Applied Sciences: Hysteretic Behavior of Steel Reinforced Concrete Columns Based On Damage Analysis

applied
sciences
Article
Hysteretic Behavior of Steel Reinforced Concrete
Columns Based on Damage Analysis
Bin Wang 1, *, Guang Huo 1 , Yongfeng Sun 1 and Shansuo Zheng 2
1 School of Civil and Architecture Engineering, Xi’an Technological University, Xi’an 710021, China;
huoguang@163.com (G.H.); sunyongfeng@163.com (Y.S.)
2 School of Civil Engineering, Xi’an University of Architecture and Technology, Xi’an 710055, China;
zhengshansuo@263.net
* Correspondence: wangbin2346@xatu.edu.cn; Tel.: +86-135-1912-1853

Received: 19 January 2019; Accepted: 14 February 2019; Published: 17 February 2019
Abstract: With the aim to model the seismic behavior of steel reinforced concrete (SRC) frame
columns, in this research, hysteresis and skeleton curves were obtained based on the damage test
results of SRC frame columns under low cyclic repeat loading and the hysteretic behavior of the frame
columns was further analyzed. Then, the skeleton curve and hysteresis loops were further simplified.
The simplified skeleton curve model was obtained through the corresponding feature points obtained
by mechanical and regression analysis. The nonlinear combination seismic damage index, which was
developed by the test results and can well reflect the effect of the loading path and the number of
loading cycle of SRC frame columns, was used to establish the cyclic degradation index. The strength
and stiffness degradation rule of the SRC frame columns was analyzed further by considering the
effect of the accumulated damage caused by an earthquake. Finally, the hysteresis model of the SRC
frame columns was established, and the specific hysteresis rules were given. The validity of the
developed hysteresis model was verified by e comparison between the calculated results and the test
results. The results showed that the model could describe the hysteresis characteristics of the SRC
frame columns under cyclic loading and provide guidance for the elastoplastic time-history analysis
of these structures.
Keywords: steel reinforced concrete; frame columns; hysteretic behavior; damage index; hysteresis
model
1. Introduction
Due to their high bearing capacity and good seismic performance such as energy dissipation and
ductility, steel reinforcement concrete (SRC) members, especially frame columns, are widely used in
high-rise and super high-rise hybrid structures [1–4]. To ensure that the seismic performance of the
structure meets the design requirements, the engineering designer must apply a strict elastoplastic
calculation analysis of the structures. As the basis of elastoplastic calculation analysis, the hysteresis
model that can accurately reflect the strength and stiffness degradation, deformation performance,
and energy dissipation capacity of structures and members under cyclic repeated loading has been
studied [5,6]. Generally, the existing hysteresis model includes two parts, namely, the skeleton curve
and the hysteresis rule. The former is determined by all the hysteresis feature points and the latter
reflects the highly nonlinear character of the structures [7]. For the structural seismic analysis, the
existing hysteresis model can be divided into two categories: the polyline hysteresis (PH) model and
the smooth hysteresis (SH) model. The PH model consists of piecewise lines obtained by simplifying
the hysteresis behavior of the structure members, and is relatively easy to express and apply [8–11].
The SH model needs more computational burden to obtain more accurate results and is consequently
Appl. Sci. 2019, 9, 687; doi:10.3390/app9040687 www.mdpi.com/journal/applsci

Appl. Sci. 2019, 9, 687 2 of 19
relatively difficult to implement [12,13]. Therefore, in this paper, the polyline hysteresis model was
adopted for seismic analysis of SRC members.
Meanwhile, in most of the existing PH models, the strength and stiffness degradation rule of
structure members under an earthquake is described by inducing the cyclic degradation index that
only considers energy dissipation [14–17]. However, an earthquake is a random load, and the loading
path has a significant effect on the ability of seismic energy consumption of the members. Therefore, it
is difficult to obtain a uniform expression of the energy dissipation capacity of the members. Compared
to methods mentioned above for establishing the hysteresis model, the hysteretic model of structural
members is established by introducing a damage index (DI) that can better describe the stiffness and
strength degradation rule [18,19].
At present, there are several damage analysis models for structure members. There are two main
approaches to establish the damage index: one is sensor based and the other one is hysteresis based.
The sensor-based DI often employs distributed sensors that are embedded or surface bonded to the
structure to capture the change of certain physical parameters, which can reflect the severity of damages
to the structure [20–24]. A commonly employed sensor is the Lead Zirconate Titanate (PZT) transducer
in structural damage detection. PZT is a type of piezoceramic material with very strong piezoelectric
effect that enables the PZT transducer to have dual actuation and sensing capacity. In addition,
PZT transducers have wide bandwidth and high sensitivity, and are widely used to generate and
detect stress waves [25–28]. With the active sensing approach, a pair of PZT sensors and an actuator
located along a stress wave path can monitor the damage on the path and a damage index can be
established to reflect the severity of the damage. The active-sensing based DI has been used in concrete
structures [29–31], SRC structures [32–34], and concrete filled tubular structures [35–37]. Damage
indices based on impedance have also been reported [38–40]. The advantage of the sensor-based
DI is that it is suitable for real time monitoring of the structure via sensors integrated with the
structure [41–43], however, this index does not directly relate to an inherent structural parameter.
The other approach to establish a damage index is based on hysteresis. The hysteretic curves
of structure members reflect the changes of strength, stiffness, energy dissipation capacity, and
displacement ductility with the loading displacement and the number of cycles. Therefore, the rule of
strength and stiffness degradation of the members can be described by the change of the hysteresis
loop under an earthquake. Early on, researchers proposed non-cumulative damage indexes based
on a single parameter, such as stiffness, strength, and hysteretic energy dissipation [44–47]. Since an
earthquake is a repeated loading process, the damage caused by an earthquake is cumulative. To better
describe the damage of structures under cycle load, the cumulative seismic damage indexes, which
consist of the combination of plastic deformation and cumulative hysteretic energy dissipation, were
developed and are widely used to analyze structural damage [48–50]. Though the hysteresis-based DI
involves a destructive test and cannot be used for real time monitoring of in-service structure, this
damage index is directly link to inherent structural parameters, such as stiffness. In this research, we
use the hysteresis-based DI to analyze the hysteresis rule of SRC columns.
The existing research results show that the cyclic loading history and the number of cycles are the
main influencing factors for the damage analysis of structure members. However, the damage index
selected in the existing hysteresis model cannot well consider the influence of the number of loading
cycles and loading path on the ultimate energy dissipation and deformation capacity of the members.
Meanwhile, to the best of knowledge of the authors, this method has been applied only to the analysis
of RC structures and steel structures. Few researches on the hysteretic behavior of SRC members have
been reported. Since the SRC composite members consist of steel and RC, their damage process is
more complex than pure steel or reinforced concrete structures [51,52]. Therefore, the current method
of establishing hysteresis model based DI cannot accurately reflect the seismic performance of the
SRC members.
In this paper, a new PH hysteretic model by considering the accumulated damage is proposed
for the seismic analysis of SRC structures. Firstly, the experimental studies on SRC frame columns
Appl. Sci. 2019, 9, 687 3 of 19
were conducted to study the hysteresis behavior under low cycle repeated loading. Based on the test
results, the hysteresis behavior was analyzed, and the skeleton curve and hysteresis loops were further
simplified. The simplified trilinear skeleton curve model was obtained through the corresponding
feature points determined by regression analysis. Secondly, the nonlinear combination seismic damage
index, which was developed by the lots test results and can well reflect the effect of the loading path
and the number of loading cycles of SRC frame columns, was used to establish the cyclic degradation
index. The strength and stiffness degradation rule of the SRC frame columns was analyzed further by
considering the effect of accumulated damage caused by an earthquake. Finally, a hysteretic model of
SRC frame columns was established and the specific hysteresis rules given. The proposed hysteretic
model was validated by comparing the experimental and the numerical results on SRC frame columns.
2. Description of Test Program

In this paper, eight SRC frame column specimens were designed and constructed based on the
current Chinese standards [53,54]. All of the specimens have the same cross-sectional shape and
dimensions, however they have different compression ratios n (=0.2, 0.4, 0.6), stirrups ratios ρv (=0.8%,
1.1%, 1.4%), and steel rations ρs (=4.6%, 5.7%, 6.8%). The configurations of the cross sections are shown
in Figure 1. The details of all specimens are summarized in Table 1.
Table 1. Design parameters of specimens.
Cross Compression Stirrups

Specimen Aspect Ratio Steel Shape
Section (mm) Ratios Type
SRC-1 150 × 210 3.0 0.4 I 14 Φ6@110
SRC-2 150 × 210 3.0 0.4 I 14 Φ6@110
SRC-3 150 × 210 3.0 0.2 I 14 Φ6@110
SRC-4 150 × 210 3.0 0.6 I 14 Φ6@110
SRC-5 150 × 210 3.0 0.4 I 10 Φ6@110
SRC-6 150 × 210 3.0 0.4 I 12 Φ6@110
SRC-7 150 × 210 3.0 0.4 I 14 Φ6@80
SRC-8 150 × 210 3.0 0.4 I 14 Φ8@120
All the steel shape was made of Q235 steel. The concrete strength was C80. The measured concrete
cube average compression strength f cu , axial compression strength f c , and elasticity modulus of
concrete Ec are 83.9MPa, 75.49MPa and 42.042MPa, respectively. The measured mechanical properties
of steel shape, longitudinal rebars, and stirrups are shown in Table 2.
Table 2. Mechanical properties of the steel and rebars.
Steel Shape Stirrups Longitudinal

Properties
Flange Wed Φ6 Φ8 Rebars Φ 10
Yield strength f y (MPa) 319.7 312.4 397.5 354.5 386.3

Tensile strength f u (MPa) 491.5 502.5 438.0 457.3 495.7
Elasticity modulus Es (MPa) 2.07 × 105 2.07 × 105 2.06 × 105
Appl. Sci. 2016, 6, x FOR PEER REVIEW 4 of 20
Appl. Sci.
Appl. Sci. 2019,
2016, 9,
6, 687
x FOR PEER REVIEW 44 of
of 19
20
4 10
110 20 20
4 10
4 10 I 14
150 150 150

210
20 11020
6@80 12I 10
14
210
( 6@110)
6@80 12 10
(( 8@120) I 14
20110
6@110) 210
6@80 20 170
12 10 20
( 8@120)
( 6@110) 20 210
170 20
20
( 8@120) 1-1
210
750 750 750

20 170 20
4 10 6@80 1-1
1050 10501050
210
4 10 1 1 6@80
400 400 1-1
4400
10 1
6 10@100 1 400
2 6@80
300 300 300

