CHAPTER FIVE
oc
such
as
-0.05
were
found
to
be favourable
at
according
to
among
others
[Abaq89], giving
problems
very
generally nonlinear,
are
the above
equations
are
solved for
automatically changing
chosen for a
lengths
are
available in
[Abaq91],
the former
being
problems
For nonlinear
an
problems,
the internal element forces, which result upon the Solution of with the
increment
are
apped
external force.
Thereby,
Only
problem is given
here.
thereby generaUy require very smaU increments in order not to deteriorate the
until
the
nonlinearity oceurs.
Solution
reach
few
the
most
arbitrary
and
large Systems. They may be divided into fuU and modified Newton-Raphson procedures.
Whereas the former
generally update
by recalculating
on
the element
the unbalanced
problems computed
tolerance values
Newton-Raphson procedure
one
was
Typical
percent of maximum
die
preselected tolerances
the
are
assembling
the
beginning of
the obtained
predicted
element forces
(stresses)
are
computed, using
as
computed
the
performed, updating
predicted values,
until the
NUMERICAL EXAMPLES
93
5.2.4
Damping
for the
For
Since the
was
very
decided to
viscous
additional viscous
that
damping may be better controUed physicaUy than the algorithmic damping. damping
is
This additional
provided
for
the introduction of
Rayleigh damping
matrix C,
consisting of one mass proportional part and one stiffness proportional part.
(5.11)
a + a2M
mass
The
were
manner
by
selecting
lowest
desired
damping
ratio at two
frequencies,
e.g.
frequency
is chosen well below the elastic first mode, to account for the due to
yielding during
the time
an
damping
will be
examples with
die
chapter,
damping was
frequency
0.12.
giving ax
0.0009 and a2
as
The Software
motion to be
apped
in three
perpendicular
directions into
simultaneously.
examples studied
motion in used
were
simplified
Time
planar analysis,
horizontal
ground
one
direction only
was
apped.
were
as
input.
the
design
ground
Friu
Tolmezzo.
are
Both
accelerations
are
discussed further in
94
CHAPTER FIVE
Examples
of
numerical model is
by comparison
shear wall
with
experimental
carry out
some
the models
developed
chapters
and
experimental
is
specimen.
description of this
models in
specimen
presented below
in section 5.3.1.
In addition to the
reliabity
check it is desirable to
use
the
developed
Performance check on
check is
the behaviour of a
buding is presented in
section 5.3.2.
5.3.1 Test
specimen
structural waU
large
A
number of tests
on
out.
carried out
realistic Simulation of
earthquake action
to
desirable
to use a test
as
specimen
as
close
fuU scale
as
possible
effects such
reinforcing and
the
A
aggregate
properly,
etc.
dynamic
with full-scale
or
specimens, documented in
an
the terature,
to
length)
as
regarded
squat walls).
specimen
was
assumed to be part of
taU wall,
representing
weU
as a
large
considerable
Bertero, and Popov [ VBP79] included an attempt to simulate the overturning moment as
action
on
the
specimen.
provided extensive
test data
which lends
itself to
models
NUMERICAL EXAMPLES
95
76mm-
2.388m-
1.680m-
-1016m
1016m
0.30Smt
|
2 AT 76
mr
T^ -r.
ml
.-.
...
.-..-,
ui
i
_-.-|ii
m.t
nr
0914 IS mm
m
l 0.152
I 1SLA8
76 rrfrt
mm
0.102
WALL
_i
4.:S2m 0.914 m
.
76mmSLA3
WALL 0.O2m THICK
r-,i
76
mm
SLAB
T
76mm l.2!9m
0.254
SO COLUMN
S-r1
76mm
G.H. 7 HOOPS
Llj
4>
mm
0.356m
95 mm
ZTmTSrl
Tmim.m.m.m.rmm.m.
C8m
1.1.
i. it
it-<<-
V
F00TW3
jl
I! :i
ii
ii 'i
ii
ii
"
jLjJ
ii ii M aUu.JL.
'rr
--
'!
|l
ll
11
F00T1NG
LJ."
at._
dti
1_ .JL
-0.660m
95 mm
3.098
a) Elevation
0.254 u*.
m
LstfS
A
Gft6EMD.7ATI^
L
=
2AT76rm)
nsz
2.388m
comprised
the
wall
specimens
in the
previous studies
on
in
[VB87],
and in
[Lind89]. We
use
96
CHAPTER FIVE
18 601
~T
6.1m
9 AT 2 745m
203mm
"FLAT SLA3~j
o
28 36m
24.7m
6IOmmx6lOmmCOL
o
\
7 AT 6.1m 42 7m
a
jo'ii'g
3 66m
54.9m
^508mm*5O6mm
COL
S5SSSSB
SPANDREL
BEAMS
&J Elevation
-SHEAR WALL-
0.644
V'l
J0.644V
J434
kN
434kNl
6.1m
J
6.1m
61m
6.401m
a)Plan
on
model
with model
developed
specimen are
The section
as a
shown in
figure
5.1. The
ten-storey
building,
in
figure 5.2.
were
shear
as a
bending
moment
(applied
vertical force
couple,
to
applied
at
opposite
sign,
and
coupled
shown in
figure 5.2).
moment
although usuaUy
history analysis
not
better than
entirely,
this is also
assumes
that
during a time
in Table 5.1.
the
case.
properties
ofthe test
specimen
are
given
NUMERICAL EXAMPLES
97
34.8 MPa
27 900 MPa 3.48 MPa
0.0001
444 MPa
Young's modulus,
all bars
test specimen
[VBP79]
5.3.2
The
use
reability
tests are
satisfactory,
is to
zone
according
and slender
gravity
carried
by
the
buding
was
are
shown in
figures 5.3
to
and 5.4.
eight-storey building
we
designed according
a
the
[PBM90], and
numerical tests
in
here
give only
brief
presentation
More
[BWL92b].
The lateral
static
an
elastic
zone
according
to
performed according
global displacement
ductility
Q equal
Appendix F.
=
The two
3, and \La.
5,
restricted
to
period of Vibration
used
found
be 1.38 Hz
according to the simplified code formula [SIA160] which does not consider
involving
only
resulted in 0.74 Hz
(elastic),
98
CHAPTER FIVE
444A'-
i i i
v-^
VO
4444
i i
8
VO
.,.
I
i
I
i
444
VO
4444 >-
8
VO
_4+-
6.00
| 4
6.00
6.00
00
Figure 5.3 Plan of capacity designed eight-storey wall building [BWL92b]
+Jr-
IffiP
gfSfit
r...
%m*m
rr\"<rxir'"i"
>Y%
>'
iM
mm*
IJi
v////
yrr/7 7777777
>j
6.00
6.00
6.00
NUMERICAL EXAMPLES
99
With the
in
was
obtained
as
[BWL92b] which
With
a
study
as a
realistic
design
two extremes.
storey
of 3.08 MN
as
(2.78
=
MN for
were
obtained
ME
16.1 MN
bending
and
a
moment at
=
the base
VE
ductility design,
obtained
ME
VE
resistance factor jr
=
equal
as
Mt
16.1(1.2)
19.3
9.7(1.2)
ductility design.
The material
speeifications
were set
according
[SIA162]
as
(design strength fc
19.5 MPa), is
Reinforcing
steel: S500
(design strength fy
buding
analysed
by nonlinear time history analysis, the most realistic nonlinear behaviour is desirable (no
reduetion in the
the time
aecounted for in
mean
strengths
to their
effective
values
as
follows:
fy
in
550 MPa,/c
30 MPa.
mm
The wall
cross
thick
was
by
6.0
buckling
criterion
according [PP92]
were set as
ductility:
0.26% distributed
uniformly
over
the
cross
6D20
D10/200
Sym
''4
r-i
500
1^/2
3000
Figure
5.5 Wall
cross
6D12
D10/250
Figure 5.6 Wall cross section fromfirst storey of fll ductility design, \i\,
100
CHAPTER FIVE
confined
zone
of 500
mm
length with
the
values
a
were
respectively. The
and the
ratio of 0.20 %
remaining
=
at
foows:
MR
19.4 MNm
and
bending
=
and
VR
ductity
of 1.2
design,
fu
we
MR
15.8 MNm
X0
flexural
overstrengths
=
as
Restricted
ductity:
Oow
1.2-^
19 4
1.45
16.1
Fu
ductity:
<&.
1.2^
15.8
1.95
9.7
These values
are to
using
dynamic
magnification factor of
)
=1.3 + = 1.3+ 30 30
1.57
where
is the number of storeys, and inserted in the formula for shear demand [PBM90]
co
<E>
V*.
gives shear
demands of
=
1.41(1.57)0.72
ductility,
and
1.95(1.57)0.43
to
1.32 MN for fu
Vc
vcbwd
0.91(0.3)0.8(6.0)
1.31 MN
as
0-91 MPa
the
remaining
spaced
at
250 mm,
NUMERICAL EXAMPLES
101
V,
aj./A^^-)m^^>0.25
139 MN
which
together
with the
concrete
contribution results in
shear
capacity
of 2.7 MN,
exceeding
ductity
an
assumed material
overstrength
curtailment
effective shear
capacity
the
as we
later.
Reinforcing
over
height of the wa
introduction
to
the
Due to the
ground motion
analysis
perpendicular to the buding in the direction of the was, only half the buding needs to
be
one
mass
used in the
for the
5.4.1
Comparison
with
experimental results
performed. These
an exact
carried
out as a
comparison
of the
analysis results
obtained from the numerical model with the results obtained from experimental tests. It is
clear that
one
agreement
cannot
be obtained with
developed
tests, two
major principles
weU
reasonably
tests. The
structural wa,
simple
numerical
modeing
of this
example.
This
example
wl be foowed
clarity
time, and
102
CHAPTER FIVE
specimens
two
was
subjected
to a
appUcation of
increased tl
and
gravity
loads
a
by
specimen,
shear force V
a
apped
at
was
monotonically
faure occurred at
was was
shear force
sghtiy
unloading
reloading couple
shown in
performed.
also
apped
and increased
according
of 0.644 V,
as
figure
are
suited to
properties
As for all finite element models the chosen element mesh will affect the obtained
macro
model,
we test
by discretising
the
three-storey test
Figure
with its load pattern and the three meshes for the
Macro elements formulated
a as
macro
model
are
developed
in
different number of
figure.
0.434 MN
0.434 MN
0.644 V
0.644 V
A'<%tYifYt'X<
0.914
m
MeshA
K&ii
%l
&Y&:?>>< 0.914 m
*. Jri
..mmmtiim
-v>7%
(Ar
'
MeshB
mumAwhmii
<.r*Y*y&s&
1.219
,5 m
^^m**-
MeshC
Lw= 2.388
Test
specimen
Numerical discretisation
NUMERICAL EXAMPLES
103
With wall geometry and material data from section 5.3.1, the the
macro
input
set
was
prepared
for
basic
of input values
kept
for all
0.193 m4
Ec
27900 MPa
spring, cccr
0.5
Ratio of yielded to uncracked stiffness for vertical spring, oc? Yield moment My, (selection shown in Shear force
at
Appendix C)
expr.
shear
to
cracking, Vc,
acc. to
(3.49)
Ratio of cracked
0.18
test
examples
the fourth
Figure
floor
at
versus
foundation
horizontal
footing
of the test
additional 10 to 15 %
in mind when
displacement included
comparing with
In the
mesh is
changed
kept constant
and
foUowing numerical examples some of the parameters of the basic set are varied, only when specifically mentioned,
1.0 MN,
are
however
are
original value.
For the
investigation of the mesh effect a monotonic shear force is apped at the fourth
to
floor, increased
according
to
to the
including
the
coupled
appUcation
of force
performed in
giving a total
shown in
of 50 increments.
displacement
figure 5.9
slip
modeing.
It is
shows
already
at
flexural
cracking
oceurs,
again visible
upon
smaer stiffnesses
(cracking and yielding) occur over a larger part of the structure than in
comparisons
not
exceed
shear
force of 1.0 MN
or a
mm.
104
CHAPTER FIVE
BASE SHEAR
V(KN)
DISPLACEMENT Sjtmm)
Figure 5.8
displacement
versus
Mesh A
Mesh B
Mesh C
0.01
0.02
0.03
0.04
displacement [m]
Figure 5.9
Mesh
NUMERICAL EXAMPLES
105
ct..
0.5
0.8
cx..
0.01
0.02
0.03
0.04
displacement [m]
subsequent
are
studied, mesh A is
to
check
obtained
by studying
the
input properties.
as
well
as
the
yielded stiffness
Springs appear.
well
as
mentioned in
chapter three as
when
in
Appendix
should be
the
global behaviour,
they
are
varied,
The first parameter tested is the ratio of cracked to uncracked stiffness of the outer
vertical
Springs,
in
Gfcr. This ratio represents the relation between the two stiffnesses
Kcr and
Ke,
of
seen
figure
and
3.19.
Varying
the
over
0.5 to 0.8,
by using
the mesh A
figure 5.2,
It is
applying
same
monotonic force
as
shown in
figure
5.10.
seen
from
figure 5.10
test
0.2 appears
to
correspondence to the
(figure 5.8)
until
one must
keep
in
106
CHAPTER FIVE
0.01
0.02
0.03
0.04
displacement [m]
Figure slip
5.11
uncracked stiffness
some
cyclic testing
making
(as may be
seen
in
figure 5.8)
some
thus
a
together
with
a
somewhat
quality
a
of the test
that
(giving
global flexural
according
to
figure 3.22)
will still be
Optimum,
if the
to
Ky
Ke
figure
0.5
constant at
It may be dent
on
seen
from both
figures 5.10
the chosen
on
yield stiffness
in the
are
(figure 5.11).
Both the
seen
figures
ductility
generally
dependent
ductility
flexural
on
Although
a
the obtained
displacement
only
varies with
cracking from
0.2, it is also
seen
asma difference in
NUMERICAL EXAMPLES
107
o.
0.14 0.18
0.25
_
a.
a.
0.01
0.02
0.03
0.04
displacement [m]
Figure
5.12
as
ranging
from 0.005
to
0.02,
we
also obtain
yield stiffness, it appears that the yield stiffness will have the most
in the first
as
place interested
may obtain
the
time
history
at
analysis. This quantity will be used for the proper design of the wall cross section
location in order
to ensure
that
that it
a
can
take this amount of curvature without faure. In the wa structure, the curvature
with
ductity
yield stiffness
also
will be shown.
In the
same manner as
ratio
spring
shear stiffness to the uncracked shear stiffness, which varied from the recommended minimum value 0.14
over
shown in
The
figure
5.12.
to
next
parameter
as
suggested
height
of the
macro
108
CHAPTER FIVE
ii
0.6
o o -C in
in
in
0.5
0.4
_
///
V)
0.3 ~r 0.2
0.1
0
0.01
0.02
0.03
0.04
displocement [m]
centre
of relative
rotation
expressed by
as
i.e. the value to be the distance of the centre of the relative rotation to the bottom of the
element related to the element
give
account, which
usuay
zero.
starts
from the bottom of the element, this value would be lower, here show the influence of this ratio
result is shown in flexural
perhaps close to
We wl
figure
of relative rotation is
moment
placed low,
in
the
yielding
is
normally the
figure 5.13.
cyclic loading.
5.4.1.2
Cyclic behaviour
For
similar
test
specimen
data for
to
the
one
reliability
tests
behaviour,
some test
cyclic
tests is
we use to
reabity
cycc loading.
NUMERICAL EXAMPLES
109
'
_
OlSPLttEMOfl'
Figure 5.14
test
specimen [VBP79]
Figure
5.15
Fourth
floor
horizontal
displacement
versus
base shear,
experiment
[VBP79]
The
test
specimen
was
subjected
as
to
couple
shown
previously
set was
then
reversed in
displacement history is
figure
5.14.
110
CHAPTER FIVE
1.00 0.75
0.50
i~~7i
1
^
^yj
/
~
Y/^mr^T^^ / /
0.25
o
:7^
A^^
/
0
O w o
0.25
0.50 0.75
1 00
i
i
-0.04
-0.03
-0.02
-0.01
0.01
0.02
0.03
0.04
displacement [m]
Figure 5.16
cyclic
test,
analysis
2.0
1.5
1.0 0.5
0
~T~
/\
Y
Y
/
/
\
1
0)
-0.5
/
//
^^^
-1.0
-1.5
mr^*^
V
1 1 1
-2.0 -0.01
0.01
0.02
0.03
0.04
Elongation [m]
versus
displacement
versus
experimental
for
test
is
figure
5.15. For the numerical model the response but with the number of
shown in
figure 5.16,
of the
eycles reduced
of around 20
to two
mm
with
displacements
and 40
Since
repeated eycles
same
maximum
NUMERICAL EXAMPLES
111
1\
/ /
/
/
/
[MN]
p cn o
Force
p cn
/ mS^
^^
//
0.01
0.02
0.03
0.04
Elongation [m]
Figure 5.18 Spring force versus elongation for right vertical spring offlrst storey macro
element, cyclic
1.00
test.
