15
Research Article Comparative Study of Seismic Behavior between Monolithic Precast Concrete Structure and Cast-in-Place Structure Chao-gang Qin , 1 Guo-liang Bai , 1 Ya-zhou Xu , 1 Ning-fen Su, 1 and Tao Wu 2 1 School of Civil Engineering, Xi’an University of Architecture and Technology, Shaanxi 710055, China 2 School of Civil Engineering, Chang’an University, Shaanxi 710061, China Correspondence should be addressed to Guo-liang Bai; [email protected] Received 15 June 2017; Accepted 16 August 2017; Published 2 January 2018 Academic Editor: Stefano Sorace Copyright © 2018 Chao-gang Qin et al. is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. We doubt whether the monolithic precast concrete structure could be designed as the cast-in-place structure in high seismic intensity area. To solve the puzzle, the 1/5 scaled monolithic precast concrete structure model and cast-in-place structure model were designed and tested by shake table. Comparative analysis between them was made to better understand their seismic behavior. Based on the experimental results, the failure pattern and mechanism were different, which was concentrated damage in coupling beam and then extended to shear walls of CIPS, and the weak connections presented cracks between precast elements besides the damage coupling beam of MPCS. e natural frequency of MPCS possessed a typical feature for the weakness of connections, which was the initial one greater than that of CIPS and decreased fast aſter the first waves with PGA of 0.035g. Acceleration amplifying factors presented variation trend under the different earthquake waves. e distribution of seismic response presented linearity along the height of models in plastic stage and turned into nonlinearity later for severe damage. In general, the MPCS and CIPS had similar seismic responses, except typical characteristics. And they were proven to have better seismic performance without collapse under the high-intensity earthquake waves. 1. Introduction Precast concrete structure is constructed with prefabricated elements made in factory, which is popularly used for resi- dential buildings, industrial buildings, and public buildings, such as apartment, parking, and stadium. It possesses the high-quality precast elements: the speed of erection and freedom in the architectural shape of the members. However, the integrity and security of connections between the precast units are significant for global structure, especially under the earthquake excitations. As is well known, shear wall structure is an effective lateral resisting system for high-rise residential buildings [1, 2] in earthquake protection region. e shear wall precast elements along story height are connected to form the lateral resisting system. In order to keep the validity of the horizontal connection, various ways are taken to connect the longitudinal reinforcements, such as grouted sleeve, posttensioning, and splice sleeves [3–7]. Further, the isolated shear walls with different horizontal connections were tested considering the contact surface and the mentioned connection of longitudinal rebar [8– 10]. Vertical connection located between the story precast elements was investigated by Vaghei et al. [11]. Nowadays, the improved grouted sleeve is an effective connection of longi- tudinal reinforcements, and a monolithic vertical connection between the story precast elements is conducted to enhance their integrality. Namely, the vertical connection is the edge components of shear wall of cast-in-place structure. e properties of prevalent connections of precast ele- ments and the overall structure were conducted with pseudo- static test and pseudodynamic test [12–15], whereas the tests did not consider the influence of the duration of earthquake waves. One kind of precast wall structure large-panel precast concrete building of three single-bay simple walls was tested by Oliva et al. [16] and a 3-story model structure was tested by Lee et al. [17]. And the seismic properties of a 1/4 scaled precast structure with high damping rubber bearings were studied by Wang et al. [18]. Nevertheless, some research reports detected that precast structure was not with excellent seismic behavior during the previous earthquake for failures Hindawi Shock and Vibration Volume 2018, Article ID 7029287, 14 pages https://doi.org/10.1155/2018/7029287

Comparative Study of Seismic Behavior between Monolithic

  • Upload
    others

  • View
    2

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Comparative Study of Seismic Behavior between Monolithic

Research ArticleComparative Study of Seismic Behavior between MonolithicPrecast Concrete Structure and Cast-in-Place Structure

Chao-gang Qin 1 Guo-liang Bai 1 Ya-zhou Xu 1 Ning-fen Su1 and TaoWu2

1School of Civil Engineering Xirsquoan University of Architecture and Technology Shaanxi 710055 China2School of Civil Engineering Changrsquoan University Shaanxi 710061 China

Correspondence should be addressed to Guo-liang Bai guoliangbai126com

Received 15 June 2017 Accepted 16 August 2017 Published 2 January 2018

Academic Editor Stefano Sorace

Copyright copy 2018 Chao-gang Qin et alThis is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

We doubt whether the monolithic precast concrete structure could be designed as the cast-in-place structure in high seismicintensity area To solve the puzzle the 15 scaled monolithic precast concrete structure model and cast-in-place structure modelwere designed and tested by shake table Comparative analysis between themwasmade to better understand their seismic behaviorBased on the experimental results the failure pattern and mechanism were different which was concentrated damage in couplingbeam and then extended to shear walls of CIPS and the weak connections presented cracks between precast elements besides thedamage coupling beamofMPCSThenatural frequency ofMPCS possessed a typical feature for the weakness of connections whichwas the initial one greater than that of CIPS and decreased fast after the first waves with PGA of 0035 g Acceleration amplifyingfactors presented variation trend under the different earthquake waves The distribution of seismic response presented linearityalong the height of models in plastic stage and turned into nonlinearity later for severe damage In general the MPCS and CIPShad similar seismic responses except typical characteristics And they were proven to have better seismic performance withoutcollapse under the high-intensity earthquake waves