10 12
10 12 6 10@100
400 1 1 2
400
300 300 300

10 12 6 10@100 2 10 12
10 12
2 10 12 200
1010 2-2
2 . 200
1010 2-2
2
. 200
1010 2-2
Figure 1. Cross-sectional dimensions and distributed steels of specimens (unit: mm).
.
A schematic
Figureand a photo of thedimensions
1. Cross-sectional test setup andare distributed
shown in Figures
steels of 2specimens
and 3, respectively.
(unit: mm). In order to
A schematic
A
prevent schematic and
sliding ofandthe aaspecimen
photo of
photo of during
the test
the testthe
setup
setup are
are
test, theshown
shown in Figures
in
specimen Figures 22 and
was fixed andon 3,strong
3, respectively.
respectively.
ground Inby
In order
order to
to
using
prevent
prevent sliding
sliding
A schematic
two support of
steel the
ofand
beams specimen
the aspecimen
photo
and two during
of during
thefixed the
testthe
setup
steel test,
test,arethe
the
beams, shownspecimen
specimen
which in havewas
was
Figures fixed
fixed
2 andon
a certain on
3, strong
strong
rigidity ground
ground
respectively.
as by
Inby using
using
order
a constraint to
of
two support
two support
prevent
the frame steel
steel
sliding
column. beams
ofbeams
the and
and two
specimen
Each two
specimen fixed
fixed
during was steel
steel
the beams,
beams,
test,
tested the which
under which
ahave
specimen have
was
constanta certain
a certain
fixed
axial rigidity
rigidity
ongravityas aas
strong a constraint
constraint
ground
load by of
during of
the
using
the frame
frame
two support
experiment column.
column. Each
steel
before Eachand
beams
applyingspecimen
specimen was
atwo fixed
lateral was
tested tested
under
steel
force Pbeams, under
toasimulate
constant
whicha seismic
constant
axial
havegravity axial
a loading.
certain gravity
load during
rigidity
The asload
lateralthe during
experiment
a constraint
force the
P wasof
experiment
before
the frame
applied before
applying
by column. applying
a lateral
Each force a
specimen
an electro-hydraulic lateral
P servo force
to simulate
was P
tested
actuator to simulate
seismicunder
placed at seismic
loading. loading.
The of
a constant
the top lateral
axial The
the gravity
column. lateral
force Pload was force
applied
during
Linear P was
by
the
variable
applied
an
experimentby an
displacement electro-hydraulic
electro-hydraulic
before servo (LVDTs)
applying
transducers actuator servo
a lateral placedactuator
force
were at to placed
P the
adopted topto of
simulate atseismic
the
monitorthe toploading.
column.
the ofLinear
the column.
horizontal variable
The lateralLinear
forcevariable
displacement
displacement P was
of the
displacement
transducers
applied
columns. antransducers
by (LVDTs) (LVDTs)
were adopted
electro-hydraulic towere
servo monitoradopted
actuator the placed to monitor
horizontal the ofhorizontal
displacement
at the top the of thedisplacement
column.columns. of the
Linear variable
columns.
displacement transducers Reaction
(LVDTs) wall were adopted to monitor the horizontal displacement of the
Reaction frame
columns. Reaction wall
Reaction frame
Reaction wall
Reaction frame
Hydraulic jack
MTS actuator
Hydraulic jack
MTS actuator
Fixed steel beam
Hydraulic jack
MTS actuator
Fixed steel beam
Specimen
Support steel beam Fixed steel beam
Specimen
Support steel beam
Specimen Bottom beam
Support steel beam
Bottom beam
Figure 2. Test setup. Bottom beam

Figure 2. Test
Figure 2. Test setup.
setup.
Figure 2. Test setup.
Figure 3. Photo of testing apparatus.

The monotonic loading was applied on the specimen SRC-1, and the remaining specimens were
The monotonic loading was applied3.on the specimen SRC-1, and the remaining specimens were
subjected to one cycle of loading Figure Photo of testing
with displacement apparatus.
ductility µ equal to 0.2, 0.4, 0.6, and 0.8, and
The monotonic
subjected to one cycleloading was applied
of loading on the specimen
with displacement SRC-1,
ductility  and thetoremaining
equal specimens
0.2, 0.4, 0.6, and 0.8, were
and
followed by three successive cycles of loading with the displacement ductility µ equivalent to 1.0, 2.0,
subjected
The
followed to one
monotonic
by three cycle of
loadingloading
was
successiveload with
applied
cycles displacement
on
of loadingthe ductility
specimen
with SRC-1,
the peak  equal
and
displacement the to 0.2,  equivalent
0.4,
remaining
ductility 0.6, and
specimens0.8, and
were
3.0, . . . . until the horizontal dropped to 80% of the load of the specimens or the load to 1.0,
could
followed by
subjected to three successive
one cycle cycles
of loading of loading
with with the
displacement ductility  equal
displacement ductility  equivalent
to 0.2, 0.4, toand
0.6, and 0.8, 1.0,
followed by three successive cycles of loading with the displacement ductility  equivalent to 1.0,
Appl. Sci. 2019, 9, 687 5 of 19
2.0, be
not 3.0,continued
…. until the horizontal
because damageload
wasdropped
apparenttoand
80% ofloading
the the peak load
was of the[55].
stopped specimens or thecyclic
The specific load
could not
loading be continued
history is shownbecause damage
in Figure 4. was apparent and the loading was stopped [55]. The specific
cyclic loading history is shown in Figure 4.
6
Displacement ductility
2
-2
-4
-6
0 2 4 6 8 10 12 14 16
Number of cycle
Figure 4. Loading
Figure 4. Loading history
history of
of specimens.
specimens.
3. Analysis of Hysteretic Behavior
3. Analysis of Hysteretic Behavior
The loading-displacement (P−∆) hysteresis curves of the members under cyclic loading are the
The loading-displacement (P−  ) hysteresis curves of the members under cyclic loading are the
basis of the hysteretic performance for the structural members. Based on the test results, the loading–
basis of the hysteretic
displacement hysteretic curves
performance for the
and skeleton structural
curves members.
for the SRC Based onwhich
frame columns, the test
can results, the
be obtained
loading– displacement hysteretic curves and skeleton curves for the SRC frame
by connecting the peak points of the hysteretic loops, are shown in Figure 5. It can be seen that:
columns, which can
be obtained by connecting the peak points of the hysteretic loops, are shown in Figure 5. It can be
seen that:
1. All of the test members have a similar failure process under repeat cyclic loading that experiences
1. Allthreeoffailure
the test members
stages, namely,have a similar
the elastic stage,failure process
the inelastic stageunder repeat and
with cracks, cyclic
theloading that
failure stage.
experiences
At the initialthree
stage failure stages,several
of loading, namely,horizontal
the elasticcracks
stage, the
appearinelastic
first stage
at thewith
root cracks, and the
of the column.
failure stage.
The stress andAtstrain
the initial
of thestage of loading,
rebars are small, several
and thehorizontal
action ofcracks appear
the steel has first
not at yetthe root of
occurred.
the
Thecolumn.
specimen Theis stress
mainly andin strain of thestage.
the elastic rebarsAt arethis
small,
time, andthethe action ofbehavior
hysteresis the steel of hasthe
notSRC
yet
occurred. The specimen
frame columns is similaris to
mainly in the elastic
the reinforced stage. At
concrete thiscolumns.
frame time, the hysteresis
Therefore,behavior of the
the hysteresis
SRC
loopsframe
havecolumns
differentisdegrees
similar toof the reinforced
pinching concrete frame
phenomenon. Withcolumns. Therefore,
the increase the hysteresis
of the displacement
loops have and
amplitude different degreesofofcycles,
the number pinching phenomenon.
the cracks continuallyWithexpand
the increase of the displacement
and develop. Meanwhile,
amplitude and the number of cycles, the cracks continually expand and
the concrete cover appears severely cracked with partial spalling. At this point, because of the develop. Meanwhile,
the concrete
effect cover
of the steel appears
and severelythe
core concrete, cracked
pinchingwith partial spalling.
phenomenon of theAt this point,
hysteresis loopsbecause of the
is gradually
effect of the
improved. steelthe
When and core concrete,
displacement the pinching
ductility µ reachesphenomenon
3, some of theof the hysteresis
horizontal loops is
cracks develop
gradually
into shear improved. Whenbut
diagonal cracks, thethedisplacement speed of 
developmentductility thereaches
diagonal3, cracks
some of the horizontal
is relatively slow.
The pinching
cracks develop phenomenon of the hysteresis
into shear diagonal loops
cracks, but thedisappears
development basically.
speedAtofthis
thetime, the hysteresis
diagonal cracks is
loop becomes a plump fusiform shape. With continuous increase
relatively slow. The pinching phenomenon of the hysteresis loops disappears basically. At in load, the longitudinal rebars,
this
steel the
time, flange, and most
hysteresis loopofbecomes
the steelawebplump yield at the crack
fusiform shape.section. Meanwhile,
With continuous due tointhe
increase action
load, the
of the repeated
longitudinal load,steel
rebars, the flange,
horizontal
and crack
most ofgradually
the steel runs through.
web yield at the The whole
crack process
section. is in the
Meanwhile,
due to thestage
inelastic actionwithof the repeated
cracks. load,
Finally, thethe
with horizontal
increase crack
of load gradually
further, therunsspecimen
through. comesThe wholeinto
the failure
process is stage.
in theThe concrete
inelastic cover
stage of the
with compression
cracks. zone at
Finally, with thetheincrease
root of the column
of load is a large
further, the
spalling area.
specimen comes The stirrups
into and stage.
the failure longitudinal rebarscover
The concrete are exposed, and some ofzone
of the compression the atlongitudinal
the root of
rebars
the buck.isHowever,
column since the
a large spalling core
area. The concrete
stirrupsisand
constrained
longitudinalby arebars
steel flange frame, the
are exposed, andability
some
of
of the longitudinal
compression rebars buck.
deformation of However,
the SRC framesince column
the coreis concrete
obviously is constrained
enhanced. by a steel flange
Meanwhile, the
frame, the ability of compression deformation of the SRC frame column is obviously enhanced.
strength and stiffness degradation of the members are slow, due to the constraint of the transverse
stirrups on the
Meanwhile, the concrete,
strength and there is a lateral
stiffness bracing on
degradation the members
of the outside ofare theslow,
steel,due
which
to thecan prevent
constraint
thethe
of global and local
transverse buckling
stirrups of the
on the steel. The
concrete, therewhole process
is a lateral of the hysteresis
bracing cycleof
on the outside shows that
the steel,
which
there iscan prevent
a good the global
ductility and an and local of
ability buckling
energy of the steel. The
consumption of whole
the SRC process
frameof the hysteresis
column.
2. cycle showstothat
Compared the there is a good
reinforced ductility
concrete and an
members, theability
initialofstiffness
energy of consumption
the skeletonof the SRC
curve of theframe
SRC
column.
frame column is larger due to the steel, and the declining part of the skeleton curve is relatively
2. Compared
flat. At thetosame the reinforced
time, the peak concrete
loadmembers, the initial
of the positive loadingstiffness
skeletonof the skeleton
curve curvehigher
is slightly of the
SRC
than frame
the one column
of the is largerloading
reverse due to skeleton
the steel,curve,
and the declining
since part of has
the specimen the askeleton curve is
certain residual
relatively flat. At the same time, the peak load of the positive loading skeleton curve is slightly
higher than the one of the reverse loading skeleton curve, since the specimen has a certain
Appl. Sci. 2019, 9, 687 6 of 19