... .
^
a o
c5
Lu
o
UY / IY
^/
Yl
/
-0.25
-0.50
"
-0.75
-1.00
/Y^
i
YY
i
-0.002
-0.001
0.001
0.002
Shear deformotion
[m]
spring offlrst storey
macro
versus
elongation for
horizontal
ness or
strength degradation,
eycles
are
suitable for
com
Springs
work
during cycc
beha
versus
spring elongation is
shown in
cyclic
112
CHAPTER FIVE
2.0
1.5
1.0
-
^
Q> o
0.5
0
-0.5
-~
-1.0
-1.5 -2.0
iii.
-0.01
0.01
0.02
0.03
0.04
Elongation [m]
versus
element, cyclic
test.
In
figure 5.16
change of stiffness
(e.g.
seen
may be noted
on
the
reloading branch
at a
force
in the upper
spring
seen
in
figure 5.17)
to
compressive
place
yielding in
this side
prior
yielding in compression.
We wl also need examine the behaviour of the horizontal
to
test.
This
spring,
employs cyclic
bilinear three.
origin
oriented
was
discussed in
Spring
macro
force
versus
test
storey
element in
figure
the
In order to
complete
versus
elongation
spring
this
macro
element is shown in
The small
figure 5.20.
in the
figure
was
spring
to
only
in
compression.
attributed
numerical
causes.
The
the
global cyclic
behaviour
as
as
shown in
figure 5.16 is
characterised
by
the
shape of dynamic
hysteresis loops,
not
defined in
figure 3.18.
shape of these
loops does
deviate
too
experimentally
NUMERICAL EXAMPLES
113
1.00
0.75
/"~7f
-
0.50
0.25
o CD
mY
Lr^^^^^
/
^T^^ /
/^-^
Y
JZ
in
o in
/
/
^^^^
CD
-
-0.04
-0.03
-0.02
-0.01
0.01
0.02
0.03
0.04
displacement [m]
test, ctd
=
Figure 521
0.8
2.0
1.5 1.0
r-~
~~~7l
/
\
1
1
I s
[MN]
p cn o p In
/ /
Force
*r
^^^'^^^
-1.0 -1.5
-2.0
tC^^^
I
1
1
-0.01
0.01
0.02
0.03
0.04
Elongation [m]
Figure 5.22 Spring force versus elongation for left vertical spring offirst storey macro
element, cyclic test, ctd
=
0.8
area
within these
loops provides
to
as
a measure
of the
question,
referred
the
hysteretic damping.
the "fatness" and
an
important
the
characteristics of the
hysteretic shape
are
"pinching" of
on
loops.
extensive discussion
this
topic.
114
CHAPTER FIVE
1.00
0.75
0.50 0.25
o
w
0 -0.25
%
o
co
-0.04 -0.03
-0.02
-0.01
0.01
0.02
0.03
0.04
displacement [m]
test, ctd
=
Figure 523 Fourth floor horizontal displacement versus base shear, cyclic
12
2.0 1.5
1.0
-
"
11
s\
/
/
~~
/1
/ l
\
\
[MN]
p cn o p cn
Force
/ //
-
^y^
-1.0
-1.5
I //
mr^^
Z''^
J 1......
1
-2.0
-0.01
0.01
0.02
0.03
0.04
Elongation [m]
versus
macro
12
In the model
hysteretic shape
mainly determined by
chapter three
the
chosen
as a
at
a
closing on
was
com
pressive
meaning
to
branch
during unloading.
as
1.0,
Li order
study
NUMERICAL EXAMPLES
115
equal to 0.8
zontal
are
shown in
figures
Figure
to
displacement
versus
figure
spring
behaviour.
5.24.
It is
seen
Corresponding results
from
for ac/
equal
1.2
are
shown in
figures
5.23 and
figure 5.21
that
a more
pinched shape
of the
The
loops
origin.
upper
Less
pinching
is obtained in
case
cycle,
seen
in the
right corner).
In this
crack closure
No
particular attempt
would not
is be made to
adjust good
the
input properties
so as to
obtain
a set
an
Optimum agreement
with the
experimental
dever
example.
Such
of
properties
necessarily
comparisons. Rather,
the
general effect of some important properties were shown, with experimental results
the lower stories of a structural wa
It
was seen
as a
from
basis.
to
yield stiffness
uncracked stiffness is
important
for the
nonlinear behaviour,
which however is obtained
tance
weU
as to some
only important
in the first
eycles. proved
mainly empiricaUy.
a
be of lesser
impor
for
waU of this
configuration.
flexural cracks close
For the
largely
determined the
shape
of the
Reasonable agreement could be obtained between both static monotonic and static
set
experimental
data and
analysis
for
default
of
be somewhat lower
for the
MN
compared
model
to
0.57 MN for
analysis),
which
of this model.
5.4.2
Multi-storey wa
of the struc
The
buding subjected to
seismic action.
was
designed
buding,
and
116
CHAPTER FIVE
the
sght
moment
resisting
effect of the
gravity
neglected
in
the
analysis.
We wl
perform analyses
ductity.
for both
'
4.0
4.0
i;<"i:i::f
4.0
m
r
H
32.0
m
Elastic region
4.0
m
r
4.0
t.inl(i.,il.i,Oi.fl
Plastic
hinge region
4.0
Uli""
"
2.0 2.0
2.0
m
m m
2.0
m
1
,t
Figure 5.25
Numerical model
One mesh
only
design.
The mesh
corresponds to
the
one
used
by
eigenfrequencies
for the
were
elastic
(sghy cracked)
=
wa without
=
stiffening beams
=
found to be fi
0.67 Hz, {2
9.8 Hz, f4
16.7 Hz.
Eigenfrequencies
for
damage
are
Appendix
figure 5.25,
as
seen
that the
the
height Lp taken
Lw
elements.
The
to the
ground motion in form of an artificiaUy generated acceleration history, compatible design spectrum [SIA160]
figure 5.26.
In
SIA
for seismic
zone
shown in
ground
acceleration of 10 seconds
length.
phase
to
lasts about
seven
seconds. In
figure
ground
damping.
NUMERICAL EXAMPLES
117
The
input properties
for the
macro
model,
as
described in
Appendix A,
are
1.8 m2
5.4 m
34000 MPa
0.7 0.01
to
3)
7.9 MNm
5)
4.6 MNm
3.0 MN
shear
cracking, Vc,
acc. to
expr.
(3.49)
0.16
no
experimental
foUowing
results
one must
were set at
design phase.
subsequent
designed buding
buding.
This is
a
so
that at first
in
performed
the
chapter. Then,
in
chapter six,
are
number of
more
fundamental
on
questions regarding
the
capacity design
of was
results of the
examples tested in
this
chapter.
We
direcy
time
by
direct time
integration,
acceleration
as
given by
the 10
12
second
ground
history
figure
5.26.
integration
residual force
tolerance of 1/100 of the maximum element forces. A maximum of six iterations per increment
were
equibrium
found to be
allowed, whereby
of 2%
two to
three
were
normay
suffieient
was
prescribed.
be
history analysis
to
displayed
is the
displacements of the
For
roof level
(ninth floor),
a
design ductity
figure 5.27.
comparison
purposes,
plotted.
major difference with respeet to the elastic linear Solution hes in the yielding which
for both
capacity designed
seven
seconds.
118
CHAPTER FIVE
2.00
-iiiiir
-iiiir-
in
1.00
c o
-1.00
<
-2.00
4 Time
10
[s]
Spectrum
SIA 160
of
generated ground
motion
~~
Design spectrum
-i
**
1.0
f
10.0
100.0
[Hz]
stiff ground and five percent
to
damping
Figure 526 Artificially generated ground motion
It may be in the
seen
figure
ductity
levels is
not
a
may be
softer system, for which the first mode is excited less after
plastic
global displacement
ductiUties
set
at the
NUMERICAL EXAMPLES
119
0.15
-j
ri
"
-1
rence
Solution
_
rf
g
E
u
0.10
Design Design
p.
= =
3
5
A\
p_
0.05
j
A
r
j\\
i
A
i \/
'
s
Q.
w
Vv if^ \\ J
\ V
i/\v/ \
Xj
ji
f
/
/P
h
\\
\\JrJ
\
v'
-0.05
i 1i
ArA
\y/
V.
Vi
-0.10 -0.15 0
1 1 1
...
...
6
Time
10
12
[s]
roof displacements
Figure 5.27
Horizontal
0.15 0.10
c
>'r
"
--
T
A
=
0.005
...
0.01 0.02
s s
AT
0.05
J\,\ A
V
/ \i
ffl
ftt
-
Q.
Vi
-0.05
r\\
-0.10h
-0.15
0
i
i i
'
6 Time
10
12
[s]
horizontal
increment
length,
ly
from
figure 5.27,
be close
at
the
onset
of
yielding at
the
base,
was
found
to
to zero, or even
negative due
to
ductility design is
shown in
120
CHAPTER FIVE
E
z
c o
E
o
Time
[s]
Figure 5.29 Overturning moment at wall base versus time, restricted ductility design
E
z
E
o
Figure 530 Overturning moment at wall base versus time, fll ductility design
and for 0.02 It is not
s.
possible
to
distinguish
about
displayed through
moment at
the
over
overturning
the base,
NUMERICAL EXAMPLES
121
12.0
-i
r..,,...,
Right
Left
c
curvature
urvature
-
10.0
E
Q
8.0 /
z*
b.O
Q
<D
l_
4.0
O
>
i_
2.0
r'
l l l
/,
iiiji1_
ii
...
10
12
Time
[s]
12.0
i|iiir
Rinht
mirvntii
ro
11
o
Left curvature
10.0
E
a>
8.0
f'
6.0
Q
ffl
u
4.0
o
>
.1
l
2-
10
12
Time
[s]
versus
time
versus
time in
in
as
the fll
ductity design.
ykME (where
equivalent
we
ME
is the moment
at
figures.
displacement known,
122
CHAPTER FIVE
12.0
i'ir~
10.0
ay ay
0.01
0.02
E
<o
8.0
6.0
o
o
l_
4.0
>
2.0
6
Time
10
12
[s]
12.0
-o c o
,i.i,i
et
_
.1,111,
1"
"
"1
""
T"
'
0 01 0.02
-
10.0
ay
E
<B
8.0
(-t
6.0
Q
<D
1_
4.0
"5
>
i_
z.
2.0
llllll
i.
..
i...
6 Time
10
12
[s]
curvature
ductility
2 MN
one
to
(displacement) times
as
(gravity
conservatively estimated
around
NUMERICAL EXAMPLES
123
as a
are
plastic hinge,
and 5.32
The
separately
right curvature
demand
versus
The any
ductity
is defined
as
based
on
given
to
were
obtained
equal
vertical
Springs.
0.02
some
influence
influence is
in
figure
clear
as
right
are
compared. It is
seen
from
reducing effect
In order to get
bending
moment
distribute
for each
over
plots
are
displayed
plot on
ele
curvature
plots which
(starting
with the
page):
of the
elastic
In
figure 5.34,
from
corresponding quantities
the
are
ductity design.
degree,
some
It
is
seen
figures 5.33 and 5.34 that the curvature clearly decreases towards
bending
State at
moment
the
decreases
to a
lesser
seen
yielded
that
yielding
place
at
fll
ductity design.
impcations of this
treated
more
Since
curvature
relationship
between chosen
and
expected
a
ductility
in the
plastic hinge
is
static
analytical example
noted that the
it would be
possible to
compare this
history analysis
the curvature
over
ductility for the lowermost element of the plastic hinge height. Due
average value
to
die entire
the
value
suggested
as an
assuming
chapter
uniformly
six.
ductility.
This
matter
S'
Moment
Moment
I
M
-
I
l
MM-- Ol O Ol O
M
Moment
[MNm]
[MNm]
[MNm]
I I
Ol
o
Ol
tili, - I MM l I
Ol cn Ol --MM Ol O O Ol
--MM
Ol
Ol
Ul
oi
p
p
O
oi
p
Ol
in
--MM
Ol
Ol
lb b
b
I
b b
b b
lob
Ol
b
b
5 ?
o
I o
>M
3 3
3 3
^^
1
Moment
Moment
[MNm]
[MNm]
Moment
l l
[MNm]
Ol --MM o Ol o Ol o
s 2
o o
I
MM-- Ol O Ol
~
I
l
I
I
--MM O Ol O Ol MM-- Ol O Ol O
,
,
M Ol
Ol
I I
l
Ol
I I
Ol
--MM O Ol O Ol
1 o
o
o
lob b
o
in
T
T
1-
1 1
i
lob' p 1 I cn
5
tn
C/i
1 o
1
-
I
o
o gm
3 3
o
ET
c
ro
o
r^
3
tt
!t*
-
^
3 3
<*
Jt *o
M
3
^
o
i_Up
a
JH Ci'
a-
.
>t <*
oo
o o
Sl *9
-
\
(*
N
o a
1 1
en
1
o
O
Ol
Oq 2.
CVI
126
CHAPTER FNE
spring
the
spring
force
versus
the
spring elongation
of the
plastic hinge
we
as
of the elastic
region,
in
and 5.34.
Figure
ductity design,
more
and
figure
ductility design.
figures
it is
clearly
seen
than in
figures 5.33
elastic.
intended to remain
ductity design,
unloading branch,
was
taken
as
closure force level somewhat below and above the effective normal force
as
seen
that the
ductity design
far the
ductity design.
PhysicaUy,
in this may be
thought of as reasonable in
more
so
as
compressive yielding
due
to a
unloading should be
weU due to the
to a
somewhat
pronouneed
for the fu
ductity design
area over
which the compressive forces may be transferred that has occurred in tension before of
as
larger plastification
more
unloading.
This leads
curvature most
somewhat
"pinched" shape
be 5.34.
moment vs.
clearly
e.g. in the
element in the
displayed
versus
time in
figure 5.37
In the
for the
the
ductity design
at
and in
figure
ductity design.
=
plots,
Vw,
Vw
is the
at
overstrength factor, t,
VE
equivalent
as a
the
plots.
seen
It is
that the
capacity is mueh larger than the equivalent static force. This is mainly
length (6.0
m
due to the
large give
wa
horizontaUy)
requirement
contribution
this
together
a
with the
concrete
around 2.7 MN
capacity,
see
wa with less
length
It
capacity
seen
would be smaUer,
although
the minimum
may be
ductity designs,
NUMERICAL EXAMPLES
127
2.0
1l
I
O
-2.0
S.
-4.0
k /
'/
-6.0
-8.0
-0.002
0.002
0.004
0.006
Elongotion [m]
region
1
-2.0
ff
m-
-4.0
//
-6.0
-8.0
-0.002
0.002
0.004
0.006
Elongotion [m]
-0.002
0.002
0.004
0.006
hinge
128
CHAPTER FIVE
2.0
-2.0
i
-
-4.0
-6.0
-8.0
'
'
-0.002
0.002
0.004
0.006
Elongation [m]
region
1
k
-
//IJ/I
-2.0
-4.0
-6.0
-8.0 -0.002
0
0.002
0.004
0.006
Elongotion [m]
-2.0
l^^ ^s i
'//
"^
m^l l
-4.0
1
1
m-^^
-6.0
-8.0 -0.002
.....I
..
0.002
0.004
0.006
Elongotion [m]
Lowermost element
ofplastic hinge
NUMERICAL EXAMPLES
129
Figure 537
design
Figure 538
is reached which for fu
close to the
ductility design
come
capacity (2.3
MN for restricted
ductity
history results
to
so
far
are
on an
buding subjected
130
CHAPTER FIVE
4.00
Vi
2.00
c o
o o
2.00
<
-4.00
10 Time
15
[s]
recorded at Tolmezzo, Friuli, 1976
Figure 539
to a recorded
a
N-S component
of ground motion
ground motion. For this purpose we repeat some of the computations, using
at
earthquake. length,
with
ground
motion is shown in
seven
figure
seconds, starting
around 3 seconds.
ground
m/s2,
i.e.
higher than
as
the
ground
acceleration
are
compatible to
to
design spectrum.
The
walls
used
previously
was
subjected
realistic
the recorded
done
intentionally
so as to
get
Performance
designs,
when
subjected
to a
ground
motion with
significantly higher
acceleration levels.
As
first
plot,
figure 5.40.
an
A maximum
displacement
of around 0.13
=
drift of 0.13/32.0
compatible ground
Rather than
motion
input,
both
repeating
all the
previous plots,
we
some
critical
versus
quantities,
dramatically. The
base shear
time is shown in
figures
designs respectively.
NUMERICAL EXAMPLES
131
0.15 0.10
'
'
11
c
o
1*1
r.
0.05 h
l\
v.
1 1 /
E
tu
1 \ \i 1 11
\
/Ma1.
/\V
0
\
a.
1
*
U
\
IV
1
'
l'
M
y
Vi
3 5
i
\\ /
-0.05
o o Cc.