deformation after positive loading. Under reverse loading, it is necessary to offset the influence
ofresidual
residualdeformation
deformationafter positive
in the loading.
members. Under reverse
In addition, positiveloading,
loadingitgenerates
is necessary to offset
a certain the
damage
to the members, which reduces the bearing capacity of the members under repeated loading.a
influence of residual deformation in the members. In addition, positive loading generates
3. certain
With damage of
the increase to the
theaxial
members, which reduces
compression ratio, thethe bearing
initial capacity
stiffness of the
and peak members
load under
of the specimen
repeated loading.
gradually increase. However, after reaching the peak load, the larger the axial compression ratio,
3. the
With thethe
faster increase of the
specimen axial compression
strength and stiffnessratio, the initial
degrade. stiffness
The energy and peakcapacity
dissipation load of ofthe
the
specimen gradually increase. However, after reaching the peak load, the larger
specimens significantly reduces, and the ductility becomes poor. With the increase of steel ratios, the axial
compression
the peak load ofratio,
the the faster the
specimen specimen
gradually strengthAfter
increases. and stiffness
reaching degrade.
the peak The energy
load, withdissipation
the increase
capacity of the specimens significantly reduces, and the ductility becomes poor. With the
of the steel ratios, the strength degradation of members is relatively slow, and the ability of the
increase of steel ratios, the peak load of the specimen gradually increases. After reaching the
ultimate deformation is strong. The stirrups ratio has little effect on the peak load by comparison
peak load, with the increase of the steel ratios, the strength degradation of members is relatively
of different stirrups ratios but the strength degradation of the specimen is relatively flatter with
slow, and the ability of the ultimate deformation is strong. The stirrups ratio has little effect on
the increase of the stirrups ratios and the energy dissipation capacity and the ductility of members
the peak load by comparison of different stirrups ratios but the strength degradation of the
are enhanced.
specimen is relatively flatter with the increase of the stirrups ratios and the energy dissipation
capacity and the ductility of members are enhanced.
200 200 200
150 150 150
100 100 100
50 50 50
P(kN)
P(kN)
P(kN)
0 0 0
-50 -50 -50
-100 -100 -100
-150 Hysteretic curve -150 Hysteretic curve -150 Hysteretic curve

Skeleton curve Skeleton curve Skeleton curve
-200 -200 -200
-30 -20 -10 0 10 20 30 -30 -20 -10 0 10 20 30 -20 -10 0 10 20
Δ(mm) Δ(mm) Δ(mm)
(a) (b) (c)
200 200 200
150 150 150
100 100 100
50 50 50
P(kN)
P(kN)
P(kN)
0 0 0
-50 -50 -50
-100 -100 -100
-150 -150 Hysteretic curve -150 Hysteretic curve

Hysteretic curve
Skeleton curve Skeleton curve Skeleton curve
-200 -200 -200
-20 -10 0 10 20 -30 -20 -10 0 10 20 30 -40 -30 -20 -10 0 10 20 30 40
Δ(mm) Δ(mm) Δ(mm)
(d) (e) (f)
200
150
100
50
P(kN)
-50
-100
-150 Hysteretic curve

Skeleton curve
-200
-30 -20 -10 0 10 20 30
Δ(mm)
(g)
Figure
Figure 5. 5.Hysteresis
Hysteresiscurves
curvesand
andskeleton
skeletoncurve
curveof
ofspecimens.
specimens.(a)
(a) SRC-2;
SRC-2; (b) SRC-3; (c) SRC-4;
SRC-4; (d)
(d) SRC-
SRC-5;
(e)5;SRC-6;
(e) SRC-6; (f) SRC-7;
(f) SRC-7; (g) SRC-8.
(g) SRC-8.
4. 4.
Determination
DeterminationofofSkeleton
SkeletonCurve
Curve
4.1. Simplified
4.1. SimplifiedSkeleton
SkeletonCurve
Curve
AtAtpresent,
present,most
mostskeleton
skeleton curves
curves are
are obtained by connecting
obtained by connectingthe
thepeak
peakpoints
pointsofofthe
themember
member
hysteresis curves. In fact, the members would
hysteresis curves. In fact, the members would produce a certain degree of damage under the process
a certain degree of damage under the process
of of
cyclic loading.
cyclic loading.With
Withthe
theincrease
increaseof
ofthe
the number cycles and
number of cycles anddisplacement
displacementamplitude,
amplitude,thethe damage
damage
accumulation of the members continuously increases. Therefore, it is difficult to determine the
Appl. Sci. 2019, 9, 687 7 of 19
accumulation of the members continuously increases. Therefore, it is difficult to determine the skeleton
curve
Appl. by
Sci. the
Appl. Sci. 2016, existing
6, x FOR PEER
2016, 6, x FOR PEER method
REVIEW
REVIEW and truly reflect the degradation of the mechanical properties 7 ofof20the
7 of 20
members under an earthquake. A monotonic load–displacement curve can be used as a skeleton curve
to skeleton
skeleton curve
curvethe
better reflect by the
the existing
existing method
bydegradation method and
and truly
of the structural truly reflect
reflect the
members the degradation
degradation of
performance the
the mechanical
ofcaused by cyclicproperties
mechanical properties
loading and
of the
of the themembers
members under
under an earthquake.
an earthquake. A monotonic
A monotonic load–displacement
load–displacement curve
curve can can be
be used
used as
as aa
describe hysteresis characteristics of structural members under an earthquake [56,57]. Therefore,
skeleton curve to better reflect the degradation of the structural members performance caused by
in skeleton
this paper, curvethetoload-displacement
better reflect the degradation curve under of monotonic
the structural members
loading performance
of members wascausedused as bythe
cyclic loading
cyclic loading and
and describe
describe the hysteresis characteristics of structural members under an earthquake
skeleton curve as shown in theFigurehysteresis
6. Based characteristics
on the results of structural
of this test, members
it canunder be seen an earthquake
that the SRC
[56,57]. Therefore, in this paper, the load-displacement curve under monotonic loading of members
frame column slightly cracks before yielding. In addition, the lateral deformation of themembers
[56,57]. Therefore, in this paper, the load-displacement curve under monotonic loading of member is
was used
was used as
as the skeleton
theslope
skeleton curve
curve as shown
as shown in Figure
in changes 6. Based
Figure 6. little.
Based At on the
on the results
the results of this
of thisthetest,
test, it can be
be seen
it canstiffness seen of
smaller
that
and
the SRC
theframe
of
column
the loading
slightly
curve
cracks before yielding. In addition,
samethe
time,
lateral
initial
deformation of the
that
themember the
skeleton SRC frame
curve column
ofandthe the slightly
members cracks
is the before
relatively yielding.
large, there In addition,
is nolittle.
obviousthe lateral deformation
inflection point of
when thethe
is smaller slope of loading curve changes
member is smaller and the slope of the loading curve changes little. At the same time, the initial At the same time, the initial
specimen
stiffness yields. The strength after yielding is is still greatly large, improved. Therefore, at the initial point stage of
stiffness of of the
the skeleton
skeleton curve curve of of the
the members
members is relatively
relatively large, there there is is no
no obvious
obvious inflection
inflection point
loading,
when the skeleton curves are basically coincident and the peak load is close. After the peak load,
when thethespecimen
specimen yields. yields. TheThe strength
strength after after yielding
yielding is isstill
still greatly
greatlyimproved.
improved. Therefore,
Therefore,at atthe
theinitial
initial
although
stage the strength of the member decreases, the slope of the falling part is not large, which indicate
stage ofof loading,
loading, the the skeleton
skeleton curvescurves are arebasically
basically coincident
coincident and and thethepeak
peak load
load is is close.
close. After
After the the peak
peak
that thealthough
load, strengththe degradation ofthe
themember
members is relatively slow.ofHowever, the damage accumulation
load, although the strength of the member decreases, the slope of the falling part is not large, which
strength of decreases, the slope the falling part is not large, which
of indicate
the membersthe continually degradation
increases underthe repeated loading, which induces the degradation of the
indicate thatthat the strength
strength degradation of of the members
members is is relatively
relatively slow.slow. However,
However, the the damage
damage
bearing capacityof
accumulation
accumulation and
of the
thestiffness,
members
members and the falling increases
continually
continually part of theunder
increases skeleton
under curve loading,
repeated
repeated becomeswhich
loading, steeper.
which The ultimate
induces
induces the
the
degradation
deformation isof the
reduced.bearing capacity
Considering and stiffness,
that the and
main the
purposefalling
degradation of the bearing capacity and stiffness, and the falling part of the skeleton curve becomes part
of the of the skeleton
elastoplastic curve
seismic becomes
response
steeper.
analysis
steeper. isThe ultimate
to study
The ultimate thedeformation
mechanical is
deformation is reduced.
behavior
reduced. ofConsidering
the memberthat
Considering after
that the main
main purpose
theentering of
of the
the inelastic
purpose the elastoplastic
stage, in order
elastoplastic
seismic
to seismic response
simplifyresponse analysis
the calculation, is
analysisinisthis to study
paper,the
to study the mechanical
the mechanical
skeleton curve behavior
behavior of
beforeofthe the member
theyield
member after
of theafter entering
SRCentering
frame column the
the
inelastic
caninelastic stage,
be simplified inbyorder
a to simplify
connection from the calculation,
the coordinate in this
origin paper,
to the
stage, in order to simplify the calculation, in this paper, the skeleton curve before the yieldthe skeleton
yield point. curve before
Therefore, the
the yield
skeleton
of
of the
curve ofSRC
the frame
the SRC
SRC frameframe column
column can
can be
column simplified
becan be simplified
simplified by
by aa connection
to an ideal
connection from the
the coordinate
trilinear
from skeleton origin
coordinate curve to
origin the
model
to the yield that is
yield
point.
shown Therefore,
point.inTherefore,
Figure 7.the the
The skeleton
segment
skeleton curve of
OAofis the
curve the
theSRCSRC frame
elastic
frame column
stage,columnand can can be
the segmentsimplified
be simplified AB thatto an
to anhas ideal
ideal trilinear
a significant
trilinear
skeleton
skeleton
strength curve
curve model
hardening modelsectionthat
thatinis
is shown
shown
the in
in Figure
skeleton Figurecurve 7.
7. The
The
after segment
segment
the specimen OA
OA is is the
the elastic
elasticisstage,
yielded, stage, and
and thethe segment
the elastic-plastic segment stage.
AB
AB that
that has
has a
a significant
significant strength
strength hardening
hardening section
section in
in the
the skeleton
skeleton curve
curve after
after the
the specimen
specimen yielded,
yielded,
In addition, the segment BC is the plastic descent segment.
is
is the
the elastic-plastic
elastic-plastic stage. stage. InIn addition,
addition, the the segment
segment BC BC is is the
the plastic
plastic descent
descent segment.
segment.
200
200
crack point
150 crack point
150 Yield point
Yield point
Peak point
100 Peak point
100 Ultimate point
Ultimate point
50
50
P(kN)
P(kN)
0
0
-50
-50
-100
-100
SRC-1 monotonic curve
SRC-1 monotonic curve
-150 SRC-2 hysteretic curve
-150 SRC-2 hysteretic curve
SRC-2 skeleton curve
SRC-2 skeleton curve
-200
-200-50 -40 -30 -20 -10 0 10 20 30 40 50
-50 -40 -30 -20 -10 0 10 20 30 40 50
Δ(mm)
Δ(mm)
Figure
Figure 6. Typical monotonic and cyclic experimental
experimental response of the steel reinforced concrete (SRC)
Figure 6. 6. Typical
Typical monotonicand
monotonic andcyclic
cyclic experimental response
responseofofthe
thesteel
steelreinforced concrete
reinforced (SRC)
concrete (SRC)
frame
frame column.
column.
frame column.
PP
PPm BB
m
Ks K
Knn
PPy A Ks
A C
y C
PPu
u
K
Kee
O
O y m u 
y m u
Figure
Figure Simplified skeletoncurve.
7. Simplified curve.
Figure 7.
7. Simplified skeleton
skeleton curve.
4.2.
4.2. Characteristic
Characteristic Points
Points
Appl. Sci. 2019, 9, 687 8 of 19

4.2. Characteristic Points
In order to obtain the proposed trilinear skeleton curve model of SRC frame column in this paper,
In order to obtain the proposed trilinear skeleton curve model of SRC frame column in this paper,
the six quantities should be determined firstly: the yield load Py , the yield displacement y , the
the six quantities should be determined firstly: the yield load Py , the yield displacement ∆y , the peak
peak load , the peak displacement 
load Pm , the mpeak displacement ∆m , the ultimate
P m , the displacement ∆u , and the
ultimate displacement u , and load Pu .load Pu .
the ultimate
ultimate
The following assumptions
assumptions areare used
used toto calculate
calculate the
the six
six quantities:
quantities:
a) the section strain is subject to the plane section assumption.
is subject to the plane section assumption.
b) the tensile effect of
of the
the concrete
concrete is
is ignored
ignored in in the
the tension
tension zone.
zone.
c) the
the steel
steeland
andrebars
rebarshave
havethe
thesame
samematerial
materialproperties
propertiesunder compression
under compression andand
tension, andand
tension, are
subject to ideal elastoplastic stress–strain relationship.
are subject to ideal elastoplastic stress–strain relationship.
Yield Displacement ∆y
Load PPy and Yield Displacement y
4.2.1.
4.2.1. The
The Yield
Yield Load y and
(1) Large eccentric Compression Columns

(1) Large eccentric Compression Columns
For large eccentric compression, the yield point of the SRC columns corresponds to the steel
For large eccentric compression, the yield point of the SRC columns corresponds to the steel
yielding at the tension flange. According to the test results, when the longitudinal tension rebars yield,
yielding at the tension flange. According to the test results, when the longitudinal tension rebars yield,
the steel tension flange is also close to yielding. The maximum strain of concrete ε 0 adopted in this
the steel tension flange is also close to yielding. The maximum strain of concrete  0 adopted in this
paper is more than 0.002 that is larger than the yield strain of the longitudinal rebars and the steel
paper isTherefore,
flange. more thanit0.002 that
can be is larger
assumed than
that thethe yield strain
concrete stress of
in the
the longitudinal
compressionrebars
zone isand the steel
a quadratic
flange. Therefore, it can be assumed that the concrete stress in the compression zone is
standard parabolic distribution, and the calculation diagram of the large eccentric compression SRC a quadratic
standardand
column parabolic distribution,
the stress–strain and the calculation
distribution diagram
of the section of theinlarge
are shown eccentric
Figure 8. compression SRC
column and the stress–strain distribution of the section are shown in Figure 8.
As As σ cy
ar
Asσs
σ sf
x
N
h0
h
My
fs As f y
ar
As b As
Figure 8. Calculation diagram of large eccentric compression at the

the yield
yield state.
state.
As
As shown
shown in in Figure
Figure 8,
8, the
the section
section curvature
curvature when
when thethe root
root section
section of
of the
the frame
frame column
column yields
yields is
is
defined as:
defined as: εy
ϕy = . (1)
h0− x
 
y
. (1)
where ϕy is the yield curvature of members, ε yy is the h0 yield
x strain of longitudinal rebars in the tensile
zone, h0 is the distance from the concrete compression edge to the center of the resultant force of the
where 
tension is the
rebars,
y x isyield curvature
the height of theofconcrete
members,  y is the zone.
compression yield strain of longitudinal rebars in the
Key
tensile for calculating
zone, h0 is the the yield from
distance curvature of the section
the concrete is to determine
compression the center
edge to the heightof
of the
the resultant
concrete
compression zone x, referring to Figure 8, the equations of internal
force of the tension rebars, x is the height of the concrete compression zone. force in the cross section are
expressed
Key foras follow:
calculating the yield curvature of the section is to determine the height of the concrete
compression zone x , referring to Figure 8, the equations of internal force in the cross section are
1 2 1
N + f y Aas
expressed f s Asf + f s tw (h − x − as − t) = σ0 s A0 s + σcy bx + σ0 sf tw ( x − a0 s − t) + σ0 sf A0 sf , (2)
s +follow:
2 3 2
1 t 2 1
f y As hN  f A  f A  f t ( h 1x  a  t )   s Aas s− t)(
0 + fys As sf ( hs −sf as 2− s2 w) + 6 f s tws ( h − x − cy bx
x+  2h sf−t w2a
( xs−as2t
 )t )+  
 0.5Nh
sf Asf , (2)
3 2 , (3)
0 0 0 0 0 0 t 1 0 0 0
= My + σ s A s a r + σ sf A sf ( a s + 2 ) + 6 σ sf tw ( x − a s − t)(2x + a s + t) + 4 σcy bx 1 2
t 1
f y As h0  fs Asf ( h  as  )  fs t w ( h  x  as  t )( x  2h  2as  2t )  0.5Nh
where, 2 6
0 , (3)
t 0 =1 x − a r f 1
        
 M y   s As ar   sf Asf ( as  )s  sf t w ( x yas  t )(2 x  as  t )   cy bx
σ 2 (4)
2 6 h0 − x 4
where,
x  ar
 s  fy (4)
h0  x
Appl. Sci. 2019, 9, 687 x  as 9 of 19
 sf  fs , (5)
h  x  as
x − a0 s
σ0 sf = x k f s , (5)
 cy  h − x − ays fc , (6)
h0  x  0
x kε y
σcy = · f , (6)
k h0−(1x vε)0. c (7)
where N is the axial expression force, f s kis=the

α(1are
+ the
λv )tensile
. (7)
strength of the steel, f y is the tensile
where
strength N of
is the longitudinal
axial expression rebar, As fand
force, As are
s is the arethethetensile
area of strength
longitudinal of the steel,and
tensile f y is the tensile
compression
0 are the area of longitudinal tensile and compression rebar,
rebar, respectively,  sf is the compressive
strength of the longitudinal rebar, A s and A s stress of the steel,  s is the compressive stress of the
respectively, 0 is the compressive stress of the steel, σ 0 is the compressive stress of the longitudinal
longitudinal rebars,  cy is the compressive stress of sconcrete edge when longitudinal bars yield
σ sf
rebars, σcy is the compressive stress of concrete edge when longitudinal bars yield under tension, A0 sf
under tension, Asf and Asf are the cross sectional area of top flange and bottom flange of steel,
and Asf are the cross sectional area of top flange and bottom flange of steel, respectively, tw is the
respectively,
thickness tw issteel
of the the web,
thickness of thickness
t is the the steel web,
of thet is the flange,
steel thickness
b isofthethesection
steel flange,
width of b isthe
thecolumn,
section
hwidth
is theof the column,
section heighthof is the
the column,
section height of the
as is the column,
distance a is the
froms the steeldistance
top flangefromtothe
thesteel
columntop flange
edge,
0
to the column edge, a  is the distance from the steel bottom flange
a s is the distance froms the steel bottom flange to the column edge, ar and a r are the distanceto the 0
column edge, a r and
froma r
the
are center of gravity
the distance fromofthe thecenter
longitudinal rebar
of gravity of to
thethe edge of therebar
longitudinal tension zone
to the andoffrom
edge the the center
tension of
zone
gravity of the longitudinal rebar to the edge of the compression
and from the center of gravity of the longitudinal rebar to the edge of the compression zone, respectively, f c is the concrete
zone,
compression f c is theM
respectively, strength, y is the compression
concrete section yield strength,
bending moment,M y is thek issection
the strength enhancement
yield bending moment, factork
considering the restraint effect of stirrups and steel on concrete, α is the consider the constraint effect
is the strength enhancement factor considering the restraint effect of stirrups and steel on concrete,
of steel on core concrete, according to reference, α is equal to 1.1 in this research [53], λv is the stirrups
 is the consider the constraint effect of steel on core concrete, according to reference,  is equal to
characteristic value.
1.1 in this research [53], v is the stirrups characteristic value.
The height of the concrete compression zone x and the section yield bending moment My can be
The height of the concrete compression zone x and the section yield bending moment M y
calculated by Equation (2) and Equation (5), respectively.
can be (2)calculated
Small eccentric by Equation (2) andcolumns
compression Equation (5), respectively.
(2) Small
For small eccentric
eccentric compression
compression,columns
according to the test results, when the member yields, the steel
For small
compression eccentric
flange compression,
has yielded according
or nearly yielded, tothe
the concrete
test results, when
stress theedge
at the memberof the yields, the steel
compression
compression
zone flange
also reaches thehas yielded orstrength
compressive nearly yielded, the and
of concrete, concrete stress
the steel at theflange
tension edge ofhasthe notcompression
yet yielded.
So, the calculation diagrams of the small eccentric compression SRC column and the stress–not
zone also reaches the compressive strength of concrete, and the steel tension flange has yet
strain
yielded. So, the calculation diagrams
distribution of the section are shown in Figure 9. of the small eccentric compression SRC column and the
stress– strain distribution of the section are shown in Figure 9.
As As kf c
ar
As f y
as
f s
My
x
h0
N
h
σ sf
as
As σ s
ar
As b As
Figure 9. Calculation
Figure 9. Calculation diagram
diagram of
of the
the small
small eccentric
eccentric compression
compression at
at yield
yieldstate.
state.
It can be seen from Figure 9 that the curvature of the section can be calculated as follows when
It can be seen from Figure 9 that the curvature of the section can be calculated as follows when
the root section of the frame column yields,
the root section of the frame column yields，
ϕy =  0
ε0
(8)
y x (8)
x
According to Figure 9, the equations of the internal force in the cross section are given as follows:
According to Figure 9, the equations of the internal force in the cross section are given as follows:
1 2 1
N + σs As + σsf Asf + σsf tw (h − as − t − x ) = f 0 y A0 s + f 0 s A0 sf + k f c bx + f 0 s tw ( x − a0 s − t) (9)
2 3 2
1 2 1
N   s As   sf Asf   sf tw ( h  as  t  x)  f y As  fsAsf  kfc bx  fst w ( x  as  t ) (9)
2 3 2
t 1 t
 s As h0   sf Asf ( h  as  )   sf t w ( h  as   x)( x  2h  2as  t )  0.5Nh
Appl. Sci. 2019, 9, 687 2 6 2 10 of 19
, (10)
t 1 1 t
 M y  f y As ar  fsAsf ( as  )  kfc bx 2  fst w ( x  as  )( x  2as  t )
2 4 6 2
σs As h0 + σsf Asf (h − as − 2t ) + 16 σsf tw (h − as − 2t − x )( x + 2h − 2as − t) + 0.5Nh
, (10)
= My + f 0 y A0 s a0 r + f 0 s A0 sf ( a0 s + 2t ) + 14 k f c bx2 + 16 f 0 s tw ( x − a0 s − 2t )( x + 2as + t)
Where,
where, h  x  as
h −
sf 
x −xas a fs , (11)
σsf = f ss, (11)
x − a0 s
h x
h0 − x0 f .
σs = s 0x  f ya. r y (12)(12)
x−a r
wherewhere 
a0 s is athe
s isdistance
the distance
fromfrom the resultant
the resultant forceforce of longitudinal
of longitudinal compression
compression rebars
rebars to to
thethe edge of
edge
0 and f 0 are the compression strength of steel and longitudinal rebars,
of thethe compression zone, s fs andy f y are the compression strength of steel and longitudinal rebars,
compression zone, f
respectively, σsf and σs are the tensile stress of steel and longitudinal rebars, respectively, the rest of the
respectively,  sf and  s are the tensile stress of steel and longitudinal rebars, respectively, the rest
symbols are the same as before.
of the symbols are the same as before.
The height of the concrete compression zone x and the section yield bending moment My can be
The height of the concrete compression zone x and the section yield bending moment My can
calculated by Equation (9) and Equation (10), respectively.
be calculated
Based by Equation
on the practical (9)condition
stress and Equation
of the(10),
framerespectively.
column, the frame column can be simplified
Based on the practical stress condition of
as a cantilever which bears axial pressure and horizontal force, the frame column, the frame
as shown column
in Figure 10. can be simplified
Meanwhile,
as assume
we can a cantileverthat which bearscurvature
the section axial pressure
of the and
SRC horizontal force, asdistributed
column is linearly shown in Figure
along the10. column
Meanwhile,
weWhen
height. can assumethe rootthat the section
section of the curvature
frame column of the SRC
hass columnthe
yielded, is linearly
horizontaldistributed along the column
yield displacement of
height. When the root section
the top of the column can be expressed as: of the frame column hass yielded, the horizontal yield displacement of
the top of the column can be expressed as:
1 2
∆y = ϕy H 1 . (13)
3y   y H 2 . (13)
3
where H is the distance from the loading point to the column root.
where H is the distance from the loading point to the column root.
Axial pressure N
Δ
P
Horizontal force
(a) (b)
Figure
Figure 10. Load
10. Load and curvature
and curvature distribution
distribution of theofframe
the frame columns.
columns. (a) Force
(a) Force diagram;
diagram; (b) distribution
(b) distribution
of curvature.
of curvature.
According to thetoforce
According balance
the force condition,
balance whenwhen
condition, the root
the section of theofframe
root section column
the frame has yielded,
column has yielded,
the horizontal load load
the horizontal of theoftop
theoftop
theofcolumn is expressed
the column as: as:
is expressed
My −M y yN y
N∆
Py = Py  . . (14)(14)
H H
4.2.2. Peak load Pm and peak displacement ∆m