VT
\f
-0.10
--
Design Design
i
pk
fx.
. i
-0.15 0
6 Time
10
12
[s]
roof displacements
Figure 5.40
It is which
Horizontal
seen
designs,
an
means
depending on
how conservative
again
macro
model does
partly
the
not
desirable and
partly
difficult to obtain
realistically in
shear
capacity
here
should be estimated
modelling
Some
is obtained
by
the
region
capacity is reached.
thought
should rather be
given
to
the
question
capacity,
was
fu normal
force of 4.15 MN
(including
tributary
1.2,
a
to
the wall)
used, without
multiplying
it
by
equal
to
as
capacity
taken
as
design examples
If, however,
dividing by
strength
(j>
acting
on
the wa to 90
shear
strength of 0.86
when
MPa, and
failure is
like the
total shear
capacity
indicating
a
shear
the
ductility designs,
subjected to
ground motion
one
normal force
132
CHAPTER FIVE
Figure 5.41
design
Figure 5.42
The curvature
design
also
clearly larger
the
plastic hinge
is shown
designs
in
figures
respectively. For the restricted ductity design, initial yielding in right curvature occurred
during reloading
ductity.
More from
yielding
large
curvamre
pinched hysteretic
NUMERICAL EXAMPLES
133
12.0
>
Right
curvature
Demand Ductil ty
|
D
Left curvature ob
00 b
/
r-1
-
4.0
2.0
t"
i
0
.. i
10
12
Time
[s]
restricted ductility
Figure 5.43
Curvature
ductility,
design
12.0
-a c o
iii,..,
,111,1.1,1.' i
Right
-
curvature
E
o
8.0
u 3
6.0
I
.
.
o
a>
4.0
./-
*""
"5
>
i
2.0
i
-
,.i
...
...
10
12
Time
[s]
Figure 5.44
Curvature
ductity.
to
displayed
from the
eight-storey building
the
were
intended
give
basic
serve
over-view of the
as a
and to
background
the
tests
on
are
capacity design
in
method in
chapter
six.
Thereby,
to some
model
chapter is devoted
basic
tests
developed
chapter four.
134
CHAPTER FIVE
was
developed
will
use
in
some
reabity
macro
tests. For
the sake of
simplieity
we
the
specimen
as
for the
et
We wl
mainly
reliabity tests,
shment of meshes
we
as
are
both
quite considerable
wl
when this A
multi-storey
structural was
eight-storey building
the
reabixy tests.
5.5.1
Comparison
with
experimental
results
The
specimen
set
and force
application
was
are
the
same as
macro
model. A
basic
of model
properties
are not
made unless
specificaUy stated, in
table 5.4.
model. The
507 MPa
stiffening
terms
0.002
of elastic modulus
0.01
test
examples
Three meshes
are
regarded
as a
elements. Forces
at
discretised and
applied at as follows.
apped
(gravity
couple)
applied
at
corner
elements.
NUMERICAL EXAMPLES
135
V /*Y
-
^C
:;';
YP
Vi
:>,
'
s
'
,'
<-
>;:,
v-
'.
'
':
*':
<
,,
'>
^\
\-
^/
^
'%
:"
,"
'
.
:"
J-**:
1.
"
o-s >.r
A
^^^
'rYj:
v
v;* 'k*j
"
%? 'o!';
....
V
^
A
s.
.'.
ii
i '/r
-.;
<f'
"
A ?'
-,
~!
5
m
*_>
r-
-.
<
,
x*'m<3.
$
5.
-..
#
Y-'i
rs,/,M
'
*
"
-J^
*?
.V
..
/ \'A
/r
*: ?"
y-
i
Vf
^ i'i
*
,4-
^
/i
'
>'
Y''
4
?
v>
'<
'
/ '
$
\
v
fr *!
'"<
^'
-S
'.$.
;..
^
'
0^;
a) Mesh A
Ayy-
*%
tr-'
..
:&
:,-.
v<& -%
$
:#'
*
''
'*%
i>
Y.
'
%' i"
r-
* 9
i?
*f
--
H
-
b)MeshB
Figure 5.45 Meshesfor micro model
c) Mesh C
1.0 0.9
.-1
'
0.8
--
^0.7
2
1-1
/-
#y'*
0.6
-
Mesh A Mesh B
Mesh C
ja
to
0.5
0.4
0 10
0.3
-
0.2
0.1
0
j
1
0.01
0.02
0.03
0.04
displacement [m]
shear is shown in
from
B and C
as a
closer,
indicating convergence.
For the
B wi be used,
compromise
136
CHAPTER FIVE
ft
ft
ft
3.5 MPa
0.01
0.02
0.03
0.04
displacement [m]
strength
is shown in
B.
This
strength
that
the influence is
mainly visible in
the
pre-yielding region,
gives
the
somewhat
skeleton curve,
as
to
was
assumed
at
analysis,
assumed elastic.
analysed is
0.01
to
0.02. The
global
effect is
essentially
proportional to
ratio represents
as
may be
seen
from the
figure.
direcdy than
in the
macro
model the
the reinforcement
only.
Furthermore, the influence of the tension stiffening effect is shown in figure 5.49. The
strain
at
where
zero
tensile
at
stress
is reached
to
0.002
(approximately
steel
yielding)
to
several
integration points,
to
foowed
by
give
reasonable result
NUMERICAL EXAMPLES
137
ay
aY
0.010 0.020
0.01
0.02
0.03
0.04
displacement [m]
ratio
1.0
0.9
0.8
rr
--
mr
"z
-
'/^
~
'
--
e0
0.001
O
CO
,/
r y
'
fr'
e0- 0.002 e0
=
//
aj (0
0.003
0.3
0.2
0.1
0 L
_.
0.01
0.02
0.03
0.04
displacement [m]
quantity to be analysed is
expression by [Rots84]
experimental expression
see
work
by
Paulay
and Loeber,
see
equation (4.16).
with die
equation
138
CHAPTER FIVE
acc.
to to
to
acc.
acc.
0.01
0.02
0.03
0.04
displacement [m]
on
the
global
figure 5.50.
It is
seen
expressions
a
are
relatively
[Prui85] gives
stiffer
global
expression
a
give
larger
there is mueh
bending (i.e.
a
normal tensile
stress)
in the
too
large
specimen.
curve
one
Reducing this
12
mm
gave
for
[BG84] however.
simplest
of the
to
expressions,
the
by
[Rots84] appears
give
reasonable results
compared
experimental
data in this
example.
Some attempts with
more a constant
gave
mueh
too
equibrium
the tensile
shown in
iteration
problems
be
due the
the sudden
drop
passing
the
strength. Therefore,
the
as
ones
plot,
seem to
We will
now mm to
discussed
at
the end of
better
be carried
cracks. As discussed in
stress
tense
sections.
NUMERICAL EXAMPLES
139
It
was
shown in
expressions
possible
tests,
modulus may be
performed.
In the
previous reabity
using
example
with the
steel modulus
was
a
made.
we
on
the
global
behaviour of such
to
modification
here
perform
the
three-storey specimen,
chapter four.
given below in
Concrete tensile
2.8 MPa
0.4/,
1.2 MPa
211000 MPa
444MPa/507MPa
0.0021/0.0024
3.50 %/0.73%
0.73 %/0.73%
stiffening
not reach a stress
as
wl be
we
seen
chapter, does
perform
simpfied modification
of steel
modu,
based
on
follows: In the
boundary
steel). In the web both vertical and horizontal steel is correcdy modified based
vertical reinforcement ratio of the web, which is the
same as
the
follows:
Boundary element:
fts=^s- ^=
0.035
32 MPa
f.
AAA
32
m + u
=-------=
=226700 MPa
y
Web:
0.0021
/=-2s-=
J"
-^0.074
151 MPa
=-4lA=444 + 151
y
0.0024
274200 MPa
140
CHAPTER FIVE
0.01
0.02
0.03
0.04
displacement [m]
to
tension
stiffening
we
obtained
an
Young's
226700/211000 equal
274200/211000 equal
to
The effect
global
a
figure
and
5.51. It is
a
seen
that the
gives
slight increase in
cracking phase
result which is
not
little closer
to
the
experimental data.
however,
importance
studied
in relation
were
previously in
this
chapter.
presented
here
to
conclude the
monotonic
figures
5.52
applying
to
integration point
edge element
vs.
was
plots.
strain,
across
plane is displayed.
shear strain
displayed.
strain is
NUMERICAL EXAMPLES
141
4.0
r\
3.0
10 to
mr^
H
mt~
2.0
CO
L.
S
sz
i.o
C/l
11I.-.J.11J11
Normal strain
vs.
strain
cracks
o_
w
co
tn
0.0002 Strain
0.0004
Stroin
vs.
strain
vs.
strain
Figure 552 Internal stress quantities of micro model, left edge at base
vertical and horizontal bars,
A
respectively. cracking is
shown in
concrete
different
shear force levels up until 1.0 MN. In 5.53a the crack pattern from the prototype after
completion of test is
is shown
shown. In the
plots
using
shown
142
CHAPTER FIVE
HWlwlw)
-^-\
w
^
_
hKxkywK
\
-^\
w
>>
_-
->.\ w
S> V.N \
-~.
w \
-SWiW
\1\
Figure 553
phase.
the
Thereby,
be directed
reliability
to
tests are
direcy
the
incorporates
the
cycc
behaviour.
NUMERICAL EXAMPLES
143
5.5.2
Multi-storey
wa
eight-storey
is very
macro
model,
wi here be
a
repeated
as
numerical effort,
mited
comparison
results wl be
direcy compared
shown in
between the
macro
eight-storey waU is
distributed
to
68 elements.
Gravity
loads
were
lumped
into two
given in Appendix
E.
4.0
Restricted
4.0 m
f'%
displacement ductility:
Py
4.0 40 4.0
m
Py=
0.26'
Py=
0.26%
Py
1.2%
12%
J05J.
Fll
1.5
X.
1.5
M
Py
4.0 m
displacement ductility:
2!m
2.0 2.0 2.0
m m m
Py=
Py=
Py=
0.21 %
0.46 % 021 %
0.46'
JOS)
1.5
J.
1.5
Jpj].
J,6-0mJ,
Figure 5.54
The Micro model mesh for
eight-storey wall
edge
elements of 0.5
lengi correspond
elements
containing
over
ductity design
were
the
edge
elements
the
vertically) only
given
reinforcement, and for the fu ductity design 0.5 percent. The artificiay generated
to
only
wl be used.
and 5
Hz)
was
used.
displacement
macro
time is
displayed
in
figure
designs.
ment
The
finding
from the
displace
the micro
designs is
not as
large
as
expected,
is here confirmed
by
model.
144
CHAPTER FIVE
0.15
->r
~i''r~
Solutlo/i
'^
0.10
c
<D
Design Design
\ik
\x.
.'""
5.,
0.05
Q.
w
-0.05
o
oc
-0.10
-0.15 0
j L
6 Time
10
12
[s]
as cross
curvature
ductility
are not
defined
a manner
comparable
the
macro
nonlinear
dynamic capabilities
a
chapter is
devoted to
models.
5.6
Comparison
In this section,
macro
some
comparisons
will be
presented
by
the
comparison
limited
to a
presentation
at
wi be
few
roof level.
ductility design,
the
comparison is
macro
shown in
figure
5.56. It is
seen
that
During
history analysis
the
macro
model exhibits
somewhat smaller
The
same
amplitudes.
results for the fll
for the restricted
corresponding
as
ductility design
ductility design
are seen
figure 5.57.
The
observations
seconds, however, the micro model clearly shows larger amptudes, indicating more pro-
NUMERICAL EXAMPLES
145
0.15
TT"-TTpT-ri-i-|jr-
-l
r-
Macro Model
If
0.10
Micro Model
-0.15
0
'
'
1-
6
Time
10
12
[s]
ductility design
Figure 556
0.15
'
->ir-
Macro Model
"e
0.10
Micro Model
1
E
o
0.05
I
O(0
-0.05
-0.10
-0.15 0
*
'
,*
6
Time
10
12
[s]
Figure 557
nounced
Horizontal
yielding
than the
macro
shape
of die
hysteresis
loops
responsible
originales
damping
in the post
as
yielding phase.
shown in
for this
the concrete
hysteretic
behaviour
figure 4.9 combined with the reinforcement hysteretic behaviour of figure 4.14.
the reinforcement is
Of diese which
two
explanation is
that the
macro
model
146
CHAPTER FIVE
keeps
the
yield stiffness
set at
the
amplitudes somewhat more than is the case for the micro model
also tested for the
for which
no
was
did
not
change
significany. display
The
same
holds for
finer discretised
highly nonlinear
may be valid for
at
history
ductility design,
further
proof
reabity
especially
two
small overa
comparison
whereby
their
respective
use
of each model wl be
The
macro
as
model
in
a
was
from the
beginning of
this
study designed
nonlinear
parameters
etc.
design
plastic hinge,
Thus it may be
readily
modeUing and
global buding
a
structures
relatively simple mechanical nature, it is capable of simulating the global behaviour of subjected
as
structural walls
to
earthquake
action. The
typical structural
wall behaviour
characteristics such
other
macro
models. It is furthermore
typical hysteretic
implementation,
building
as
described in the
as
may be modelled
long
not
as
in the
same
direction
minor test
was
performed
on a
modified
model
prepared
for three
dimensional
with
satisfactory results.
such
as
A fll three
dimensional version is
cores
model is
capable
of
providing
aU desired
cross
sectional
quantities, such
as
ductity.
AU these
cross
quantities
sectional
a
quantities
are
dynamic Performance
in
design
NUMERICAL EXAMPLES
147
quantities,
macro
which
are
obtained
readily,
the
constitute
in the
macro
one
of the of
major advantages
smaller
of the
model
compared to the
clear
micro model
a context
advantage of
no
model is its
computational
effort.
Deliberately,
direct CPU
time
that
comparisons etc.,
reason.
performed
are
Another
important aspect of
development,
less
advantage
typically
was
three times the number of degrees of freedom, and this fact Solution of the nonlinear time
during the
storey buding.
as
macro
seen more as an
existing
use a
mainly
case
suffices for
no
some
nonlinear time
with the
model for
where
experimental
satisfactory
dynamic
may be
quite
die
as
presented
in this
new
development
to
die
macro
use as a
comparison
development steps
may be
may in their
a
essence
be
logical
development,
of
phenomena
are
simplified
efficient
areas
model. Particular
a
the
damage, such
cracking,
may be
yielding
damage analyses
performed
accurately with die micro model than with the macro model.
die micro model may be connected with
a
In the present
implementation
Zrich
are
by
in
Cooperation with
in
Software
seen
figures 5.53b-d, currentiy comprising magnified deformed shape and crack direc
yielding
of reinforcement in different directions and
are to
a
be
implemented.
macro
These
damage analysis capabities make the micro model clearly superior to the analysis cases,
as
model in certain
die
analysis
of damaged
existing wall
structures.
148
CHAPTER FIVE
superior is
irregulr,
geometrie
which is
quite often
the
case.
The
uses
membrane
its
It
as
basically consists
described in
of
addition,
local-
concrete
behaviour
cracking.
The
major drawbacks
it is its
complexity
Ue
partly
model
or
development:
a
improve
model due
to
complexity. Secondly,
a more
refined
discretization
and after that
work we.
more
clearly
during
as
for extensive
rapidly increasing Computing power, and thereby decreasing CPU time, computational effort of the micro model wl become less powerful mesh generators
the
and
important
post-processing
Software
sectional
macro
not
be
overturning
moment
sectional
not
quantities
level of the wa
as
obtained
by
the
macro
model may
also
generaUy
provide
For
be avaable. However,
could
theoretically
that. the
checking
macro
dynamic Performance
presendy offers the
design
process
involving multi-storey
The micro model
some
was, the
has still
model
most
some
development potential
analysis.
special
examples is concluded.
A discussion
number of design
problems within
in
149
CHAPTER SIX
CAPACITY
DESIGN
CONSIDERATIONS
6.1
Introduction
Among
against seismic
actions
on
reinforced
concrete
buildings,
the
capacity design
method
ly comprehensive
A
major reason
chapters,
structure
Performance of a
Performance expectations
in the
set out at
the
design phase.
Based
with
on
presented
questions regarding
on
Performance
of
capacity designed
The interest
focuses
multi-storey buding
structures in
against seismic
as
action is and
provided by
essentiaUy
cantilevers
The
capacity
as
design procedure
are
briefly
introduced of the
far
as
complete presentation
[PP92].
capacity
design
[PBM90]
plastic hinge
which may
zone
it is of
major importance
flexural
to assess
the
strength
develop in
the
severe
referred to
as
remaining
The
importance
elastic
to
overstrength
a
plastic hinge.
way that at
acts in the
plastic hinge
Another
corresponding
reason
to the flexural
overstrength,
yield.
the
important
for
knowing
the flexural
zone
on
magnitude
plastic hinge
region [PBM90].