4.2.2. Peak load Pm and peak displacement  m
When the section reaches the maximum flexural capacity, the peak load of the frame column can
be calculated by:
Mmax − N∆m
Pm = (15)
H
where ∆m is the horizontal displacement corresponding to the peak load of the frame column, Mmax is
the maximum flexural capacity at the root section of the frame column.
Appl. Sci. 2019, 9, 687 11 of 19
According to the simplified skeleton curve obtained above, the peak displacement ∆m can be
expressed as:
Pm − Py
∆m = + ∆y . (16)
Ks
where Ks is the hardening stiffness, Ks = αKe , Ke is the elastic stiffness, αs is the hardening coefficient
which has the relationship with the axial compression ratio n, 0.2 ≤ n ≤ 0.4, αs = 0.3, 0.4 < n ≤ 0.6,
αs = 0.4.
4.2.3. Ultimate Load Pu and Ultimate Displacement ∆u

The ultimate load can be 80% of the peak load when the root of the frame column is failure,
namely, Pu = 0.8Pm . The ultimate displacement can be expressed as follows:
Pm − Pu 0.2Pm
∆u = ∆m − = ∆m − . (17)
Kn Kn
where Kn is the softening stiffness, Kn = αKe , αn is the softening coefficient which has the relationship
with the axial compression ratio n, based on the statistical analysis of test data, 0.2 ≤ n ≤ 0.4,
αn = −0.07, 0.4 < n ≤ 0.6, αn = −0.1.
5. Damage-Based Hysteretic Analysis
5.1. Simplification of the Hysteresis Loop

An intact hysteresis loop consists of a loading curve and an unloading curve. According to
the results of this test, after yielding of the members, the slope of the loading curve decreases with
increasing displacement, indicating that the stiffness of the members has degraded during each
repeated loading process. Like the loading curve, the slope of the unloading curve also decreases
with the increase of the number of cycles, and the unloading stiffness of the members continuously
degrades. After completing unloading, the member has some residual deformation that accumulates
continuously with the increase of the number of cycles.
In order to better reflect the good seismic performance of the SRC members and enhance the
simulation accuracy of the hysteresis curve in the later stage of loading, according to the results of this
test and the characteristics of the hysteresis curves, the hysteresis loop is simplified as follow,
1. When the horizontal load does not reach the yield load, the hysteresis loop is simplified into three
parts, namely, the elastic section OA (CD), the strengthening section AB (DE), and the unloading
section BC (EF) in Figure 10.
2. When the horizontal load reaches the yield load, the hysteresis loop is simplified into four parts,
namely, the elastic section HI (LM), the strengthening section IG (MN), the softening section GK
Appl. Sci. 2016,and
(NO), 6, x FOR PEER REVIEW
the unloading section KL (OP) in Figure 11. 12 of 20
B
P G K
A
I
C
H
P F 0 L 
O M
D
N E
Figure
Figure 11. Simplified hysteretic
11. Simplified hysteretic loops.
loops.
5.2. Proposed Damage Index

To obtain the seismic damage index for SRC members, based on the test results of the SRC frame
columns under low-cycle loads, the nonlinear combination seismic damage index, which can reflect
Appl. Sci. 2019, 9, 687 12 of 19
5.2. Proposed Damage Index

To obtain the seismic damage index for SRC members, based on the test results of the SRC frame
columns under low-cycle loads, the nonlinear combination seismic damage index, which can reflect
the degradation of strength and stiffness under an earthquake, was developed by considering the
effect of the loading path and the number of loading cycles [58]. This damage index is used in this
paper to analyze the hysteretic behavior of the SRC column, and is expressed as:
!c !c
N1 ∆max,j − ∆y Nh
Ei
D = (1 − γ ) ∑ + γ∑ (18)
j =1 ∆u,i − ∆y i =1 Eu,i
where,
∆u,i = A + Be−α , (19)
Eu,i = A + Be−α , (20)
c = 5.69 + 0.87 ln ρv + 0.056λ + 10.46ρs − 2.1n. (21)
where D is the damage index, ∆max,j is the largest inelastic deformation corresponding to the j-th
half-cycle, N1 is the number of half-cycles of ∆max,j generated for the first time, Ei is the hysteretic
energy dissipation of the i-th half-cycle, Nh is the number of half-cycles, γ is a composite parameter and
is equal to 0.15 in this paper, c is the experimental parameter, Eu,i and ∆u,i are the normalized ultimate
energy dissipation capacity and ultimate deformation under constant axial load and monotonic lateral
load after the i-th half-cycle, respectively, α is the normalized cycle cumulative energy dissipation under
Nh
the i-th half-cycle, α = ∑ Ei /Eu,0 , Eu,0 is the ultimate energy dissipation capacity under monotonic
i =1
loading directly, for ultimate energy consumption, A = 0.46, B = 0.54, and for ultimate deformation,
A = 0.76, B = 0.24, λ is the shear span ratio and equals 3.0 in this paper.
5.3. Cyclic Strength and Stiffness Degradation in the Hysteretic Model
5.3.1. Damage-Based Cyclic Degradation Index

The cycle degradation index β i can be used to describe the degradation of the performance of
the structures and members under an earthquake. The previous research results have shown that the
loading path has a significant influence on the energy dissipation capacity of the members, therefore,
the cyclic degradation index that was established by hysteretic energy dissipation cannot satisfactorily
consider the influence of the loading path. Meanwhile, it is difficult to give the unified expression
of component energy dissipation capacity under an earthquake. Therefore, in this paper, the cycle
degradation index, which was based on the damage index obtained above, was proposed to better
consider the effect of the loading path and cycle number on the seismic performance of the frame
column, and it is expressed as:
β i = [∆Di /(1 − Di−1 )] ϕ (22)
where ∆Di is the increment of the component damage value for the i-th loading cycle; Di−1 is the
cumulative damage value of the member before the i-th loading cycle; and ϕ is the correlation
coefficient, based on the test results, and is equal to 1.2 in this research. Based on the degree of damage
of the member under cyclic loading, the cyclic degradation index can describe the degradation of
member performance. Meanwhile, the impact of the loading path on the cyclic degradation index is
also considered. The value of the cyclic degradation index β i is between 0 and 1. When the value is
closer to 1, it indicates that the degradation of the member performance is more serious. If β i < 0 or
β i > 1, it means that the damage value increment of the member exceeds the residual damage value
of the member under a certain cyclic loading, which indicates that the structural member is invalid.
Therefore, the structure members failure criterion can be expressed as:
Appl. Sci. 2019, 9, 687 13 of 19
∆Di > 1 − Di−1 . (23)
In the following, the degradation rule of the member performance is described by using the
damage-based cyclic degradation index β i .
Remark: The damage index (Equation (18)), the cyclic degradation index (Equation (22)), and the
associate failure criterion (Equation (23)) are hysteresis based. Both are obtained based on the
hysteresis curves through destructive tests. However, currently there is a lack of studies to link
the sensor-based damage index with the hysteresis-based damage index through destructive tests.
It would be interesting to conduct tests of SRC specimens integrated with embedded piezoceramic
smart aggregates and to correlate these two indices. If these two indices correlate well, the health
status of an in-service structure can be monitored in real-time through embedded sensors based on the
sensor enabled damage index. In addition, the failure criterion (Equation (23)) can help to establish the
corresponding criterion for the sensor-based damage index.
5.3.2. Degradation Analysis of Strength and Stiffness