150
CHAPTER SDC
For
chosen
and
furthermore necessary
curvature
wl
zone
develop
is based
in the
on
to the
FoUowing
ductity demand
brief
a
investigation is carried
out
of the The
develops during
nonlinear time
history analysis.
help to explain
the different
Then,
question of the distribution of flexural strength demand during design. The curtailment of the flexural reinforcing forming of plastic hinges
at
important aspect
one
other locations
at
die wa base.
on
Lasdy,
time
the
overstrength,
curvature
ductity demand,
dynamic
shear
strength demand,
and
6.2 Flexural
Overstrength
of structural wa,
In die
plastic hinge
zone
large deformations
may
develop during
at
those
which the
design strength
tensile
material
is determined,
(usuay
at
compressive edge).
reserves
At the of the
edge strain
are
mobised, referred
material
wl
develop, which are correspondingly larger than the section design strengths.
In the
for sectional
overstrengths
on
are
material
over
strength and assumptions regarding the section deformation mainly for an R/C beam.
For
history analysis
important parameters
is the
one
yielding
overstrength.
151
6D20
Figure
6D12
D10/250
Sym
ifcjo?4
Lw/2
3000
we
perform
here
small numerical
study
of the
overstrength as
to
it may
develop
for relevant wa
cross
procedure designed
introduced in
compute
three in
model is
cross
sections which
was
chapter
macro
same
material models,
chapter three,
to
more
realisticaUy.
as a
The
eight-storey buding
and 6.2. The
calculation
example.
displacement ductity
was
design
Ec
are
given again
fc'
figure 6.1.
foUowing
set
of input data
used:
33000 MPa,
19.5 MPa, Es
210000 MPa,
fy
460 MPa,
soh
Strain
hardening stiffness
Strain at concrete
Equibrium
tolerance: 0.05 MN
152
CHAPTER SDC
Three
computations
were
performed.
strain
The first
one
utised the
design
values for
strength
mean
as
no
hardening.
The second
was
yield strength
adjusted tofy
550 MPa,
computation, the
mean
value of 30
gradual
move was
as
performed
as
from the
behaviour
shown in
realistic
are
figures
6.3 to 6.6.
are
based
on
restricted
and
figures
6.5 and
based
was
on
fll
displacement ductity.
In
4.15 MN
included in the
computation
seen
significandy
to
figure
6.3 it is
mean
that
increase of flexural
strength
design values
effective
seven mean
to
effective
values for
around
percent
a
Thus
concrete
as
total
cross
sectional flexural
stl accounts for almost half of that amount Without the normal force,
at
obtain
using
effective
mean
yield strength (corresponding to the relation between die yield strengths: 550/460),
more at
and
to
the strain
hardening.
The concrete
an
additional
percent,
indicating
moment
dropped
over
40 %
the
curves
including
For the fu
mean
shows the
case
strength
simar
Thus the concrete accounts for over the half of the total
of 15 percent
that for
zero
normal force
design is found.
relatively
are
low
[PBM90] which
intended for
on
calculations of
are
suggestions values
given
for
153
25
-i
>
r"
'
'
'
20
"E
15
-t-t
il
it
E
o
10
if il il
fy fy
fy
i
460, ay
550, ay
0.0 %,
0.7 0.7
fc
=19.5
=
%, f. %, fc
19.5
30.0
-
550, ay
0.000
0.002
0.004 Curvature
0.006
0.008
(rad)
design
Figure 63
25
,.
..j
,..,-.......
,.,,,..
20
fy
fy
-
460, ay
550, ay 550, ay
0.0
%, fe =19.5
?
z
0.7 %, 0.7
fc
19.5
30.0
2
4->
15
fy
%, fc
c o
mTm-
F
o
->
10
/
i
...
...
..
0.000
0.002
0.004
Curvature
0.006
0.008
(rad)
zero
Figure design
6.4 Moment
vs. curvature
relation with
ductility
mainly due
to
the steel is mueh smaUer, and in addition that the effect of concrete is of the that of the steel. Thus, for was with axial force the total flexural
than
order
as
overstrength is
lower
the
anticipated
accounts
for
signifieant part of
overstrength.
154
CHAPTER SIX
25
-i
->
-i
r-
E
z
fy fy
460, ay
550, oty 550, ay
0.0
%, fc
19.5 19.5
30.0
0.7 %, 0.7
fc
fy
0
%, fc
0.000
0.002
0.004
Curvature
0.006
0.008
(rad)
Figure 65
25
-ii
"
'
20
fy fy
460, ay 550, ay
550, ay
0.0
%, fe
19.5 19.5
30.0
?
z
0.7 %,
0.7
fe
2
-->
15
fy
%, fe
c tu
6
2
10
"
^m.
/j"
~~
/
0
j
jiiii1_.
_ii
ii...
0.000
0.002
0.004
Curvature
0.006
0.008
(rad)
zero
Figure 6.6
The conclusions that may be drawn from the numerical smdies sections
can
on
die two wa
cross
be stated
as
foUows.
no or
usuay suffi
steel
eient concrete
area to
accommodate the
crushing of die concrete. This allows for large tense strains and forces. The large
strains
account
as
155
However, for was with considerable normal force, the compressive strain exceeds
the
point
of concrete
strength
maximum
compressive
force has
developed
strains
position
move
developed
are
only
as
large
as
bending
part of the
compressive force.
for
less dominant
overstrength.
limiting
factor in this
relatively large part ofthe total overstrength. overstrength plays emphasised that
which
There
are
as a
an
important role.
In
[PBM90]
develop in
moments
structural wa
develop.
And if larger
larger shear
forces.
Accordingly,
is considered in the
capacity design
wl be
State
procedure in the calculation of the demand of shear force. When it was found in the above
numerical studies that for was with smaUer than
the flexural
overstrength
anticipated with simpler hand calculations according [PBM90], one may Suggestion is a conservative estimate, i.e.
obtain
use
an
on
however, of importance
to
and
more
used above.
die flexural
design of
region
be
the wa
cross
section in the
region
where the
developed
plastic hinge wl
it is
important to have
be
seen
as
good an estimate
chapter,
later in this
it may be necessary to
requirement.
Ductility
Demand
In order to
adequately deta
zone
plastic hinge
zone
it is necessary to
assess
the
amount
of
deformation this
zone
may
undergo.
The curvature
ductity demand in
in two aspects.
the
plastic hinge
this
measure
of
an
structural wa is
particularly important
Firsdy,
gives
formed.
Secondly, it wl
withstand.
156
CHAPTER SIX
"9-1-- 20
ii
"^
4.
18 16 14 12
5
Jmt
-o
to
o> c
E
:3
10
8 6
4
L
/
"'"
QJ
,.,,^*
I
o
i
2h
0 0
2
4
10
12
14
16
geometrische
Wdndschlonkheit
h^/l^
ratio and
Figure
6.7 Curvature
ductility demand
as
function of slenderness
global
experimental
and
have been
performed
in
experimental
Analytical
a
investigations comprised
increasing
ductility
at
die
hinge,
and
subjected to
static
point load
at
give
a
similar results,
area
see
figure 6.7.
rather than
global
area
displacement ductility
Covers
level
shaded
is
given
different
assumptions concerning
represents
the
length
of the
plastic hinge.
edge
of the shaded
areas
yielding assumed
over a
hinge length
taken
half the
assumes
length.
[PBM90]
it
was
over
this assumed
plastic
hinge
area, and
totally elastic behaviour is assumed over the rest ofthe wa. analysis considers
modes
are
the
dynamic
only,
higher
disregarded.
chapter three a dynamic analysis series is performed in order to investigate die dynamic
curvature
ductity demand.
157
Curvature
Ductility
i
i
Demand
N
K
o
N N
\lA
5, Cty 5, Oy
0.01
8
CS
12 11
uA
0.03
0.01
0.03
II
HA=3, Oy
liA
=
10 9 8
7
3, cty
6
5
4
-m
3
2
1
I I
I.
aspect ratio
10 11 12
curvature
ductility demand,
as
same
eight-storey
structure as
used in
chapter five
=
wl here be used
a
as
reference.
5.33. For
die wall
comparison
walls
was
give
proper flexural
fundamental
frequency of Vibration was considered for a comparison walls giving different demands
from the static
reinforcement
equivalent
more
The
elements, with
total
length of 6m
(taken from the reference waU), and this arrangement was kept for mesh consistency for
aU tested walls,
not
entirely
correct
a
However, this
measure
is rather to be
regarded as
"construction"
length
with
detaing
wl be
allowing
for
major yielding,
area
yielding
by
analysis.
analyses
the
and the
analytical entirely
was
was
assumed to be
the
plastic hinge
the
determined
during the
nonnear time
particular attempts
made to suppress
158
CHAPTER SDC
plastic hinge,
wa
design
was
the
according
to
the recommended
which
yielding
compatible ground
motion
were
input
as
discussed in
under the
chapter five
was as
performed
same
premises
chapter five.
Figure 6.8 shows the dynamic curvature ductity demand obtained from the numerical comparisons. The values
stiffness ratios cty, for the
section 3.5
were are
outer
flexural
spring of the
macro
elements,
as
discussed in
taken
as
ductility demand,
two
as was
of the
chosen
represented by
shaded
curvature
yield
A
figure,
to
frequency
displayed.
tendency figure.
at
lower curvature
ductity demand
to
higher aspect
ratio is
seen
The
tendency appears
be
more
This difference
figure 6.8
was
are
integration
energy is takes
dissipated
place,
over
major cracking
at
yielding
concentration of rotation
ofthe
the
plastic hinge
load for floor
is decreased.
gravity ground
eight
floors
giving
on
the
in
section
at
the
were
employed throughout
the
analyses shown
figure 6.8. Since the results of static calculations in figure 6.7 apparendy did not consider
the effect of normal force, the smaUer gravity loads, down to
same
zero
numerical
6.8
only pertains
to
the
design phase,
easy,
i.e. it is
as
measured
during
the time
history
the
analysis,
reason
Briefly stated,
is that, due to influence of higher modes, the roof level may not be
displaced in the
159
'^roof
Figure 6.9
Global
used to obtain
same
as
for
at
positive
or
the instance
may
yielding begins
manner.
at
the wa base.
result in this
One
possibility
of
obtaining
reasonable
based upon
history analysis,
is described
displacement
static
history analysis
monotonic
analysis is performed on the same wa, using an inverted triangulr equivalent static
apped incrementaUy.
reached
The increments
are
added until
staticay
the
same
displacement is
as was
history analysis.
In the
analysis
6.9.
kept as
The
displacement
principle
figure
The
based
on
displacement
history analysis
wl here be referred to
as
the
dynamic
It wl thus be obtained
(6.1)
160
CHAPTER SIX
The assumed
not
values
at
the
be reached
exacy by this
eight-storey
global
ductiUties
to
be:
ductility (Pa
=
3):
jua*r"
=
2.9
ductility O^a
the
5):
fiA*~
5.3
In this
example
correspondence
between the
ductity values
obtained
by
this
method
to
the chosen
A number of
[Wenk93].
6.4
Energy Dissipation
comparison
between walls of different aspect ratios
It
was
In section 6.3
was
made
regarding
the curvature
ductility demand.
from static
was, the
than the
curvature
ductility demand
of the
plastic hinge
mmed
out to
expected
smdy
assumption that more energy is dissipated over die upper storeys for flexible was.
here the energy
We
time
history analysis
in order to get
as
these
assumptions. Thereby
energy
introduced
are two
foUows.
There
E.t,
ESe=~uTk.u
where the is the element
(6.2)
displacement
matrix.
The second strain energy type is die inelastic strain energy
E^ (irrecoverable strain)
also known
as
the
hysteretic energy,
,*
1
=
~u
kuu
(6.3)
where
ku
161
140 120
-iir-
Plostic
E
80
533
energy types
are
defined
as
Ev,
*r
cu
(6.4)
E.=\Pmdt
where is the
(6.5)
the
damping matrix,
and
Ek, defined
as
Et
*
mu
(6.6)
where
is the
mass
E,
transmitted
to a structure
by
ground
Et
Et.
Etl. + E9
Ek
(6.7)
We
are
here
two
especially
interested in
dissipated
ex
pressions,
dissipation, namely
viscous energy,
162
CHAPTER SIX
140
-'
'
'
'
'
'
'
'
'
'
'
'
'
'
120
--
Tl
Total
100
Plastic
f^A
r-Y1
E
ZT
80
cn
l_
60
j\)
.PA/W
r
^r-'~~
''
+-*
~~
40
\ f
20 0
-_*-LS-TTi
J
""*
,y~
1 \ 1
_i.
I_l
10
Time
[s]
ra
=
8.0
E,=E.
d
sie
+E
'-'-,
(6.8)
We
mainly
focus here
rest
on
how the
dissipated energy
as
plastic
5.33.
hinge
For
was
area
and die
the elastic
an
region.
eight-storey
wall of
chapter
selected, with
comparison
8.0, which
for
a
figure 6.8,
were
designed
here
restricted
3).
clarity,
we
display
only
the
a
dissipated
inelastic strain
well
as
the
is
displayed
as
accumulated wiU be
dissipated energy
plastic
displayed separately
so
dissipated
Figure
163
It is
seen
dissipated energy
is increased,
and this in absolute terms and in relation to the total energy. This may be
partly explained
by
Secondly, dissipated
one
dissipation
of the upper
storeys is related
flexural
cracking,
and partial
yielding.
die
to
Based total
on
the stiffer
to
dissipated
figures
a
as
from 35 %
around 24 %, which is
It
was
third
possible
to track the
by
means
of the
macro
subsequent section
of this
in deta.
6.5 Flexural
Strength
6.5.1
Impcation of numerical results plastic hinge zone is detaed, the question of the flexural strength for die rest of assumptions
of
an
Once the
equivalent
static force,
Suggestion has been made [PBM90] concerning the distribution of flexural strength from
the
plastic hinge
zone to
Suggestion
are as
foUows:
The flexural
strength is kept
plastic hinge
zone,
stretching
taken
as
length Lp upwards
horizontal
of the flexural
zone to
Lp
are
usuaUy
a
the
linear decrease
suggested stretching
or
a
plastic hinge
to
flexural
strength
corresponds
minimum reinforcement
requirements.
die wa
164
CHAPTER SIX
Biegewid erstand
der Mindesfbewehrung
Erforderlicher
Biegewiderstand
Momente
infolge
der Ersatzkr'ff e
--V)7rV7777?T
N
Biegewidersfond
am
Wandfuss
Figure 6.12
With the
Distribution
ability
to
perform
nonlinear time
we now
distribution
acting on a
the
The
relatively
stiff. In
figure 6.13
a more
height
shown to the
or
right.
frequency
frame
were
stiffening
respectively.
as
follows.
ra
Fundamental
5.33
ra
8.Q
0.67 Hz
0.40 Hz
Af,-
y^ M-
Design strength Mr
165
M
-25-20-15-10-5 0 5 10 15 20 25
/x__
.
/
-25-20-15-10-5 0 5 10 15 20 25
f1=0.67Hz
(MN111)
fi=0.40Hz
(MNm)
Figure 6.13 Effective moment distributions extractedfrom nonlinear time history analysis
for eight-storey wall [BWL92b]
more
andfor
It may be
right),
the
higher
moments are
obtained at mid
height of the
structure.
as
It should
again be pomted
moment
that the
moments at a
particular
history
same
die maximum
generaUy
not
obtained at the
In order to further
explain
the
tendency shown
in
figure 6.13
we
examine
eigen
frequencies
From
seen
figure 6.14 it is
to
seen
spectral
be im
value
compared
Relatively
seen, the
portant for the behaviour of the flexible wa. As the nonlinearities occur during the time
history analysis,
a the
eigenfrequencies
more
wl
generaUy
move
figure
flexible when
damage oceurs.
166
CHAPTER SIX
Flexible wall
|fl
{f2 j
f3
Stiff wall
jJ4%g
16% g
0.2
0.5
1.0 2.0
5.0
10
100
f[Hz]
shape
of the
an
important
role, and thus the relation between the structures' eigenfrequencies, and the shape of the
important or not
frequencies
moment
consequence, it is possible
to State
and the
over
shape of the input spectrum wl determine the distribution of bending height of the
structure.
the
The
relationship.
moment
distribution.
ground motion
ground motion
Hl
Moment distribution
consisting
higher modes
167
to
estabsh
for
given
ground motion,
upon the above
e.g. the
Suggestion, based
findings, for a
the
foUowing.
6.5.2
as
and
on
of the basic
goals
of the
capacity design.
Suggestion
is based
results obtained
by numerical analyses with the macro model developed in this report. strength was reduced over the height of a multithe
develop
as
we
at
plastic hinge,
the
elastic
region.
This
simplification
is based
on
findings
curvature
over
the
figure 6.16.
figure 6.16a,
distribution of curvature
moment
along
the
plastic hinge
height is shown,
moment curvature
and in
figure
6.16b the
distribution. The
corresponding
relation is
plastic
only
a
hinge
may
only
be about half or
third of the
M\ and the
plastic design
it appears reasonable
for
especially
dynamic analysis.
thus
The
to
the elastic
region is
168
CHAPTER SIX
Wa
height
height
03
Mt.
02 01
?
**
M2
M,
M
a) Curvature distribution
b)
moment
distribution
c) Moment versus
curvature
relation
Figure 6.16 Qualitative moment curvature distribution over plastic hinge for multi-storey
structural wall, discretised into three
macro
elements
Figure
known
6.17 shows
Suggestion
The
on
strength
in order to
avoid undesired
to
yielding.
figure.
no
major yielding is
which may
permitted. Directly
plastic hinge
zone
the flexural
overstrength
develop
The
in this
zone
region.
The elastic
region
constitutes the
rest
length Le
in the
figure.
strength in
This
the elastic
region bordering
the
is denoted Re in the
figure.
strength must be such that it can take the overstrength from the plastic hinge
169
i,
i,
1 ' 1/
/4
h
IL
.
Lec
f-ec\
"A.
Flexible wall
Stiff wa
Figure
6.17
zone
without
yielding,
an
must not
means
zone.
general way to State the required constant strength of the elastic region denoted Re in
6.17 is in
terms
figure
of the avaable
strength
in the
Rp
in
figure
Re
k0Rp
is the
(6.9)
where
kg
1.2.
strength
findings
kept
constant over a
distance Lec, is
seen
in die
on
figure. According
to
earer in this
chapter,
this
length
dependent
the
eigenfrequencies of
the wa in
we
ground motion.
find that the
For
generaUy
more
to
length Lec
be
longer die
figure 6.17
a
the
strength
distribution
for
typical stiff wa is
typical flexible
is
wa to the
right
Above the
height Lec,
strength
suggested, by curtailing
It is necessary to estabsh
we
an
length
of constant flexural
strength.
As
have
already
Lec/Le.
seen, this
length
length, i.e.
(6.10)
ac
170
CHAPTER SDL
vary between
zero
and
a
one.
However, these
motion
a
extreme
ground
input compatible
related to SIA that the modes
design spectrum,
it should be
to
estabsh
length
of constant flexural
strength
spectrum).
immediately higher
and since
than die fundamental mode will govem the result in addition to the
procedure
a
would be
quite involved
relation
for
design purposes,
we
goal
is
to
estabsh
design guideline
which may be
our
ready usable
so
wl
attempt
to use
and cabrate
history analysis.
as an
input
the fundamental
eigenfrequency
to be
of
as
designed for
that
flexural
strength (corresponding
over
to
the
overstrength
in the
the
height of the
to
wall in
strength
to
be
proportional
0.2 T\, i.e.
frequency,
to
i.e.
proportional
ct.
die fundamental
as
period T\.
proportional
be taken
ac=0.27;
(6.11)
ae=0.2
h
(6.12)
simple Suggestion
too
cannot be
regarded
as
generally
valid for
frequency regions
be used, e.g.
a
study.
high
6.5.3 Numerical
example
will here be iUustrated
two
The flexural
by
numerical
example
[BL93] involving
The two
respectively wiU
designed for
three.
we
obtain
171
0.2
0.2
1/0.67 1/0.40
0.3
0.5
Placing
these
strengths
as
over
the effective
moment
distributions
gives
reasonable
Ulustrated in figure 6.18 for the stiff wall with aspect ratio ra
are
taken
from
to
simple
normal
curtailment
possibility is
to evaluate
strength
at
force) with nominal minimum reinforcement ratio, and curtail the flexural reinforcement
of constant elastic
strength towards
reinforcing is used.
the
same
history analysis of
wa is shown in
figure
6.12. It is
seen
plastic hinge
zone,
[PBM90]
at two
more
eigen-
frequency
fi
is shown in
figure 6.20.
The
The
5.3
4.0
m.
approach suggested
zone
[PBM90] with
the
plastic hinge
several
plastic
the mid and upper storeys, which would The demand from the
figure,
the shaded
areas are
[PBM90]
or
followed. In these
stretching
over no
more
less
overstrength
may
develop
over
large parts
of the
plastic
hinge
This
a moment
means
that it is necessary
to
provide
for
an
increased
172
CHAPTER SIX
normal force
history analysis
Moment from
equivalent static
force calculation ME
(MNm)
30 25 20 15 10 5
0
Hz
^=0.67
Figure
6.18
aspect ratio ra
5.33
Design strength
minimum
zero
with nominal'
reinforcement and
Demand from
normal force
6.0 m
Design strength MR
^M*^
25 20 15 10 5
>
yR ME
(MNm)
f1=
0.67 Hz
Figure
6.19
Comparison of moment
history analysis
ra
=
to
5.33
173
Design strength
zero
with nominal
history analysis
Overstrength M
Design strength MR
20 15 10 5
>
yR ME
(MNm)
fi =0.40 Hz
Figure 6.20 Suggested strength distribution for eight-storey wall, aspect ratio ra-8.0
Design strength
minimum
zero
with nominal
reinforcement and
normal force
EE3
Unintended
caused
yielding by moment
Demand from
equivalent static
force calculation
ME
5.3
(MNm)
20 15 10 5
1=
0.40 Hz
Figure
6.21
nonlinear time
history analysis
ra
=
to
8.0
174
CHAPTER SIX
elastic
region.
to
must
yield
moment
of that
is
a
cross
section
corresponds
which
overstrength
moment
of the
phenomenon
essentiaUy
has to be
separated
approach.
shown in
the
a
difficulty pertains to the tacit assumption that the moment demand wl foow
one
a curve
similar to the
figure 6.12.
It is mentioned that
higher modes
wiU of the
may be
apparentiy
line with
never
larger than
straight
linear decrease
curve
same
figure.
not
As
we
higher modes
curve
for
flexible walls
so
enough.
on
The aspect ratio of die wall does not enter into the discussion
pertains to
[SIA160]
well
as to
the
expected demand as
are
seen
figure
6.12 is
severe
they
designed
for
in countries with
a
considerably stiffer,
would be
and for
given
typical
Europe.
The demand in
as
figure
more
typically reflect
the demand
not
The here
impact of the above results on the design of the flexural reinforcement is discussed
the wa with aspect ratio of 5.33. The flexural
briefly for
strength
in the
plastic
of
hinge region
amounts to
was
computed
at a
strain
at
the
to
0.0035 and
was
no
computed using
ordinary assumption
the
design strength
plastic hinge as
Re
strain
hardening. Using
overstrength factor
above the
Xo we may obtain
XoRp
1.2(19.4)
23.3 MNm
According to the above it is required that the wa in the elastic region reaches a flexural
strength of 23.3
the confined
MNm. This
can
be achieved
at
zones at
the ends.
Adding
figure
four D20
each end
as
shown in
cross
figure
6.23
gives
an
elastic
the
section of the
to
plastic hinge
guarantee that
zone
are not
quite enough
no
the
ns
6D20
D10/200
Sym
500
-r
1^/2
3000
Figure
6.22 Wall
cross
section
ofplastic hinge
zone
10D20
A+wm
500
D10/200
Sym
Jsection
Lw/2
3000
Figure 6.23
Wall
cross
calculated to be 23.3 MNm. However, this difference of around three percent may be tolerable. The
cross
section
strengdiened
example
be necessary to
safeguard against uncontroUed development of plastic hinges in proposed that the additional flexural reinforcement bars be practicaUy allow for enough
and the bond. in order to
arranged
in
U-shaped loop
region
Figure
24
between the
plastic hinge
(left) and the vertical section A-A (right) which is indicated in plan in figure upwards
on
up
opposite bar,
on
or
may be
spced
place
at
using
the
sptice length
depending
24
(right)
it is
important
to
corners
ofthe
U-shape. by
the arrangement of the reinforcement
proposed here,
zone
better
opportunity compared
to
to
determine the
more
geometrical
extension of the
as
plastic hinge
a
is
given
of the
previous
vague
now
assumptions such
e.g.
height of
e.g. be chosen to be
exacy equal
the
height
advantageous
of a construction
point of view.
176
CHAPTER SIX
Sym
A*---
A-A
flffi
IIE Eiill
Two horizontal
Hill
EllSC
bars D10
U-shaped
barD20
fc
elastic
500
1^/2
3000
300
the
It is difficult to suggest
good
strength
we
to
be
used
on
die
observe
that the
moment at
appears to be
sghy
proposed
we
value of 20 % is
Findings
on
slighy increased flexural strength right above the plastic [Haas93] in his smdy
a
hinge
were
also made
by
Haas
on a
wall with
an
fundamental
frequency
using
of this report
It is clear that
approach
would work
generaUy for
extensive
in order
177
The
dynamic
shear forces
developed
at
the base of
multi-storey
structural wall
during
severe
seismic action
cross
changes
more
rapidly
than die
sign of die
bending moment.
2) The magnitude of the
the shear force obtained
cross
only from the equivalent static force method. experimental studies, [ES 89],
to
Both these
phenomena
do with the
the several
over
higher
modes which
are
contributing
a
simultaneously
force
to
the deformations
the
height
shear
brought down to the base which changes its sign more rapidly.
to
This is due
the fact that the few lowest modes have the bulk of their inertia forces
to
by
periods belonging
to these
by
the
eight-storey
wa used in
chapter
five and
previously in this chapter by transforming the overturning moment at the wa base from
the time
into the
frequency
domain.
Figure 6.25
(visible
shows
overturning moment at
It is
seen
ground motion
at
input,
and restricted
around 0.5
including
(at
4
die
yielding phase)
same
dominates.
Figure
a
6.26 shows
wall, displaying
even
something
from mode 4
(around 15 Hz).
partly
due to die
developed
flexural
overstrength
at
the
partly to the effect of low centre of gravity of the horizontal inertia just discussed, giving
lower than
a
forces of
higher
modes
resulting
centre
may be located
considerably
anticipated by
the inverted
triangulr
force
distribution of the
178
CHAPTER SIX
5
ri
-"
10 f
15
20
[Hz]
moment at
=
Figure design,
of overturning
ductility
533
0.3
0.2
^^10 f
n-^*V-
15
20
[Hz]
restricted
5.33
on
eight-storey
some
wall
and
on
in that
chapter and to
degree
previous discussion
this chapter.
179
during
history analysis
as
the
cross
macro
displayed
versus
time in
well
as
for the
fll
ductity design,
oceurs at
magnitude
amounts to
ductility
designs, respectively.
We wiU discuss here the numerical result in relation to the
design
criteria
are
set
for shear
force
method. The
higher dynamic
to
shear forces
recognised in
[PBM90] according
the
foUowing
(6.13)
K,
where
<M\,>oy
is the
V^,
design
is the
VE
static force
[NZS3101].
treated earer in this
overstrength
was
chapter,
dynamic
six storeys:
cov
0.9 +
(6.14)
Buding
with
more
<
1.3
<
1.8
(6.15)
In
our
8, giving cov
1.57.
Starting
VE
triangulr force
obtained
as
distribution of the
MN. The flexural
0.74
the rnge of 15
to
we
history analysis,
the
right hand
to
know
VE
and
we
overstrength
reached
according
dynamic magnification
may establish
a
compared
as
to
the
total shear
magnification
the
produet
of the flexural
180
CHAPTER SIX
a).O0fa,
example
we
(6.16)
obtain
For our
m.
equal
to
2.310.1 A
in
examples presented
the time
overstrength
5.19) gives
respeet
to
moment
a
during
figure
cross
overstrength
the
factor with
available
22.5/19.4
would rather
use
overstrength
=
gives <boa
22.5/16.1
1.40.
we
may
now
3.11/1.40
2.22.
clearly higher
large
number of storeys
given by
expression (6.15),
of the
same
expression using the correct number of storeys equal to eight. The difference
=
is ofthe
magnitude 2.22/1.57
as
equal
to
2.1/0.43
4.88. From
figure 5.20
19.0/9.7
the maximum
bending
moment
ductility is about 19.0 MNm, which gives the flexural overstrength with
moment as
=
respeet
to die
=
equivalent static
2.49. By
as
4.88/1.96
2.49/1.57
comparison
with the
suggested magnification
this value is
capacity
due to minimum
reinforcing requirements
smaller
of the code
not
(which
are
also
capacity design
section had
a
cross
examples,
the minimum
reinforcement
requirement
is
more
capacity
numerical
examples
several
were
presented here,
that the shear
together with
behaviour reaches
large values
Keintzel
on
dynamic
report)
During
found that
181
design,
it will
not
be
attempted
here
to
derive
an
improved
model
what
history analyses
presented
using
the
macro
Merely a few
as to
possibility would be
die
to
dynamic magnification
give die
and
by modifying
expressions (6.14)
(6.15).
expression (6.13)
is
on
by
die
overstrength
partially
for
dynamic magnification
be
suggested with
the
expression
chapter,
resulting overall
the total
fairly realistic
as
estimate ofthe
dynamic
shear forces.
However,
as
well
as
the
constituents
forming
dynamic
shear force
display
other, and this should be the subject of further research and considered in future design
formulae.
182
183
Summary
The
models
for the Simulation of the behaviour of multi-storey reinforced concrete structural walls in
budings subjected
walls. After
an
to
earthquake action,
a
with
introduction and
review of
working
in
cross
sectional
structural wa
by
means
of nonnear
Springs.
on
emphasis
fulfling simple
on a
kinematic
hysteretic
rules
are
derived
largely based
knowledge
of the basic
physical
behaviour
complemented
with
empirical
macro as
element is derived.
on
by accommodating
the
macro
composite
were
clear
manner.
were
programmed
and
implemented
in
an
Option.
The models
developed Appendix
a
according to speeifications in
Both models had to
of this report
undergo
series of tests, which served the purpose of static tests and gave
an
checking
the
estimate of parameter
behaviour of
influence. Since
suitable
multi-storey was, the micro model had to serve as comparison for the macro model.
After
a
chapter devoted
a
to
reliabity
tests
and tests
some
on
the
Performance
of
capacity
capacity analysed
further
chapter
deals with
specific problems
of the
developed here. by
macro
wall is modelled
to
elements and
dynamically. Special
paid
the
ynamic
curvamre
ductity
demand in the
184
plastic hinge,
die
moment
improved
over
the
Conclusions
was
numerical
model for
For the
macro
analysis tasks.
walls of
to
Performance check
was
multi-storey
the
model
found
especiay useful.
This is due
realistic
hysteretic behaviour,
of monitoring
cross
capabity
sectional quantities,
and curvature
-
ductity, and
The
reliabity
data from
static tests. It
was
macro
model
capable
of
simulating
the most
important aspects
have
However, for dynamic behaviour no suitable experimental data exists, and in order to
a
comparison
macro
was
concluded that
developed,
serve as a
when detaed
The
macro
model
relatively simple
number of
to compre-
hend. Extensive
knowledge
use.
standing
for
a
or
spring members
fuU
dynamic
behaviour. This is attributed to the direct way behaviour. Furthermore, the effect of
can
parameter modifications
model.
The model,
as
the
global behaviour
be
easily followed
in the
macro
implemented and
extensive
described in the
Compared viding
to
important during
a
reasonable
on
provide mueh
macro
Information
locased
yielding.
The
model
185
also
inherentiy
assumes a
regulr
wa.
wall geometry,
Symmetrie
cross
section, and
cannot
developed
possible
to
clear and
simple
uniformly
was
found that it is
concrete
usuay adequate
possible
to
divide the contributions to the material modulus matrix into different parts. The interaction
transparent
manner,
generalised into different levels. The major advantage of the micro model is
of local
an
damage,
such
as
direction of
in
a
cracks and
of reinforcement. Furthermore,
more
irregularity
or
opening
are
wa
may be
important drawbacks
as
its
general
inabity
monitor
sectional
quantities,
demands
such
ductity,
and its
high computational
black box, the
compared
the
macro
model, which is
use
in order not to
the
ought to
materials.
The
Performance
tests on the
example
of a
respects
the time
history analysis
curvamre
indicated that
particular care
in the
as a
design
must
be taken. The
dynamic
ductity
demand in the
plastic hinge
displacement ductility,
existing suggestions
capacity
design method.
The distribution of flexural
strength
is
one
problematic
by
design recommendations do
true
not
in the
case
of slender walls. It
shown
an
energy
study
that
slender wa the
case
dissipates more
for
a
how to
improve
appeared
to
two
chosen
examples,
The maximum base shear obtained from the time exceeded the ations.
considerably
recommend
anticipated dynamic
shear force
given
in the
capacity design
186
The models
developed
regulr
only
been
presented
an
in their basic
estimation of the
seen
in
perspective,
design
analysis of existing
large structures,
models be
performed
so as to
enable die
modeUing
of
more
compcated
was found in
connections
to
frames,
cores, such
as
Regarding the
an
extensive
parametric study on improved design parameters should be carried out Attempts should
be made to try to confirm the
on
by further
experimental
The
history analyses,
time
in
particular, should
be carried out
larger magnification
by
an
during
confirmed
additional
parametric study
be extracted
as
and
to
opposed
187
ZUSAMMENFASSUNG,
SCHLUSSFOLGERUNGEN
UND
AUSBLICK
Zusammenfassung
Das Ziel der
in Gebuden, insbesondere
kapa
Einfhrung
grund-
bisherige
zur
Weiterentwicklung ausgewhlt.