(1) Degradation of yield load
With the increase of the number of load cycles, the yield load of the member decreases
continuously after each reverse loading and reloading, and the degradation rule of the yield load is:
±
Pyi = (1 − β i ) Py±(i−1) . (24)
±
where Pyi is the yield load of the member after the i-th cycle loading, and Py±(i−1) is the yield load of the
member before the i-th cycle loading. The superscript "+" means positive loading, "−" means reverse
loading, and the same is shown below.
(2) Degradation of hardening stiffness
With the increase of the number of load cycles, the hardening stiffness of the members is
degenerated, and the degeneration rule is:
±
Ksi = (1 − β i )Ks±(i−1) . (25)
±
where Ksi is the hardening stiffness of the member after the i-th cycle loading, and Ks±(i−1) is the
hardening stiffness of the member before the i-th cycle loading.
(3) Degradation of softening stiffness
With the increase of the number of load cycles and the displacement, the softening stiffness of the
member gradually degrades and approaches to the origin, the degradation rule of softening stiffness
can be determined by the following equation:
±
Kni = (1 − β i )Kn±(i−1) (26)
±
where Kni is the softening stiffness of the member after the i-th cycle loading and Kn±(i−1) is the softening
stiffness of the member before the i-th cycle loading.
Figure 12 shows the degradation rule of member yield load, hardening stiffness, and softening
stiffness. The member is loaded from point 0 to point 2 along the loading path 0–1–2. The initial
+
loading stiffness is Ke , the hardening stiffness is Ks0 , the softening stiffness is Kn+ . Then along point 2
unloaded to point 3, the unloading path is 2–3. A half cycle is completed from point 0 to point 3
according to Equation (18) to calculate a damage value. Then the degradation index β i is calculated
through Equation (22), and the respectively corresponding strength and stiffness values are calculated
by Equation (24), Equation (25), and Equation (26). In the reverse loading, the yield load of the
− −
members decreases from Py− to Py1 , the slope of the hardening stiffness decreases from Ks0 to Ks1 , and
−
the slope of the negative stiffness decreases from Kn− to Kn1 . From point 6 unloads to point 7, the
according to Equation (18) to calculate a damage value. Then the degradation index  i is calculated
through Equation (22), and the respectively corresponding strength and stiffness values are
calculated by Equation (24), Equation (25), and Equation (26). In the reverse loading, the yield load
of the members decreases from Py- to Py1- , the slope of the hardening stiffness decreases from K s0
Appl. Sci. 2019, 9, 687
- 14 of 19
to Ks1 , and the slope of the negative stiffness decreases from Kn- to Kn1
-
. From point 6 unloads to
point 7, the damage value is calculated by Equation (18) again. Next the half-cycle damage value
damage 
increment value
Di isiscalculated,
calculated and
by Equation
then the(18) again. Next
degradation  i is recalculated
the half-cycle
index damage value (22).∆Di
increment
by Equation
is calculated, and then the degradation index
When the member is reloaded along the load path 7–8–9, β i is recalculated by Equation (22). When the member
the yield load of the structure reduces from
+
Py to Py1 , the hardening stiffness decreases from K s0 to ofKthe
+ is reloaded
+ along the load path 7–8–9, the yield load + structure reduces from Py+ to Py1 , the
s1 , and the softening stiffness decreases
+ + +
hardening stiffness decreases from Ks0 to Ks1 , and the softening stiffness decreases from Kn to Kn1 .
from Kn+ to Kn1 +
.
P 2
K s0
Py+ 1 Kn
Py1+  9
8 K s1 
Kn1
10
Ke
7
0 3 Δ
6  Ks1 Py1-
Kn1 4
5
Py-
Kn Ks0
Figure
Figure 12. Degradation
12. Degradation of yield
of yield load,
load, hardening
hardening stiffness,
stiffness, andand softening
softening stiffness
stiffness of members.
of members.
(4) (4) Degradation

Degradation of unloading
of unloading stiffness
stiffness
The existing research results showthat
The existing research results show thethe
that unloading stiffness
unloading of the of
stiffness member does not does
the member significantly
not
degrade during the elastic phase and the elastoplastic stage, which is almost the same
significantly degrade during the elastic phase and the elastoplastic stage, which is almost the same as the initial
stiffness K e .When the horizontal load reaches the peak load and the structure is in the
as the initial stiffness K e .When the horizontal load reaches the peak load and the structure is in the plastic force
state, the unloading stiffness of the members degenerates and can be described as follows:
plastic force state, the unloading stiffness of the members degenerates and can be described as follows:
i  (1  i )Kiu( i 1u
KKuui = (1 − β ) K ) ( i −1) (27) (27)
where
where K uiKuiisisthe
theunloading
unloading stiffness
stiffness ofof
thethe member
member after
after i-thi-th
thethe cycle
cycle loading,
loading, andandKu(Ki u1() i−is is the
1) the
unloading stiffness of the component before the
unloading stiffness of the component before the i-th cycle loading.i-th cycle loading.
TheThe unloading
unloading stiffness
stiffness degradation
degradation of the
of the member
member in the
in the plastic
plastic phase
phase is shown
is shown in Figure
in Figure 12. 12.
The loading starts from point 0 to point 2 along the loading path 0–1–2. At this time, according to to
The loading starts from point 0 to point 2 along the loading path 0–1–2. At this time, according
Equation
Equation (18),
(18), wewecancan calculate
calculate a damagevalue.
a damage value.Then,
Then,according
accordingto to Equation
Equation (22),
(22), we
we can
can calculate
calculatethe
thedegradation
degradationindex indexβ i . The unloading stiffness is calculated through Equation (27), and the unloading
i . The unloading stiffness is calculated through Equation (27), and the
stiffness decreases from the initial stiffness Ke to Ku1 at this time. When reverse loading reaches point 5,
unloading stiffness decreases from the initial stiffness K e to K u1 at this time. When reverse loading
the damage value is calculated by Equation (18) again. At the same time, the damage value increment
reaches point 5, the damage value is calculated by Equation (18) again. At the same time, the damage
∆Di is calculated from the unloading point 2 to reloading point 5. Then, the deterioration index β i is
value increment Di is calculated from the unloading point 2 to reloading point 5. Then, the
recalculated by Equation (22), and the unloading stiffness is calculated by Equation (27). At this time,
the unloading rigidity of the member decreases from Ku1 to Ku2 .
(5) Degradation of reloading stiffness
Based on the assumptions of the previous hysteresis loops, the reloading stiffness is related to the
unloading stiffness of the previous cycle. The degradation rule of reloading stiffness can be expressed
by the following equation:
Kri = (1 − β i )Kui . (28)
where Kri is the reloading stiffness of the member after the i-th cycle loading.
As shown in Figure 13, the loading starts from point 0 to the reverse loading point 3 along the
loading path 0-1-2-3. At this time according to Equation (18) to calculate a damage value, and then
according to Equation (22) to calculate the degradation index β i . During the reverse loading, according
to the Equation (28), the reloading stiffness of the member is changed from Ke to Kr1 , and then the
reverse loading starts from point 3 to the reverse unloading point 6 along the loading path 3-4-5-6.
At this time according to Equation (18) to calculate a damage value. At the same time, the damage
increment ∆Di of the half cycle is calculated, and then the degradation index β i is recalculated by
Equation (22). During reloading process, the reloading stiffness of the member is changed from Kr1
to Kr2 .
according to Equation (22) to calculate the degradation index  i . During the reverse loading,
according to the Equation (28), the reloading stiffness of the member is changed from K e to K r1 ,
and then the reverse loading starts from point 3 to the reverse unloading point 6 along the loading
path 3-4-5-6. At this time according to Equation (18) to calculate a damage value. At the same time,
the damage increment Di of the half cycle is calculated, and then the degradation index  i is
Appl. Sci. 2019, 9, 687 15 of 19
recalculated by Equation (22). During reloading process, the reloading stiffness of the member is
changed from K r1 to K r2 .
P 2
1
7
Kr2 Ku1
Ke Ke
6
0 3 Δ
Ku1 Kr1
Ku2
4
5
Figure13.
Figure Degradationofofunloading
13.Degradation unloading stiffness
stiffness andand reloading
reloading stiffness
stiffness of members.
of members.
5.4.
5.4. Hysteresis Rules
Hysteresis Rules
Based
Based onon the
the above
abovedescription
descriptionofofthe
theperformance
performance degradation
degradation of the
of the member,
member, the hysteresis
the hysteresis
rules
rules of
of the
the SRC framecolumn
SRC frame columncan
canbe
besummarized
summarized asas follows:
follows:
1. Before the component has not yielded, the loading and unloading are performed along the elastic
1. Before the component has not yielded, the loading and unloading are performed along the elastic
segment of the member skeleton curve.
segment of the member skeleton curve.
2. After the load born by the member reaches the yield load, the loading path is performed along
2. After the load born by the member reaches the yield load, the loading path is performed along
the skeleton curve of the member. During the unloading process, the corresponding damage
the skeleton
value is calculated curvebyof the member.
Equation (18) at During the unloading
the unloading process,
point. At the same the
time,corresponding
the correspondingdamage
value is calculated
degradation index by Equation (18) at the unloading point. At the same time, the corresponding
i is calculated by Equation (22), and the unloading stiffness is calculated
degradation
by Equation (27).
index β i is calculated by Equation (22), and the unloading stiffness is calculated by
Equation (27).
3. The reverse loading and reloading path: after a half-cycle is completed, the damage value of the
3. member
The reverse loading and
is recalculated, andreloading path:
the half-cycle after value
damage a half-cycle
incrementis completed, the damage
Di is calculated. Then,value
the of
the memberindex
degradation is recalculated, and the
 i is calculated half-cycle
according damage value
to Equation increment
(22). The ∆Di is calculated.
reverse loading stiffness, theThen,
the degradation index β i is calculated according to Equation (22). The reverse
yield load, the hardening stiffness, and the softening stiffness of the member hysteresis loop are
loading stiffness,
the yield load, the hardening stiffness, and the softening stiffness
calculated by Equation (28), (24), (25), and (26) respectively before the loading starts. The of the member hysteresis
loop are calculated
continued loading is by Equation
performed (28),the
along (24), (25), and
softening (26) respectively
stiffness beforeThe
of the members. thestiffness
loading isstarts.
calculated as before, and the path at the time of reloading is the same as described above. stiffness is
The continued loading is performed along the softening stiffness of the members. The
calculated as before, and the path at the time of reloading is the same as described above.
6. Validation of Hysteretic Models
6. Validation of Hysteretic Models
To verify the effectiveness of the hysteretic model established in this paper, the hysteresis curves
To verify the effectiveness of the hysteretic model established in this paper, the hysteresis curves
calculated by this method were compared to the hysteresis curves obtained by the experiment, and the
calculated by this method were compared to the hysteresis curves obtained by the experiment, and
results
theare shown
results are in Figure
shown 14, where
in Figure the data
14, where of the
the data of specimen SRC-6
the specimen SRC-6 and
andSRC-11
SRC-11 are
are used from the
used from
reference
the reference [59]. It can be seen that the results of calculation and test are in good agreement with each
[59]. It can be seen that the results of calculation and test are in good agreement with
other,each
which indicates
other, which that the hysteretic
indicates model ofmodel
that the hysteretic the SRC frame
of the SRCcolumns proposed
frame columns in thisinpaper
proposed this can
betterpaper
describe
can the hysteresis
better describe characteristics of the SRC structural
the hysteresis characteristics members
of the SRC under
structural an earthquake.
members under an
earthquake.
150 P(kN)
200 P(kN)
150 100
100 50
50
Δ(mm)
Δ(mm) 0
0 -20 -10 0 10 20
-20 -10 0 10 20
-50 -50
-100 -100 Test value

Test value
-150 Caculation value
Caculation value
-150
-200
(a) (b)
200 P(kN) 400
P(kN)
150
Figure 14. Cont. 300
100 200
50 100
Δ(mm) Δ(mm)
0 0
-50 -40 -30 -20 -10 0 10 20 30 40 50 -15 -10 -5 0 5 10 15
-50 -100
-200
Δ(mm)
Δ(mm) 0
0 -20 -10 0 10 20
-20 -10 0 10 20
-50 -50
-100 -100 Test value