ModeUtyps entwickelt
wobei der
ErfUung
ein
tereseregeln basieren
auch durch
elementes wurde ein
Das zweite ModeU, MikromodeU genannt basiert auf der Kontinuumsmechanik und
auf nichtnearen
Stoffgesetzmodellen.
Modelltyp
wurde unter
Bercksichtigung
hergeleitet
Finite Elemente
programmiert
und in einem
gegebenen
Programm implementiert.
Anhngen A und durchgefhrt,
Da keine
B ist
Sowohl mit dem Makro- als auch mit dem MikromodeU wurden numerische Testserien
die dazu dienten, die
zu
aufgrund
von
Daten
statischer Versuche
berprfen sowie
Modellparameter abzuschtzen.
geeigneten
zur
Versuchsdaten
von
Tragwnde
Verfgung standen,
Kapitel,
dienten die
Vergleichsbasis
zur
fr das MakromodeU.
Nach einem
Ueberprfung
Zuverlssigkeit numerischer
bezglich
kapazittsbemessene Gebudes
188
gewidmet ist,
Kapitel
auf
der Be
ent
Kapazittsmethode eingegangen.
dynamisch
Krmmungsduktittsbedarf
Eni
und
Querkraftbedarf untersucht.
Vorschlag
vorgesteUt
Schlussfolgerungen
In der
Einfhrung
hingewiesen,
dass
es
schwierig ist,
ein fr alle
Ueberprfung
des Verhaltens
kapazittsbemessener Tragwnde
und
zwar
geeignet erachtet
aufgrund
Hystereseverhaltens und
wegen seines
begrenzten Rechenaufwandes.
wurden
Die
aufgrund
von
Ver
festgesteUt dass
das Makro
modeU
fhig ist
die
wichtigsten Aspekte
zu
metermodifikationen
We
vom
simulieren.
Verhalten
dynamischen
mehrstckiger Tragwnde
kerne
geeigneten Ver
suchsdaten voregen, wurde ein MikromodeU entwickelt das als Vergleichsbasis bei der
dienen soU, und das auch als zustzliches ModeU verwendet werden kann.
verstndlich. Vertiefte Kentnisse der Mechanik kontinuierlicher Medien sind weder fr das Verstehen des Modes noch fr dessen
Anwendung notwendig.
Das MakromodeU
Beschrei
er
besteht
aus
der absolut
bung
der
Das ModeU
und
dynamischen
Parametermodifikationen
189
es
implementiert
und im
Anhang
A beschrieben
wurde, ist
dynami
Rechenaufgaben,
Systeme
es
steen. Whrend
nicht in der
globales
Verhalten
zeigt
ist
dagegen
Lage,
Rissrichtungen
und lokales
Fessen der
Bewehrung
zu
Wandquerschnitt
Oeffnungen,
auf
zu
demzufolge
Lage, Abweichungen,
wie z.B.
zeigt
dass
es
auch
mgch ist
die
gleichmssig
bewehrte
Tragwnde
zu
gengt
nur
realistische
globale Ergebnisse
gezeigt
dass
es
Materialmodulmatrix in separater Form darzustellen. Die Interaktionseffekte zwischen Stahl und Beton knnen in einer klaren Weise modelliert und auf verschiedenen Stufen
generalisiert richtungen
knnen
werden. Der
es
Hauptvorte betreffend
die
Einige
keine dass
schwerwiegenden
dass
es
i.a. und
Querschnittsgrssen
meistens
Krmmungsduktilitt wiedergibt,
was
es
verhltnismssig
rechenintensiv ist,
Lsung
wenn
ist Im weiteren
gut auch,
dass
das MikromodeU nicht als eine Black Box bentzt werden soU, und
so muss
der Anwender
des Verhaltens
Tragwerke deuten
mit den
heutigen Empfehlungen
der
Kapazittsbemessung
Ergebnisse
Zeitverlaufsberechnungen
deuten
sichich der
Angaben
Verteung
von
Biegekapazitt
problematisch sein,
heutigen Bemessungsempfehlungen
190
ermgchen
FaU bei in
emer
Lsung
mehr
gedrungenen und
aber noch
beschriebene das
Biegebewehrung
scheint fr
Im weiteren ist
aus
der
Zeitverlaufsberechnung resultierende
maximale
Querkraft am
Ausblick
nur
Vergleiche
zu
Versuchsergebnissen
und
eines
Tragwandkragarmes ermgchen.
ein
erweiterten und fr
Perspektive gesehen,
Arbeit noch nicht
Werkzeug
darzustellen fr
Bemessungsuntersttzung
Entwicklung
so
kompti-
zierteren
Tragwnden,
wird. Besonders
-
gedacht wird
zu
Tragwnde
mit mit
Koppelung
Rahmensystemen,
Momentenverteilung
experimenteUe
rische Studien
zu
besttigen.
Biegekapaziresul
tierte, sote durch eine weitere Parameterstudie berprft werden. Ebenso soten dyna
mische Versuche
messen, was
es
an
der Wand
zu
bei den
191
REFERENCES
[Abaq89] [Abaq91]
[Abra91]
Abaqus,
Abrams, D.P. "Laboratory Definitions of Behaviour for Structural Compo nents and Buding Systems", Earthquake-Resistant Concrete Structures, Inelastic Response and Design, S.K. Gosh, Editor, ACI Special PubU cation SP-127, American Concrete Institute, Detroit 1991, pp. 91-152
[ACISP84]
Earthquake Effects on Reinforced Concrete Stractures, US-Japan Research, J.K. Wight, Editor, ACI Special Publication SP-84, American
Concrete Institute, Detroit, 1984
[Bach67]
Bachmann, H., "Zur plastizittstheoretischen Berechnung statisch unbe stimmter Stahlbetonbalken", Dissertation, Institut fr Baustatik, Eidge nssische Technische Hochschule (ETH), Zrich, 1967
[Bach89]
Bachmann, H.
et
[Bach91]
Bachmann, H., "Earthquake Design of Bridges, The Swiss Code App roach", European Earthquake Engineering Journal, No. 3., 1991, and Bericht Nr. 187, Institut fr Baustatik und Konstruktion, Eidgenssische Technische Hochschule (ETH), Zrich, 1992
[Bach93]
Bachmann, H., "Erbebensicherang von Bauwerken", Vorlesungs-Autographie, Institut fr Baustatik und Konstruktion, Eidgenssische Techni sche Hochschule (ETH), Zrich, 1993 Bertero, V.V., Aktan, A.E.,
[BACS84]
Simulation Tests and Associated Studies of a l/5th Scale Model of a 7Story Reinforced Concrete Test Structure", US-Japan Cooperative Earth quake Research Program, Report No. UCB/EERC-84/05, Earthquake Engineering Research Center, University of California, Berkeley, 1984
[Bath82]
Bathe, K.-J., "Finite Element Procedures in Engineering Analysis", Prentice-HaU Inc., Englewood Cliffs, New Jersey, 1982
[BG80]
April, 1980,
pp. 819-42
[BG84]
Bazant, Z.P., Gambarova, P.G., "Crack Shear in Concrete, Crack Band Micro-Plane Model", J. Struct. Eng., ASCE, Vol. 110, No. 9, Sept, 1984, pp. 2015-35 Bachmann, H., Linde, P "Dynamic Ductity Demand and Capacity Design of Earthquake-Resistant Reinforced Concrete Walls", Proc. Tom Paulay Symposium, University of California, San Diego, Sept. 20-21, 1993, ACI Special PubUcation, American Concrete Institate, Detroit 1993
[BL93]
192
REFERENCES
[BP90]
von
Stahlbetontrag
wnden unter Erdbebeneinwirkung", Beton- und Stahlbetonbau, 85, 1990, H.ll, und Bericht Nr. 177, Institut fr Baustatik und Konstruktion,
[BWL92a]
Hybrid Reinforced Concrete Frame WaU Buildings", Nonlinear Seismic Analysis of Reinforced Concrete Buildings, Fajfar, P., Krawinkler, H., Editors, Elsevier Applied Science, London, 1992
[BWL92b]
Bachmann, H., Wenk, T., Linde, P., "Seismic Ductility Demand in
Gravity
[CB89]
Budings",
Chappuis, P., Bachmann, H., "Concrete material model for earthquake loading including strain rate and loading history", Bericht Nr. 172, Institut fr Baustatik und Konstruktion, Eidgenssische Technische Hochschule
(ETH), Zrich, Juni, 1989
Cervenka, V., "Inelastic Finite Element Analysis of Reinforced Concrete Panels", dissertation, Dept of Civil Engineering, University of Colorado, Boulder, 1970
Cervenka, V., Gerstie, K.H., "Inelastic Analysis of Reinforced Concrete Panels: Theory", Publ. IABSE, Vol. 31-11, Zrich, 1971, pp. 31-45
[Cerv70]
[CG71]
[Chap87]
Chappuis, P., "Modelisation non-lineaire du comportement du beton sous des solcitations dynamiques", Dissertation, Bericht Nr. 155, Institut fr Baustatik und Konstruktion, Eidgenssische Technische Hochschule (ETH), Zrich, Juni, 1987
Chen, W.F., "Plasticity in Reinforced Concrete", McGraw-Hl, New York, 1982
Cornelissen, H.A.W., Hordijk, D.A., Reinhardt, H.W., "Experimental
determination of crack-softening characteristics of normal weight and ght weight concrete", Heron, Delft University of Technology, Vol. 31, No. 2, 1986
[Chen82]
[CHR86]
[Colo93]
Colotti, V., "Shear behavior of RC Structural Walls", /. Struct. Eng., ASCE, Vol. 119, No. 3, March, 1993 Cervenka, V., Pukl, R., Eligehausen, R., "Computer Simulation of anchoring technique in reinforced concrete beams", Proc. Int. Conf. Computer Aided Analysis and Design of Concrete Structures, Part I, N. Bicanic et al, Editors, Pineridge Press, Swansea, 1990
[CPE90]
[Curb87]
Curbach, M., "Festigkeitssteigerung von Beton bei hohen Belastungs geschwindigkeiten", Dissertation, Heft 1, Schriftenreihe des Instituts fr
Massivbau und
Baustofftechnologie,
[deBN85]
deBorst, R., Nauta, P., "Non-orthogonal cracks in element model", Eng. Comp., Vol. 2, March, 1985
smeared finite
REFERENCES
193
[Ding85]
Dinges, D., "Vergleichende Untersuchungen von Stahlbetonflchentragwerken unter Bercksichtigung physikalischer und geometrischer Nichtnearitten", Dissertation, Fachbereich Konstruktiver Ingenieurbau, Tech
nische Hochschule Darmstadt, 1985
[DO90]
Dahlblom, O., Ottosen, N.S, "Smeared Crack Analysis using Generazed Fictitious Crack Model", /. Eng. Mech., ASCE, Vol. 16, Jan., 1990
[Drr80]
zur
Berechnung
von
Stahlbetonscheiben
unter
Bercksichtigung des Verbundverhaltens", Dissertation, Fach Konstruktiver Ingenieurbau, Technische Hochschule Darmstadt
to
[ES89]
study the Response Yielding Columns", dissertation, Report UILU-ENG-89-2007, Department of Civ Engineering, University of Dnios, Urbana-Champaign, 1989
of Reinforced Concrete Frame-Wa Stractures with
[FF91]
"Mathematical
Rotterdam,
[Fran70]
Franklin, H.A., "Nonlinear Analysis of Reinforced Concrete Frames and Panels", dissertation, Report No. UC-SESM 70-5, Dept. of Civ Engi neering, University of California, Berkeley, 1970
[FVSFZD90] Fischinger, M., Vidic, T., Selih, J., Fajfar, P., Zhang, H.Y., Damjanic, F., "Validation of a macroscopic model for cycc response prediction of RC was", Proc. 2nd int conf. on Computer aided analysis and design of concrete structures, Zell am See, April 4-6, 1990, Bicanic, N. Editor, 1990 Swansea Pineridge Press,
,
[Glem84]
Glemberg, R., "Dynamic Analysis of Concrete Stractures, Constitutive ModeUing, Dynamic Substructuring, Computer Implementation", disser tation, Publication 84:1, Department of Structural Mechanics, Chalmers University of Technology, 1984 Gupta, A.K, Maestrini, S.R., "Tension Stiffness Model for Reinforced Concrete Bars", J. Struct. Eng. ASCE, Vol. 116, No. 3, March, 1990
Glemberg, R., Samuelsson, A.,
crete
[GM90]
[GS83]
Structures", Nordic Concrete Research, PubUcation 2, The Nordic Concrete Federation, Oslo, 1983
[Gust85]
Modelling
Applied Strength
Analysis", dissertation, Report TVBM 1007, Division of Building Materials, Lund Institute of Technology, 1985
[GyU83]
Gylltoft, K.,
Engineering,
fatigue
in concrete
University
of Technology, 1983
[GyU89]
Gylltoft, K., "Cyclic Loading", in Ch. 6: "Other Aspects of Material Modelling", Fracture Mechanics of Concrete Structures, Elfgren, L., Editor, RILEM Report, Chapman and Hau, London, 1989, pp. 172-187
194
REFERENCES
[Haas93]
von Stahlbeton-Tragwandsystemen nach Nachdiplomarbeit, Institut fr Baustatik und Konstruktion, Eidgenssische Technische Hochschule (ETH), Zrich,
1993
[Hass91]
zone response to simultaneous normal and shear displacement", RILEM Proc. 6, Analysis of Concrete Structures by Fracture Mechanics, Elfgren, L., Shah, S.P., Editors, Workshop dedicated to Prof. A. HiUerborg, Abisko, Chapman and Hall, London, 1991
[Hass92]
zones
by simultaneously apped
tation, Report TVBM 1010, Division of Buding Materials, Lund Institute of Technology, 1992
[HHT77]
for Time
& Struct.
Hilber, H.M., "Analysis and Design of Numerical Integration Methods in Structural Dynamics", dissertation, Report No. UCB/EERC-76/29, Earth quake Engineering Research Center, University of California, Berkeley,
1976
[Hiro75]
Walls and
Hirosawa, M. "Past Experimental Results on Reinforced Concrete Shear Analysis on Them", Structural Research Pubcations, No. 6, 1975, Building Research Institute, Tsukuba, Ministry of Construction,
Japan, [HMM91]
1975
Hayami, Y., Miyashita, T., Maeda, T., "Nonlinear Analysis of Shear Walls subjected to Cycc In-Plane Shear Loading", internal report, Infor mation Processing Center, Kajima Corporation, 2-7, 1-chome, Motoakasaka, Minato-ku, Tokyo, 1991 Hillerborg, A., Modeer, M., Petersson, P.E., "Analysis for crack formu lation and crack growth in concrete by means of fracture mechanics and
finite elements", Cement and Concrete
[HMP76]
Resesach, 6,1976
[Hugh87]
[Kein88a]
Hughes, T.J.R., "The Finite Element Method, Linear Static and dynamic Finite Element Analysis", Prentice-Hall, Englewood Ctiffs, N.J., 1987
Keintzel, E., "Zur Querkraftbeanspruchung von Stahlbeton-Wandscheiben unter Erdbebenlasten", Beton- und Stahlbetonbau 83, H. 7,1988
[Kein88b]
[K0U88]
Keintzel, E., "Vereinfachte Ermitung der Querkraft aus Erdbebenlasten in Stahlbeton-Wandscheiben", Beton- und Stahlbetonbau 83, H. 12,1988
Kollegger, J., "Ein Materialmodell fr die Berechnung von Stahlbetonflchentragwerken", Dissertation, Fachbereich Bauingenieurwesen, Ge
samthochschule Universitt Kassel, 1988
[Kolm85]
REFERENCES
195
[KP71]
Struct. Div.