Test value
-150 Caculation value
Caculation value
Appl. Sci. 2019, 9, 687 -150 16 of 19
-200
(a) (b)
200 P(kN) 400
P(kN)
150 300
100 200
50 100
Δ(mm) Δ(mm)
0 0
-50 -40 -30 -20 -10 0 10 20 30 40 50 -15 -10 -5 0 5 10 15
-50 -100
-100 -200
Test value
Test value
-150 -300 Caculation value
Caculation value
-200 -400
(c) (d)
Figure
Figure 14. Comparison
14. Comparison of testing
of testing and
and calculatinghysteretic
calculating hysteretic curves.
curves.(a)(a)Specimen
SpecimenSRC-4; (b) Specimen
SRC-4; (b) Specimen
SRC-5;
SRC-5; (c) Specimen
(c) Specimen SRC-6;
SRC-6; (d)(d) SpecimenSRC-11.
Specimen SRC-11.
7. Conclusions
7. Conclusions andand Future
Future Work
Work
Inpaper,
In this this paper,
eighteight
SRC SRC frame
frame columnspecimens
column specimens were
weretested
testedunder
undercombined
combinedaxial compression
axial compression
and lateral cyclic load. Base on the test results, a hysteresis model was developed to
and lateral cyclic load. Base on the test results, a hysteresis model was developed to simulate simulate the the
hysteresis behavior of the SRC frame columns, and conclusions were obtained as follows:
hysteresis behavior of the SRC frame columns, and conclusions were obtained as follows:
1. The failure process of the SRC column specimen experiences three stages under the cycle load,
1. The failure
namely, process
the elastic ofstage,
the SRC columnwith
the inelastic specimen experiences
cracks stage, three stages
and the failure under
stage. In the cycle
the elastic stage,load,
the hysteresis
namely, the elastic behavior
stage, of
thethe SRC frame
inelastic with columns
cracks isstage,
similar andto the
the reinforced concrete
failure stage. frame
In the elastic
columns.
stage, In the inelastic
the hysteresis behaviorstage, due SRC
of the to theframe
mutual constraint
columns betweento
is similar steel
theand concrete,concrete
reinforced the
framebearing
columns.capacity
In theof inelastic
steel andstage,
concretedueistoimproved.
the mutual With the increase
constraint of thesteel
between displacement
and concrete,
amplitude and the number of cycles, the hysteresis loops become a plump fusiform. Especially,
the bearing capacity of steel and concrete is improved. With the increase of the displacement
after the peak load, the specimens show a good anti-seismic performance.
amplitude and the number of cycles, the hysteresis loops become a plump fusiform. Especially,
2. With the increase of the axial compression ratio and steel ratio, the seismic performance of the
afterframe
the peak
columnload, is the specimens
enhanced. show aratio
The stirrups good anti-seismic
has performance.
the little effect on the seismic performance of
2. Withthetheframe
increase
columnof the axial
before thecompression ratio andafter
peak load. However, steeltheratio,
peakthe seismic
load, performance
the energy dissipation of the
frame column
capacity and is the
enhanced.
ductility The
of thestirrups
members ratio has the with
is enhanced littlethe
effect on the
increase of seismic performance
the stirrups ratios. of
3. frame
the Based on the comparison
column before the of peak
the test results
load. of the monotonic
However, after the loading
peak and
load,cycle
therepeat
energy loading, a
dissipation
trilinear
capacity andskeleton curve of
the ductility model of the SRCiscolumns
the members enhancedcanwith
be established
the increaseby simplifying the load-
of the stirrups ratios.
displacement curve under monotonic loading. According to the equilibrium of the internal force,
3. Based on the comparison of the test results of the monotonic loading and cycle repeat loading,
a set of calculation equations for determining the simplified skeleton curves of the SRC columns
a trilinear skeleton curve model of the SRC columns can be established by simplifying the
are proposed.
load-displacement curve under monotonic loading. According to the equilibrium of the internal
force, a set of calculation equations for determining the simplified skeleton curves of the SRC
columns are proposed.
4. The hysteresis loop is simplified by combining with the experimental results. Based on the seismic
damage index of the SRC frame columns, the cyclic degradation index is established, which can
reflect the performance degradation of the members very well. Finally, the multi-line hysteresis
model of the SRC frame column is established by establishing the rule of strength and stiffness
degradation of the members and the hysteresis rule. It was shown that the computed hysteresis
curves are in good agreement with the experimental results, which verifies the validity of the
hysteresis model. The proposed model is able to predict the cyclic response of SRC columns with
sufficient accuracy.
Future work will involve testing SRC specimens with embedded piezoceramic smart aggregates.
During the tests, we will obtain both the sensor-based damage index and the hysteresis-based damage
index and study the correlation between the two indices. If the two indices correlate well, we can
use the sensor-based damage index to provide the real-time data of structural health monitoring
for in-service structures with integrated sensors. Meanwhile, the proposed method in this paper
has a low computational efficiency, which is not convenient for real seismic analysis of structures.
Therefore, based on the cumulation damage of the material and the results proposed in this paper,
we will conduct finite element analysis (FEA) of the SRC structure members in future investigations.
Additionally, the stress state of concrete confined by stirrups and steel is more complicated for SRC
Appl. Sci. 2019, 9, 687 17 of 19
members. Therefore, to achieve the refined finite element simulation of the SRC structures, based on
the cyclic constitutive model for concrete confined by transverse reinforcements [60,61], the cyclic
constitutive model of confined concrete established by considering the effect of cumulation damage
will be studied in future research.
Author Contributions: B.W., G.H., and Y.S. designed and performed the experiments; B.W. and S.Z. analyzed the
data; B.W. and G.H. wrote the paper.
Funding: This research was funded by the National Science and Technology Support Program: 2013BAJ08B03;
National Natural Science Foundation of China: 51678475; the Natural Science Foundation of Shaanxi Province:
2018JQ5158; Scientific Research Program was Funded by Shaanxi Provincial Education Department: 18JK0382.
Acknowledgments: The authors appreciate the support of the National Science and Technology Support Program
(2013BAJ08B03), the National Science Foundation of China (51678475), the Natural Science Foundation of
Shaanxi Province (2018JQ5158) and the Scientific Research Program Funded by Shaanxi Provincial Education
Department (18JK0382).
Conflicts of Interest: The authors declare no conflict of interest.
References
1. Xiao, Y.; Zeng, L.; Cui, Z.; Jin, S.; Chen, Y. Experimental and analytical performance evaluation of steel beam
to concrete-encased composite column with unsymmetrical steel section joints. Steel Compos. Struct. 2017, 23,
17–29. [CrossRef]
2. Yan, F.; Xiao, C.Z.; Xu, P.F.; Liu, P.; He, W. Experimental study on the irregular shaped steel reinforced
concrete columns of the China World Trade Center Phase III in Beijing. China Civil Eng. J. 2010, 43, 11–20.
3. Zheng, J.; Dong, L.G.; Zheng, S.S.; Zeng, L.; Xiang, Z.H.; He, W. Researching a fuzzy and performance-based
optimization method for the life-cycle cost of SRHPC frame structures. Appl. Sci. 2017, 7, 269. [CrossRef]
4. Zeng, L.; Xiao, Y.F.; Chen, Y.G.; Jin, S.Q.; Xie, W.; Li, X.J. Seismic damage evaluation of concrete-encased steel
frame-reinforced concrete core tube buildings based on dynamic characteristics. Appl. Sci. 2017, 7, 314. [CrossRef]
5. Park, Y.J.; Reinhorn, A.M.; Kunnath, S.K. IDARC: Inelastic Damage Analysis of Reinforced Concrete
Frame-Shear-Wall Structures; State University of New York at Buffalo: Buffalo, NY, USA, 1987.
6. Ergun, S.; Demir, A. Effect of hysteretic models on seismic behavior of existing RC structures. J. Perform.
Constr. Facil. 2015, 29, 04014160:1–04014160:9. [CrossRef]
7. Zhao, G.F.; Zhang, M.; Li, Y.L.; Li, D.W. The hysteresis performance and restoring force model for corroded
reinforced concrete frame columns. J. Eng. 2016, 2016, 7615385. [CrossRef]
8. Zeynalian, M.; Ronagh, H.R.; Dux, P. Analytical description of pinching, degrading, and sliding in a bilinear
hysteretic system. J. Eng. Mech. 2012, 138, 1381–1387. [CrossRef]
9. Tarutal, G.M.; Sriharsha, R.K.; Prakash, S.S. Hysteresis modelling of reinforced concrete columns under pure
cyclic torsional loading. Struct. Eng. Mech. 2017, 64, 11–21.
10. Zhou, C.D.; Lu, X.L.; Li, H.; Tian, T. Restoring force model for circular RC columns strengthened by
pre-stressed CFRP strips. Steel Compos. Struct. 2014, 17, 371–386. [CrossRef]
11. Lee, C.S.; Han, S.W. Computationally effective and accurate simulation of cyclic behaviour of old reinforced
concrete columns. Eng. Struct. 2018, 173, 892–907. [CrossRef]
12. Tathagata, R.; Andrei, M.R. Enhanced smooth hysteretic model with degrading properties. J. Struct. Eng.
2014, 140, 04013028:1–04013028:10.
13. Wang, P.H.; Ou, Y.C.; Chang, K.C. A new smooth hysteretic model for ductile flexural-dominated reinforced
concrete bridge columns. Earthq. Eng. Struct. Dyn. 2017, 46, 2237–2259. [CrossRef]
14. Ibarra, L.F.; Medina, R.A.; Krawinkler, H. Hysteretic models that incorporate strength and stiffness
deterioration. Earthq. Eng. Struct. Dyn. 2005, 34, 1489–1511. [CrossRef]
15. Li, L.; Zheng, S.S.; Wang, B.; Deng, G.Z.; Wang, W. Cyclic deterioration effect of the steel reinforced high
performance concrete frame. Eng. Mech. 2010, 27, 125–132.
16. Rahnama, M.; Krawinkler, H. Effect of Soils and Hysteresis Models on Seismic Design Spectra; Report No. 108;
John A. Blume Earthquake Engineering Research Center, Stanford University: Stanford, CA, USA, 1993.
17. Rodas, P.T.; Zareian, F.; Kanvinde, A. Hysteretic model for exposed column–base connections. J. Struct. Eng.
2016, 142, 04016137:1–04016137:14. [CrossRef]
Appl. Sci. 2019, 9, 687 18 of 19
18. Takeda, T.; Sozen, M.A.; Neilsen, N.N. Reinforced concrete response to simulated earthquakes. J. Struct. Div.
1970, 96, 2557–2573.
19. Kumar, S.; Usami, T. An evolutionary degrading hysteretic model for thin-walled steel structures. Eng. Struct.
1996, 18, 504–514. [CrossRef]
20. Mitra, M.; Gopalakrishnan, S. Guided wave based structural health monitoring: A review. Smart Mater.
Struct. 2016, 25, 053001. [CrossRef]
21. Song, G.; Wang, C.; Wang, B. (Eds.) Structural Health Monitoring (SHM) of Civil Structures; MDPI: Basel,
Switzerland, 2018; ISBN 978-3-03842-783-4 (Pbk); ISBN 978-3-03842-784-1 (PDF).
22. Li, P.; Gu, H.; Song, G.; Zheng, R.; Mo, Y.L. Concrete structural health monitoring using piezoceramic-based
wireless sensor networks. Smart Struct. Syst. 2010, 6, 731–748. [CrossRef]
23. Moughty, J.J.; Casas, J.R. A State of the art review of modal-based damage detection in bridges: Development,
Challenges, and solutions. Appl. Sci. 2017, 7, 510. [CrossRef]
24. Xu, J.; Hao, J.J.; Li, H.N.; Luo, M.Z.; Guo, W.; Li, W.J. Experimental damage identification of a model
reticulated shell. Appl. Sci. 2017, 7, 362. [CrossRef]
25. Xu, K.; Ren, C.; Deng, Q.; Jin, Q.; Chen, X. Real-time monitoring of bond slip between GFRP bar and concrete
structure using piezoceramic transducer-enabled active sensing. Sensors 2018, 18, 2653. [CrossRef] [PubMed]
26. Huo, L.; Wang, F.; Li, H.; Song, G. A fractal contact theory based model for bolted connection looseness
monitoring using piezoceramic transducers. Smart Mater. Struct. 2017, 26, 104010. [CrossRef]
27. Du, G.; Kong, Q.; Zhou, H.; Gu, H. Multiple cracks detection in pipeline using damage index matrix based
on piezoceramic transducer-enabled stress wave propagation. Sensors 2017, 17, 1812. [CrossRef] [PubMed]
28. Lu, G.; Li, Y.; Zhou, M.; Feng, Q.; Song, G. Detecting damage size and shape in a plate structure using PZT
transducer array. J. Aerosp. Eng. 2018, 31, 04018075. [CrossRef]
29. Tzoura, E.A.; Triantafillou, T.C.; Providakis, C.; Tsantilis, A.; Papanicolaou, C.G.; Karabalis, D. Damage
detection of reinforced concrete columns retrofitted with FRP jackets by using PZT sensors. Struct. Monit.
Maint. 2015, 2, 165–180. [CrossRef]
30. Xu, K.; Deng, Q.; Cai, L.; Ho, S.; Song, G. Damage detection of a concrete column subject to blast loads using
embedded piezoceramic transducers. Sensors 2018, 18, 1377. [CrossRef]
31. Song, G.; Gu, H.; Mo, Y.L.; Hsu, T.T.C.; Dhonde, H. Concrete structural health monitoring using embedded
piezoceramic transducers. Smart Mater. Struct. 2007, 16, 959. [CrossRef]
32. Kong, Q.; Robert, R.H.; Silva, P.; Mo, Y.L. Cyclic crack monitoring of a reinforced concrete column under
simulated pseudo-dynamic loading using piezoceramic-based smart aggregates. Appl. Sci. 2016, 6, 341.
[CrossRef]
33. Zeng, L.; Parvasi, S.M.; Kong, Q.; Huo, L.; Li, M.; Song, G. Bond slip detection of concrete-encased composite
structure using shear wave based active sensing approach. Smart Mater. Struct. 2015, 24, 125026. [CrossRef]
34. Qin, F.; Kong, Q.; Li, M.; Mo, Y.L.; Song, G.; Fan, F. Bond slip detection of steel plate and concrete beams
using smart aggregates. Smart Mater. Struct. 2015, 24, 115039. [CrossRef]
35. Luo, M.; Li, W.; Hei, C.; Song, G. Concrete infill monitoring in concrete-filled FRP tubes using a PZT-based
ultrasonic time-of-flight method. Sensors 2016, 16, 2083. [CrossRef] [PubMed]
36. Zhang, J.; Xu, J.; Guan, W.; Du, G. Damage detection of concrete-filled square steel tube (CFSST) column
joints under cyclic Loading Using Piezoceramic Transducers. Sensors 2018, 18, 3266. [CrossRef] [PubMed]
37. Xu, Y.; Luo, M.; Hei, C.; Song, G. Quantitative evaluation of compactness of concrete-filled fiber-reinforced
polymer tubes using piezoceramic transducers and time difference of arrival. Smart Mater. Struct. 2018, 27,
035023. [CrossRef]
38. Narayanan, A.; Subramaniam, K.V. Experimental evaluation of load-induced damage in concrete from
distributed microcracks to localized cracking on electro-mechanical impedance response of bonded PZT.
Constr. Build. Mater. 2016, 105, 536–544. [CrossRef]
39. Wang, F.; Ho, S.C.M.; Huo, L.; Song, G. A novel fractal contact-electromechanical impedance model for
quantitative monitoring of bolted joint looseness. IEEE Access 2018, 6, 40212–40220. [CrossRef]
40. Wang, B.; Huo, L.; Chen, D.; Li, W.; Song, G. Impedance-based pre-stress monitoring of rock bolts using a
piezoceramic-based smart washer—A feasibility study. Sensors 2017, 17, 250. [CrossRef]
41. Tibaduiza, D.A.; Mujica, L.E.; Rodellar, J.; Güemes, A. Structural damage detection using principal
component analysis and damage indices. J. Intell. Mater. Syst. Struct. 2016, 27, 233–248. [CrossRef]
Appl. Sci. 2019, 9, 687 19 of 19
42. Wang, F.; Huo, L.; Song, G. A piezoelectric active sensing method for quantitative monitoring of bolt
loosening using energy dissipation caused by tangential damping based on the fractal contact theory.
Smart Mater. Struct. 2017, 27, 015023. [CrossRef]
43. Zhang, C.; Yu, X.; Alexander, L.; Zhang, Y.; Rajamani, R.; Garg, N. Piezoelectric active sensing system for
crack detection in concrete structure. J. Civil Struct. Health Monit. 2016, 6, 129–139. [CrossRef]
44. Banon, H.; Biggs, J.M. Seismic damage in reinforced concrete frames. J. Struct. Eng. 1981, 107, 1713–1729.
45. Roufaiel, M.S.L.; Meyer, C. Analytical modeling of hysteric behavior of RC frames. J. Struct. Eng. 1987, 113,
429–444. [CrossRef]
46. Colombo, A.; Negro, P. A damage index of generalized applicability. Eng. Struct. 2005, 27, 1164–1174.
[CrossRef]
47. Cho, H.C.; Lee, D.H.; Ju, H.; Park, H.C.; Kim, H.Y.; Kim, K.S. Fire damage assessment of reinforced concrete
Structures using fuzzy theory. Appl. Sci. 2017, 7, 518. [CrossRef]
48. Park, Y.J.; Ang, H.S. Mechanistic seismic damage model for reinforced concrete. J. Struct. Eng. 1985, 111,
722–739. [CrossRef]
49. Mehanny, S.S.F.; Deierlein, G.G. Seismic damage and collapse assessment of composite moment frames.
J. Struct. Eng. 2001, 127, 1045–1059. [CrossRef]
50. Erduran, E.; Yakut, A. Component damage functions for reinforced concrete frame structures. Eng. Struct.
2007, 29, 2242–2253. [CrossRef]
51. Zheng, S.S.; Deng, G.Z.; Yang, Y.; Yu, M.H.; Zhang, J.F. Experimental study of bond-slip performance between
steel and concrete in SRC structures. Eng. Mech. 2003, 20, 63–69.
52. Wang, B.; Zheng, S.S.; Guo, X.F.; Li, L. Experimental research on mechanical behavior of SRHSHPC frame
columns under cyclic loading. J. Build. Struct. 2011, 32, 117–126.
53. National Standard of the People’s Republic of China: Code for Design of Concrete Structures; GB 50010-2010;
China Architecture & Building Press: Beijing, China, 2010.
54. National Standard of the People’s Republic of China: Code for Design of Composite Structures; JGJ 138-2016;
China Architecture & Building Press: Beijing, China, 2016.
55. Industry Standard of the People’s Republic of China: Specification of Testing Methods for the Earthquake Resistant
Building; JGJ101-96; China Architecture & Building Press: Beijing, China, 1996.
56. Liu, Q.Y.; Kasai, A.; Usami, T. Two hysteretic models for thin-walled pipe-section steel bridge piers.
Eng. Struct. 2001, 23, 186–197. [CrossRef]
57. Sezen, H.; Chowdhury, T. Hysteretic model for reinforced concrete columns including the effect of shear and
axial load failure. J. Struct. Eng. 2009, 135, 139–146. [CrossRef]
58. Zheng, S.S.; Wang, B.; Li, L.; Hou, P.J. Study on seismic damage of SRHSC frame columns. Sci. China Tech.
Sci. Ser. E 2011, 54, 2886–2895. [CrossRef]
59. Zheng, S.S.; Zhang, L.; Li, L.; Hu, Y.; Hu, C.M. Experimental research on seismic behavior of steel reinforced
high strength concrete frame columns. J. Build. Struct. 2012, 33, 124–132.
60. Aslani, F.; Nejadi, S. Cyclic constitutive model for high-strength concrete confined by ultra-high-strength
and normal-strength transverse reinforcements. Aust. J. Struct. Eng. 2012, 12, 59–72.
61. Aslani, F.; Jowkarmeimandi, R. Stress–strain model for concrete under cyclic loading. Mag. Concr. Res. 2012,
36, 330–347. [CrossRef]
© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).

Applied Sciences: Hysteretic Behavior of Steel Reinforced Concrete Columns Based On Damage Analysis

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Applied Sciences: Hysteretic Behavior of Steel Reinforced Concrete Columns Based On Damage Analysis

Uploaded by

Copyright:

Available Formats

applied

Appl. Sci. 2019, 9, 687; doi:10.3390/app9040687 www.mdpi.com/journal/applsci

2. Description of Test Program

Table 1. Design parameters of specimens.

Cross Compression Stirrups

Table 2. Mechanical properties of the steel and rebars.

Steel Shape Stirrups Longitudinal

Yield strength f y (MPa) 319.7 312.4 397.5 354.5 386.3

150 150 150

750 750 750

300 300 300

300 300 300

Figure 2. Test setup. Bottom beam

Figure 3. Photo of testing apparatus.

Appl. Sci. 2016, 6, x FOR PEER REVIEW 6 of 20

150 150 150

100 100 100

-50 -50 -50

-100 -100 -100

-150 Hysteretic curve -150 Hysteretic curve -150 Hysteretic curve

150 150 150

100 100 100

-50 -50 -50

-100 -100 -100

-150 -150 Hysteretic curve -150 Hysteretic curve

-150 Hysteretic curve

Appl. Sci. 2016, 6, x FOR PEER REVIEW 8 of 20

(1) Large eccentric Compression Columns

Figure 8. Calculation diagram of large eccentric compression at the

where N is the axial expression force, f s kis=the

4.2.2. Peak load Pm and peak displacement ∆m

4.2.3. Ultimate Load Pu and Ultimate Displacement ∆u

5. Damage-Based Hysteretic Analysis

5.1. Simplification of the Hysteresis Loop

5.2. Proposed Damage Index

5.2. Proposed Damage Index

Eu,i = A + Be−α , (20)

c = 5.69 + 0.87 ln ρv + 0.056λ + 10.46ρs − 2.1n. (21)

5.3. Cyclic Strength and Stiffness Degradation in the Hysteretic Model

5.3.1. Damage-Based Cyclic Degradation Index

∆Di > 1 − Di−1 . (23)

5.3.2. Degradation Analysis of Strength and Stiffness

(4) (4) Degradation

-100 -100 Test value

-100 -100 Test value

You might also like