Kent D.C, Park, R., "Flexural Members with Confined Concrete", J. ASCE, Vol 97, No. 7, July, 1971, pp. 1969-90
[KSOA82]
of the
Fu-Scale Seven-Story Reinforced Concrete Test Structure: Test PSD-3", Proc. 3rd Joint Technical Coordinating Committe, US-Japan Cooperative
Earthquake
1982
Research
Program, Building
[Kupf69]
Kupfer, H., "Das Verhalten des Betons unter zweiachsiger Beanspruch ung", Bericht Nr. 18, Lehrstuhl Massivbau, Technische Hochschule,
Mnchen, 1969
[Lars59]
"Bearing Capacity of Piain and Reinforced Concrete Walls", dissertation, Doktorsavhandngar vid Chalmers Tekniska Hgskola Nr. 20, Gumperts Frlag, Gteborg, 1959
Larsson. L.E.,
[Lind88]
Linde, P., "Experimental and Analytical Modeng of R/C Shear Was", Engineer's thesis, Department of Civil Engineering, Stanford University,
1988
[Lind89]
Linde, P., "Analytical Modeling Methods for R/C Shear Walls", Report No. TVBK-1005, Department of Structural Engineering, Lund Institute of
Technology, 1989
[MB87]
Moazzami, S., Bertero, V.V., "Three-Dimensional Inelastic Analysis of Reinforced Concrete Frame-WaU Structures", Report No. UCB/EERC87/05, Earthquake Engineering Research Center, University of California,
Berkeley, 1987
[MKHH91]
Meskouris, K., Krtzig, W.B., Hansktter, U., Hoffmeister, P., "Nicht
lineares seismisches Tragverhalten ebener und rumcher Stahlbeton tragwerke", Arbeitsbericht fr die Jahre 1989/1991, Wissenschaftche Mitteilungen, SFB-151-Berichte Nr. 20, Sonderforschungsbereich Trag werksdynamik, Ruhr-Universitt Bochum, 1991
[Moaz91]
Moazzami, S., "An Analytical Model Designed to predict the Behavior of Shear Walls", Proc. Int. Conf. Buildings with Load Bearing Concrete Walls in Seismic Zones, Paris 1991, pp. 359-67
,
[MP90]
Moser, K., Paulay, T., "Kapazittsbemessung von erdbebenbeansprachten Stahlbetonrahmen", Schweizer Ingenieur und Architekt, Nr. 44, Nov. 1990, und Bericht Nr. 180, Institut fr Baustatik und Konstruktion, Eidgenssische Technische Hochschule (ETH), Zrich, Nov., 1990 Nson, A.H., "Nonnear Analysis of Reinforced Concrete by the FEM", ACI Journal, Vol. 65, No.9, American Concrete Institute, Detroit, 1968, pp. 757-66
Nooru-Mohamed, M.B., "Mixed-mode fracture of
mental
concrete: an
[Ns68]
[Noor92]
experi
Delft Univer
Analysis of Reinforced Vol. 64, No.3, American Concrete Concrete Beams", ACI Journal Institute, Detroit 1967, pp. 152-63
,
"Finite Element
196
REFERENCES
[NZS3101]
NZS 3101: Part 1, Commentary NZS 3101: Part 2: New Zealand Code of Practice for the Design of Concrete Structures, Standard Association of New Zealand,
Wellington,
[01de85]
Analysis of Tensile Fracturing Stractu res", licentiate thesis, Report 1985:010L, Division of Structural Engi neering, Lulea University of Technology, 1985
Oldenburg, M.,
"Finite Element
[OTTKW89]
Ohomori, N., Tsubota, H., Inoue, N., Kurihara, K., Watanabe, S.,
"Reinforced Concrete Membrane Elements
In-Plane Shear Stress", Nuclear pp. 61-72
[PBM90] [PL74]
Paulay, T, Bachmann, H., Moser, K., "Erdbebenbemessung von Stahl betonhochbauten", Birkhauser Verlag, Basel, Boston, Beruh, 1990 Paulay, T., Loeber, P.J., "Shear transfer by aggregate interlock", Shear in Reinforced Concrete, ACI Special Publication SP-42, American Concrete
Institute, Detroit, 1974, pp. 1-15 Park, R., Paulay, T., "Reinforced Concrete Structures" John Wiley & Sons, New York, 1975
[PP75]
[PP92] [Prui85]
Paulay, T., Priesey, M.J.N., "Seismic Design of Reinforced and Masonry Buildings" John Wiley & Sons, New York, 1992
Concrete
Pruijssers, A.F., "Description of the stiffness relation for mixed-mode fracture problems in concrete using the rough-crack model of Walraven", Report No. 5-85-2, Stevin Laboratory, Department of Civ Engineering, Delft University of Technology, 1985 Pruijssers, A.F., "Aggregate Interlock and Dowel Action under Monotonie and Cycc Loading", dissertation, Department of Civ Engineering, Delft University of Technology, 1988
Rashid, Y.R., "Analysis of prestressed concrete pressure vessels", Nuc lear Engineering and Design, Vol. 7, No. 4,1968, pp. 334-44
Reinhardt, H.W., Cornessen, H.A.W., "Post-Peak Cyclic Behaviour of
Concrete in Uniaxial Tense and
[Prui88]
[Rash68]
[RC80]
Alternating
Tensile and
[Rein84]
[RK88]
[Rots84]
Rots, J.G., "Variable reduetiefactor voor schuifweerstand van gesheurd beton", Report No. BI-84-33, Institute TNO for Building Materials and Stractures, Delft, 1984 Rots, J.G., "Computational Modeling of Concrete Fracture", dissertation,
[Rots88]
[RR92]
Dept
of Civ
Engineering,
Delft
University
of Technology, 1988
Rule, W.K., Rowlands, R.E., "Predicting Behavior of CyccaUy Loaded RC Stractures", J. Struct. Eng., ASCE, Vol. 118, No. 2, Febr, 1992
REFERENCES
197
[Saat91]
Saatcioglu, M., "Modeling Hysteretic Force-Deformation Relationships Earthquake-Resistant Concrete and Inelastic Structures, Response Design, S.K. Gosh, Editor, ACI Concrete Institute, Detroit, 1991, American PubUcation SP-127, Special
for Reinforced Concrete Elements",
pp. 153-198
[SB90]
Sedarat, H., Bertero, V.V., "Effects of Torsion on the Linear and Nonlinear Seismic Response of Stractures", Report No. UCB/EERC90/12, Earthquake Engineering Research Center, University of California,
Berkeley,
[SE92]
1990
Sozen, M.A., Eberhard, M., "Design Shear for Earthquake-Resistant Walls", Nonlinear Seismic Analysis and Design of Reinforced Concrete Buildings, Suppl. Proc. of a Workshop held in Bled, Slovenia, July 1316,1992, J. A. Blume Earthquake Engineering Center, Report No. 103,
Dept
[SIA160] [SIA162]
of Civ
SIA Norm 160, Einwirkungen auf Tragwerke, Schweizerischer und Architekten-Verein, Zrich, 1989
Ingenieur-
und Archi
tekten-Verein, Zrich,1988
[Skri82]
unarmierten Betonbauten unter Erdbebenbeanspruchung", Disserta tion, Institut fr Baustatik und Konstruktion, Eidgenssische Technische
[T055]
Tomii, M., Osaki, Y., "Shearing Resistance of Reinforced Concrete Bracing Was", Transactions, Architectural Institute of Japan, No. 51, September, 1955, pp.83-95
[VB87]
Vulcano, A., Bertero, V.V., "Analytical Models for Predicting the Lateral
Response
[VBC88]
Vulcano, A., Bertero, V.V., Colotti, V., "Analytical Models of R/C Structural Was", Report No. UCB/EERC-88/17, Earthquake Engineering Research Center, University of California, Berkeley, 1988
Vallenas, J.M., A., Bertero, V.V., and Popov, E.P., "Hysteretic Behavior of Reinforced Concrete Was", Report No. UCB/EERC-79/20, Earthquake Engineering Research Center, University of California, Berkeley, 1979
[VBP79]
[Walr80]
Wahaven, J.C, "Aggregate interlock, a theoretical and experimental analysis", dissertation, Department of Civ Engineering, Delft University
of Technology, 1980
[WB90]
Wenk, T., Bachmann, H. "Ductility demand of 3-D reinforced concrete frames under seismic excitation", Proc. Europ. Conf. on Struct. Dynamics, Eurodyn '90, Bochum, A.A. Balkema Rotterdam, 1991, pp.
,
537-41
[Wenk90]
einwirkung", DAfStb-Forschungskolloquium,
198
REFERENCES
[Wenk93]
von Stahlbetonrahmen unter Erd Dissertation in bebeneinwirkung", Vorbereitung, Institut fr Baustatik und Konstruktion, Eidgenssische Technische Hochschule (ETH), Zrich, 1993
[WK85]
Wallace, B.J. Krawinkler, H., "Small-Scale Model Experimentation on R/C Assembes", US-Japan Cooperative Earthquake Research Program, Report No. 74, J.A. Blume Earthquake Engineering Research Center,
Stanford
University, 1985
Wenk, T., Linde, P., Bachmann, H., "User Elements Developed for the Nonnear Dynamic Analysis of Reinforced Concrete Stractures", Proc. Abaqus Users' Conference, Aachen, June 23-25,1993, Hibbitt Karlsson & Sorencen, lhc, Pawtucket Rhode Island, 1993 Zienkiewicz, O.C, "The Finite Element Method', McGraw-Hill, London,
1977
[Zien77]
[ZP074]
Zienkiewicz, O.C, Phillips, D.V., Owen, D.R.J., "Finite element analysis of some concrete non-nearities, Theory and examples", Concrete structures subjected to triaxial stresses, Seminar, 17.-19. May, 1974, Bergamo, IABSE, Zrich, 1974
Yanes, L. A., "A Computer Program for Nonlinear Analysis of Rein
forced Concrete Stractures 1982
[Yane82]
of Civil
Abbreviations: ACI:
ASCE: EERC: IABSE: RILEM: SIA: UCB:
WCEE:
American Concrete Institute American Society of Civ Engineers Earthquake Engineering Research Center (University of California) International Association for Bridge and Strctural Engineering Reunion Int des Labs. d'Essais et de Rech, sur les Materiaux et les Constr. Schweizerischer Ingenieur- und Architekten-Verein University of Cafornia, Berkeley World Conference on Earthquake Engineering
199
NOTATION
Greek
Upper Case
O0
&
Hexural
overstrength factor
&y
ct ct.
area,
ctcl
are
closing
ctcr
ct*mm
ct.
spring
cross
section
ay
ay
spring
cross
Ratio of global flexural stiffness reduetion of yielded Shear retention factor, factor in time Axial
section
8a 8h 8he
integration
Horizontal Shear
8S
8SC 8V
8V i <5vr
Shear
Shear
displacement
displacement at shear cracking
displacement
displacement at left element edge
displacement at right element edge
8y
e
displacement
at zero stress
Strain
Eo
Eots
zero stress
in tension
stiffening
model
Esou
50h
strength
strength
200
NOTATION
cr
/
Concreto
Ely
en r
s
Steel strain
Yield strain
Ey
h
r
Yr
K
integration
A
Aq
\~A
H
f-e
cov
Curvature
Rotational
p Px
Py
a
y-direction
Stress
Axial tress
O"0
of
af
Ps
Fibre
concrete stress
stress
Fibre steel
Confinement ratio
Tense reinforcement ratio
Pt
Ots
stiffening stress)
Pw
pWh
Latin
Upper Case
Ac
Concrete
area
area
Ag
Af
NOTATION
201
Af
Fibre steel
area
As
A si
inclined surface
Av
area
area
Aw
sectional sectional
AWi
area
inclined at 45
area
degrees
Aws
B
Horizontal reinforcement
of web
Strain-displacement matrix
Global
damping matrix,
transformation matrix
Cd
Ct
D
Dc
Dg
Dia
Di
Young's
modulus
Ec
Ei
Ek
Dissipated energy
Kinetic energy Elastic strain energy
Inelastic
Ese ESie
Es,ts
Esx
Increased
Young's
stiffening
ESy
Et Ets
Ev
F
F
Young's modulus for reinforcement steel in y-direction Young's modulus for concrete between cracks in opening phase
Viscous energy
Force
Fc Fcs
Ff
Cracking force of outer vertical spring Spring force of central vertical spring
Fibre concrete force Fibre steel force Residual force
Ff Fr
202
NOTATION
Fs
Spring
Fts
Steel force
Fy
G
Yielding
Modulus of rigidity
Gf
H
Fractore energy
Buding height
WaU
Hw
/
height
section
laIe
cross
section
Iy
K
Moment of inertia about strong axis Global stiffness matrix Stiffness of left vertical outer spring
Ki
K2 K3
Ka Kc Kcs Kt
spring
of outer vertical
Elastic
compressive stiffness
spring
Elastic
Ks Kse
Ku Kve
Horizontal
Elastic
spring stiffness
spring
Unloading stiffness
of outer vertical
spring
(shear) spring
Ky
L
spring
area
of integration
Le Lec
strength
Lp
Ly,
M M
Bending
Global
mass
matrix
ME
Mi
Mr
My
M
MN
including
normal force
NOTATION
203
N P
Normal force
Shear retention coefficient
Effective
Pu
design
normal force
on
Pv
Viscous effect
Q
Re
Transformation matrix
Rm
with
zero
normal force
Rp
T{
U
Global
Global
displacement vector
velocity vector
vector
Global acceleration
Vc VE
VR Vs Vu
Vw
Z
equivalent force
strength
Softening
ag(t)
be
Ground acceleration
history hoops
bh
bw
c
Static
height
fc
Concrete stress
fc'
fd
fs
fi
Eigenfrequency of mode i
Inertia force
fi
204
NOTATION
fs
fs
ft
fts fu
fv
stiffening
strength of steel
strength of concrete
fwy
fy
h
hc
k
Upper distance
to centre
of relative rotation
ke kib
Kie
kts
l
Concrete tension
softening factor
lc
Centroidal distance
Web
h
m
length
ms
n
Number of storeys
WaU aspect ratio
ra
s
Spacing
of reinforcement bars
Sh
t
u u
Ui
V
Degree
of freedom No. i
Vc
w
Concrete shear
stress
WQ
tense stress
Xi
205
APPENDIX
USER ELEMENT
INPUT DESCRIPTION
The
macro
model
was
coded
as a
"User Element"
[Abaq91],
in the
same manner as
the
library elements.
of this
input properties,
weU
as some
useful Output
quantities
user
in this
appendix.
are
specific
for the
user
example of
complete input
file is
given
at the end of
Appendix
A. For the
manual
[Abaq91].
Function
The
user
element is coded
as a
subroutine
on a
according
an
chapter three.
In
performs
the
These
are
element subroutine
fixed formal
by
the
from
increment to another,
vector
of Solution
dependent
changed by the
subroutine
as
to
during execution,
and this is
made by
The
user
element has to be
speeified just
like the
brary
elements. The
corner
simplest
only.
user
element type
describing
the
macro
nodes
It is
input
speeifications, two
206
APPENDIX A
VARIABLES=45
1,2
3, 1, 2, 6
Except what was stated above, the first line specifies that two coordinates are i^uired
per node, the number of
input properties
equal
to
is
equal
to
number of solution-
dependent
State variables is
prescribe the
d.o.f. numbers
are
available for the nodes; for the first node the first and second
degrees of freedom
same
degree
of
freedom is also available. It should here be noted that the coordinates of the nodes
represent the location of the end points of the flexural Springs of the
macro
model,
see
chapter three.
Input properties
The
in the ne of element
specification
are
given
on a
described
as
1) Cross sectional
area
of entire wa section
(m2)
2) Moment of inertia (about strong axis) of entire wa section (m4) 3) Young's modulus of uncracked
concrete
(MPa)
to
ratio of com
pressive to
0.03)
tense stiffness,
(useful
rnge: 0.4 to
0.8)
5) Yielding factor, equal to ratio of yielded to compressive stiffness (useful rnge: 0.01
to
6) Bending
moment at flexural
yielding of cross
section with
zero
calculation
procedure
see
7) Shear force at the onset of shear cracking, for calculation 8) Cracking factor in shear, equal
calculation
see
to
A wa element with
a cross
area
of 1.8
m2,
a moment
of inertia of 5.4
m4,
Young's modulus of 33000 MPa, cracking and yielding factors in flexure of 0.7 and
0.03
respectively,
yield
moment
no
an
uncracked
shear force
capacity
of 2.8 MN and
207
thus
the
require
the
foUowing
user
as
follows
on
two lines on
element
nodes, i.e. the element geometry, essentiaUy all other relevant element properties
Output quantities
For
plotting
purposes,
number of Output
quantities
are
element. These
written
on an
element file
see
[Abaq91],
the
foUowing two
FREQUENCY=2
SDV
which
assumes
are
"ELOUT"
(default: all elements), and that information is desired for every second increment).
and these wl in the
case
"post file",
see
of a
are
user
specific
State
input
sequence
on a
plot
ofthe
time wl here be
given
on
three lines:
speeifies,
scaling
default
here), then
plot,
208
APPENDIX A
(El. No.
one
integration point
see
number
(always
one
for
user
elements).
options,
[Abaq91].
user
element type 30
are
sted in
foUowing:
SDV22
SDV23 Base shear
(MN)
height (m)
SDV28
SDV29 SDV30 SDV34
height
SDV35
Example
of
Input File
understanding
a
of the
use
of the
macro
model, implemented by
is
of a
user
element,
smaU
presented
so as to create an
q c
w///////////r+
V///////////7A
6.00
i
Model with element and node numbers
two user
Wall prototype
elements
oftype
U30
The
are
area:
1.8
m2,
cross
m4, Young's
MPa, flexural cracking factor: 0.7, flexural yielding factor: 0.03, bending
axial load: 6.5 MNm, shear force at
gravity
applied
to
as two
209
apped
to the
wall from the left at the top. The central distance between the two flexural
spring
is
chosen
as
5.5
m.
are set as
0.05
At first, the
the structural
analysis
is
given
same
die second
versus
top
displacement,
*HEADING
versus curvature
6, 5.5, 8.
*USER ELEMENT, TYPE=U30,
VARIABLES=45
1,2
3, 1, 2, 6
*UEL PROPERTY, ELSET=WALL
1.8, 5.4, 33000., 0.7, 0.03, 6.5, 2.8, 0.16 ?ELEMENT, TYPE=U30, ELSET=W ALL
1, 1, 2, 3, 4
2, 3, 4, 5, 6
*BOUNDARY
5, 2, -0.5 6, 2, -0.5
*END STEP
210
APPENDDC A
*STEP, INC=20
0.05, 1.0
*CLOAD
5, 1, 1.0
*NODEFILE
U
*EL FILE,
FREQUENCY=1
SDV
*END STEP
Post fe for
use
input fe:
*POSTFTLE
VARIABLE-VARIABLE, BSTEP=2
DISPL (m)
SHEAR FORCE (MN)
,
Ul,, TOP-DISPL, 6, 1,
(MNm)
211
APPENDIX
developed
well
in
as a
"User Material"
[Abaq91],
essentiaUy in
as
the
input properties,
as some
useful Output
quantities of this
material
are
specific
for the is
user
example of
given
at the end of
Appendix
B. For the
Function
The
user
material is coded
as a
subroutine in
to the
discussion in
chapter
update
of the
material modulus
via
a user
properties
resulting stresses.
a
These
are
fixed format
[Abaq91].
are
integration point
subroutine
by the program.
another,
vector of
solution-dependent
changed
material
by
the subroutine and which is stored until later increments. Since the
as a
subroutine is written
to be
during execution,
The
user
material has to be
materials. The
user
material is to
as
die CPS4,
the CPS8,
way
as
see
the
Abaqus
manual
[Abaq91].
The
user
same
for
brary materials.
material is
speeified
as
foows:
212
APPENDIX B
foUowed
by
in the section
Input
properties. The abovementioned solution-dependent state variables which are 23 for the
simplest user material model are then speeified in the input file as foows:
DEPVAR 23
Input properties
The
as
specification are
given
are
given
are
described
as
concrete to
0.20)
end of tension
5) E-modulus for reinforcement steel (MPa) (useful rnge: 180000 to 220000) 6) Yield strength of steel, without regard to tension stiffening (MPa) (useful rnge: 300 to600)
7) Reinforcement ratio in horizontal direction (useful rnge: 0.002
to
0.05)
A material with
stiffening
200000 MPa, reinforcement ratios of 0.01 and 0.02 in the horizontal and vertical
directions
as
user
foUows
ne
on
the
input fe:
213
Output quantities
For
plotting
purposes,
State
number of Output
quantities
are
solution-dependent
behaviour of an element
These
are
written
on an
element fe
each
see
included:
EL FTLE, ELSET=ELOUT,
FREQUENCY=2
SDV
which
assumes
are
"ELOUT" increment
(default:
(default:
every
increment).
In
"post fe",
see
may be
given
case
are
nn
specific
State variable
to
brief
example, given
the
input
sequence
on a
vs.
time wl here be
on
three lines:
HISTORY
TTME
BASE SHEAR
State variable
a
number,
set to
here), then
plot,
(El. No.
see
in this
case) and integration point number. For closer details and options,
[Abaq91].
material
user
are
finally sted
foUowing:
214
APPENDIX B
SDV6
SDV7
Accumulated concrete strain transverse to first crack direction Accumulated shear strain upon
cracking
Accumulated concrete stress normal to first crack direction Accumulated concrete stress transverse to first crack direction Accumulated shear stress upon
cracking
SDV15
Accumulated horizontal steel strain Accumulated stress for horizontal steel Accumulated vertical steel strain Accumulated stress for vertical steel
SDV16
SDV18 SDV19
Example
of
Input File
understanding ofthe
a
use
ofthe
macro
model, implemented by
of a
user
element,
small
example of a structural wa
modeUed
by two elements
so as
will here be
create an
presented.
deberately chosen
seen
to
in
figure B.1.
10
11
12
5
7
o
6
8
9
3
4
4
5 6
cd
1
1
2
2
V///////Y/Y/.
600
I
Model with element and node numbers
Wall prototype
Figure B.1 Structural wall modelled by six library membrane elements oftype CPS4,
with material behaviour speeified according to the USER MATERIAL
option
(uniform),
strain of
The
following
basic wa
properties
are
Young's modulus of
0.0001, strain
at
concrete:
cracking
end of tension
stiffening
215
reinforcement ratios of 0.01 and 0.02 in the horizontal and vertical directions
respectively. Furthermore,
for forces.
are
set to 0.05 MN
The
input file
analysis is given
post fe is
of die
same
displacements
the base shear, and the stress normal to the first crack
HEADING
6, 6., 2.67
7, 0., 5.33
10, 0., 8.
11,3.8. 12, 6., 8. ELEMENT, TYPE=CPS4
1, 1, 2, 5,4
2, 2, 3, 6, 5 3, 4, 5, 8, 7
4, 5, 6, 9, 8 5,7,8, 11, 10
6, 8, 9, 12, 11
SOLID SECTION, MATERIAL=WALL
0.3
MATERIAL, NAME=WALL
USER MATERIAL, CONSTANTS=8
216
APPENDIX B
DEPVAR
23
BOUNDARY
1, ENCASTRE
10, 2, -0.5
12, 2, -0.5
END STEP
STEP, INC=20
10, 1, 1.0
NODE FILE
U, CF
EL FILE, SDV
FREQUENCY=1
END STEP
use
resulting
input fe:
POSTFILE
VARIABLE-VARIABLE, BSTEP=2
N STRAIN
N STRESS
217
APPENDIX
yield
chapter three,
storey
on
Appendix, the
selection of die
yield
test
wl be demonstrated, foowed
by comments
the
was
cross
section, according
to
the
procedure apply
waU there are,
certain
height of the
figure
cross
a
specimen
for the
used in
chapter five
a
is
given.
The test
specimen,
was
which is
ten
storey buding,
presented
5.4.1,
reliability
as
we
as
in
subsequent
model.
as
figure
displayed.
cracking
mainly representative
clearly
be
for the of
crack. The
seen at a moment
the the
over
the
cross
section
be carried
by
yielding behaviour
curve
is
seen
to
develop gradually,
yielding
is from the
estimated to
curve
occur at
gradual yielding
gives
rounded
shape
of die
more
straight,
which
means
hardening
alone
hysteretic model
as
model
assume a
trilinear
skeleton curve,
shown in
a
chapter three.
This
In
compressive prevail
until
branch
yielding
entire
yielding
a
yielding of the
used in
cross
section in
representative
This
means
that
yield
figure 6.1,
which may be
trilinear model.
218
APPENDIX C
5.0
4.0
0.000
0.002
0.004 Curvature
0.006
0.008
(rad)
1:3 scale test specimen
Figure Cl
Selection
5.0
-ir-
Behaviour in crack
Elastic behaviour Selected model behaviour-
i.
0.000
0.002
0.004
0.006
0.008
Curvature
(rad)
In
yield level
was
found to be
representative,
as a
cross
figure by
yielded
to
uncracked vertical
spring stiffness
few percent
as
219
discussed in
chapter
specimen
model
one
percent
was
found to
The model behaviour with the fuU axial load of 0.87 MN,
the test
specimen,
is
seen
in
figure C.2.
The
approximation ofthe
model behaviour
cross
according to
expression
sectional behaviour.
figures
C. 1 and C.2
soon as
to the skeleton
curve
of a the
hysteretic model
of these
of chapter three. As
are no
figures
longer relevant,
nonnear
of
chapter five,
the
procedure of determining
pattem
cross as
yield
moment wi not be
same
for the
ductity design
were
seen.
on
From
respectively Increasing
read, based
the
curve
design
strength.
gives
7.9
these values
by
values
respectively.
6.4 and 6.6 for the
curves
Inspection
of
figures
of
yield strength
hardening
compressive strength
suggests good agreement, and would have given about 7.9 and 4.6 MNm
as
region.
220
221
APPENDIX
presented
in this
Appendix.
belong
Abaqus
element
brary, [Abaq91].
The
presentation
to
be used in
used
theoreticaUy also be
case
of three-dimensional
analysis.
CPS3
CPS6
CPS4
CPS4R
CPS8
CPS8R
Figure D.1
222
APPENDIX D
In the
figure the
denote
dots denote
in-plane degrees
crosses
integration
see
points.
In the element
R denotes reduced
integration,
e.g.
[Abaq89].
For structural walls the
triangulr elements
case
of
irregularities
or
elements will be used, and for the nonlinear models elements CPS4 and CPS8
developed
proved
to be
preferable.
integration
in
problems
they
analysis.
three-storey
1:3 scale structural wa test
comparison
was
made between die elements CPS4 and CPS8 for die nonlinear static
specimen presented
were
chapter three
presented
in
chapter five,
used for
same
this
comparison.
The
number of elements used in mesh A with CPS4 (left) and with CPS8
same
are
A with CPS8
(right).
>l
**>
--;
S?
*# *
%
""' ^
*
<.
'
<
Figure
D.2 Meshes
for
element
comparison: Mesh
A with CPS4
(left),
mesh B with
(right)
Figure
comparison. The
It is
seen
base shear is
that Solution
described in
chapter four,
suggested by
reinforcement Different
some
integration
certain element is
authors.
223
0.01
0.02
0.03
0.04
displacement [m]
Figure D.3
Base shear
versus
224
225
APPENDIX
In order to
behaviour of the
used in the
the nonnear
dynamic
behaviour in
few
were
performed.
given
possibility
of
Springs of
chosen
damaged Springs
damage
were
completely elastic
elements have
an
State, simulates
macro
elastic stiffness.
by letting
spring of each
acr of 0.7
was
element
develop
case a
cracking factor
used,
as
described in
chapter three.
yielding in equal
the three
plastic hinge
figure 5.20).
This is achieved
by setting the
to
Springs
plastic hinge
as
elements
yielded
stiffness. The
a
yield
taken
intended to simulate
damaged
by
all
letting the
(i.e. close
central vertical
to
spring
ofthe
horizontal modes,
given
on
die
following page.
obtained
was
by prescribing
which starts
high
tense
strength
and
high yield
limit. Since
cracking
is
procedure
gradually
and needs
but-up
properly
it
was not
found feasible to
same
appes
eight
yielded State.
by
means
for the
lowest modes
state obtained
to the elastic
by the macro model. The comparison of the two elastic damaged states, simulated by the
that the
macro
states is shown in
model,
are
From table E. 1 it is
seen
are
in
fairly good
model. lowest
agreement with both models. The first mode is about 6 % lower for the
Somewhat refined meshes for the micro model devered about the
macro
same
frequencies
as
for the
macro
model.
226
APPENDDC E
Mode Number
1
2
0.79
4.25
10.4 17.4
4.60
11.3 18.4
3
4
5 6
7 8
24.3
29.9 34.2
37.0
23.3
24.5
29.1
32.3
Table E.l
Mode Number
1
2
Major Damage
0.19 2.97 7.16 13.9 21.5 28.3 33.5
3
4
16.7 23.5
5 6
7 8
29.3
33.6
29.3 33.9
36.5
36.9
36.8
Table E.2
macro
model for
eight storey
wall
of
damaged
states, obtained
the
tendency
may be
clearly
are
frequencies
were
while ie
unaffected. It should be
kept
achieved by
certain
constantiy changing.
The
damage
State
behaviour
suggested here
be
is therefore
only
simpfication.
oceurring e.g.
The
"major damage"
a
appcable
to any free
Vibration
after
alternatingly
yielding
bound
in
on
describe
unloading behaviour
complement
with
loading. Therefore,
as a
hypothetical
lower
"free Vibration",
behaviour. It
FREQUENCY STUDY
should be noted that
OF DAMAGED STATES
227
during
time
history analysis,
as
already
As
a
at the
comparison
by the
the behaviour
In
displayed
in the
frequency region.
three different
figures E.l
of the
eight storey
plots represent
time segments.
which is
Figure
yielding
behaviour.
seen
shows
suggestions
damaged
figure
frequencies
clearly
seen
in be
0.7 Hz, 4.3 Hz, and 9.5 Hz. These values should
essentially
comparable
seen
that this is
largely the
case.
figure
frequencies essentially
as:
0.5 Hz,
seem
8 to 12 seconds,
have
damaged
in
system in free Vibration. The first mode dominates the damping out ofthe vibrations. It is
here
seen
point
at which the
system
was
Figure
E.l
to 4
seconds,
228
APPENDIX E
Figure
E.2
Figure
E.3
damaged
frequencies
as
These values
"moderate
yielding"
and
"major damage"
yielding State.
states of
only
major damage
State should
as a
229
regarded
moderate
as
being capable
some
of
withstanding
the
ground
motion
suggested
here with
local
It should be
input
assumptions of modelling.
230
231
APPENDIX
DESIGN
DEFINITIONS
In the
presentation of chapters
such
number of definitions
are
mainly related to
the
design references,
as
[PBM90].
comprehensive
definitions could in
summary
is
provided
in this
Appendix.
given as
listing,
where
expressions
and
are
stated
briefly.
essentially
mited to the
subject of
area
of
Equivalent
The
according by
earthquake action
may be modeUed
an
Section 4 19 of
modelling
explanations
are
given below.
forces
frequency
for
buildings
earthquake
are
carried
by
cores or as
diagonal bracing,
according
to Clause 419
foows
41
h
where
/0
:
:
fundamental
Cs
ground coefficient:
:
:
Cs Cs
dimension in
of the
consideration
h
:
height in
of the
buding
232
APPENDIX F
The static
equivalent
force
according to
as
follows
Qacc^Cfa+^VaccQr)
o
where
Qm^m
ah
total horizontal
report)
as
Clause 4 19 64
g
:
:
acceleration due to
gravity
Ck
construction coefficient in accordance with Clause 419 71 which takes into account
tions and the
a
plastic deforma
use
Ck=Cd
*
where
deformation coefficient
account the
according
to
According to the
on
SIA
structural classes
a
(Clause
4 19
[PBM90] allows
replacement
a as
ductity:
pa
to
3, fu ductility: pa
5.
Cd
Cd
0.65.
be
seen as an
design coefficient
can
overstrength
reduetion
factor
as
DESIGN DEFINITIONS
233
Cd= t
where
=0.65
1.25 1.2 1.5
1-5
=
fo
l0yR
Gm
: mean
value of the
X VaccQr
'sum
Qr
over
the
height of the
structure
19 508
=n
K+v^&),A
Y\G*+yLVaccQr).h
where
Qm.ci
:
'
at
height h,
{Gm + ^ YaccOr).
ht
discussed in
ME
^.Cil
distributed
of moment
ME
is the redistributed
moment)
234
APPENDIX F
MIMC
v
M0
X0MR
0>o,JdE
YrMe
--ME
Figure F.l
Levels
offlexural strength
Nr.
on
Calculated
tributary to
the waU.
shear force
nents
Qm^i
die
Cross sectional
Yr
resistance factor
accounts
-
as
for:
compared to calculated
M:
as
yRME
NG
(note:
not
yR NG)
MB
strength,
calculated with
yRNG.).
equal to,
or
DESIGN DEFINITIONS
235
M0
flexural
overstrength,
is the
as
calculated
as
XBMR
or as
^>0BIME
where
X0
calculated
usually obtained
1.2
or more
o,*
Va
'
calculated
as
coJbotttVE
(ov
buildings
up to six stories
cov
a),
'
0.9
buildings over 6
six stories
=1.3
<
1.8
30
VR
shear
asVR Vc
Vc
V.
concrete
as
Vc
v.bjd
given
as
where
Pu
acting
on
the
cross
section and
as
Ag
bw
0.8
times the wa
length Lw.
Vs
strength, expressed
as
V.
\fy
A,
is the
area
of shear reinforcement
is the vertical
Material
strength
:
design value (nominal value) for steel yield strength (design strength) overstrength value for steel yield strength, calculated
fo
as
Xjy
236
APPENDDC F
fs'fy k
fo-,
f^f
-
Kf
'oJy
J m-y
TmJy
(U
Figure F.2 Levels of material strength
X0
f-y
Jy
usuaUy obtained
as
1.2
or more
e(f0 y)
/
steel strain at
m.y
effective
mean
strength
material
Jm,
as
Jm,y
nt
Jy
usuaUy obtained
as
1.2
fcw.min
'
(3 23)
fc
0.65 fcw,nUn