1 Introduction

Precast concrete structure is constructed with prefabricatedelements made in factory which is popularly used for resi-dential buildings industrial buildings and public buildingssuch as apartment parking and stadium It possesses thehigh-quality precast elements the speed of erection andfreedom in the architectural shape of the members Howeverthe integrity and security of connections between the precastunits are significant for global structure especially under theearthquake excitations As is well known shear wall structureis an effective lateral resisting system for high-rise residentialbuildings [1 2] in earthquake protection region

The shear wall precast elements along story height areconnected to form the lateral resisting system In order tokeep the validity of the horizontal connection various waysare taken to connect the longitudinal reinforcements suchas grouted sleeve posttensioning and splice sleeves [3ndash7]Further the isolated shear walls with different horizontalconnections were tested considering the contact surface

and the mentioned connection of longitudinal rebar [8ndash10] Vertical connection located between the story precastelements was investigated by Vaghei et al [11] Nowadays theimproved grouted sleeve is an effective connection of longi-tudinal reinforcements and a monolithic vertical connectionbetween the story precast elements is conducted to enhancetheir integrality Namely the vertical connection is the edgecomponents of shear wall of cast-in-place structure

The properties of prevalent connections of precast ele-ments and the overall structure were conducted with pseudo-static test and pseudodynamic test [12ndash15] whereas the testsdid not consider the influence of the duration of earthquakewaves One kind of precast wall structure large-panel precastconcrete building of three single-bay simple walls was testedby Oliva et al [16] and a 3-story model structure was testedby Lee et al [17] And the seismic properties of a 14 scaledprecast structure with high damping rubber bearings werestudied by Wang et al [18] Nevertheless some researchreports detected that precast structure was not with excellentseismic behavior during the previous earthquake for failures

HindawiShock and VibrationVolume 2018 Article ID 7029287 14 pageshttpsdoiorg10115520187029287

2 Shock and Vibration

of welded and poorly constructed connections [19 20]Further experimental research is clearly needed to fill thegap in the knowledge of seismic behavior of precast concretestructure And large-scale shake table test is a reliablemethodto research the dynamic seismic response of precast concretestructure

This paper presents a comparative shaking table testprogram realized on two 1 5 scaledmodels of a 12-story shearwall structure to understand the dynamic seismic responseof precast concrete structure One is cast-in-place structure(CIPS) and the other ismonolithic precast concrete structure(MPCS) The prototype structure was designed in two baysand two spans in accordance with the provisions of code [21]and the splitting design of MPCS was according to the code[22] and design drawings sheets [23] Based on the testingresults the dynamic characteristics of the two models suchas frequency damping ratio and mode shape are evaluatedthrough the white-noise test In comparison the failurepattern and mechanism seismic force response story shearstory displacement and interstory drift will be intensivelyexplored analyzed and discussed Finally the comprehensiveunderstanding of seismic performance of MPCS and CIPSwill be revealed especially the earthquake response of theMPCS in its entirety

2 Experimental Program

21 Model Design

211 Similitude Relationship As the performance parametersof shake table and condition of lift in laboratory scaledmodel was applied in shake table test Prototype structurewas designed observing the provisions of Chinese code [21]And the reduced scale design for the models was establishedfrom Buckinghamrsquos Pi theorem [24] The similar constants ofgeometry stress and acceleration were first defined as 0202 and 1 respectively [25] And then the other parameterswere deduced by the similitude rules and summarized inTable 1 The CIPS and MPCS had the same similitude rulesFurthermore the models were designed as elastic-plasticmodels to observe the plastic behavior under the high-intensity earthquake waves [26]

212 Material Design As the scaled parameters of physicsthe stress and elastic modulus of the model materialdecreased to 20 of that of the concrete of prototypestructure The microconcrete was adopted as model materialfor the limitation of coarse aggregate Shen et al pro-posed ceramsite powder coal ash or pumice as blendingagent could reduce the elastic modulus of microconcrete[27] So the gypsum was mixed to the microconcreteAfter testing the model material was a compound withcement fine aggregate coarse aggregate water gypsum =1 364 364 093 05 (inweight ratio)Theultimate strengthof the microconcrete was 894MPa and the elastic modulusof that was 729GPa consistently with the similitude constant02 in contrast with C40 concrete Galvanized iron wire wasused to replace the reinforcements following the similar rulesof internal forces [25]

213 Plane Design Considering the architectural space thephysical dimension and connection plane of the precastelements of high-rise residence in practice the model wasa two-bay two-span twelve-story shear wall structure thatwas regular in plan and elevation The CIPS and MPCSscaled models were 1800mm times 1800mm in plane and witha constant story height of 600mm The distance of spanwas 1100mm and 700mm in 119883 direction and 900mm and900mm in 119884 direction The thickness of the shear wall andcoupling beam was 40mm and that of the slab was 30mm Itcontained three kinds of coupling beamswith different spans500mm 300mm and 160mm respectively Figure 1 displaysthe plane view of the models

The MPCS model included three kinds of cast-in-placeconnections binding precast elements ldquoLrdquo type ldquoTrdquo type andldquo+rdquo type to forman entirety part in each story and three kindsof precast concrete shear walls (PCSW) PCSW-1 PCSW-2and PCSW-3The three connections represented the externalPCSW connected in corner external and internal PCSWconnected in side and internal PCSW connected inside ofprecast structure Moreover the cast-in-pace connectionswere corresponding to the edge components of shear wallof the CIPS model which were separated from the precastelement in the factory and then made with cast-in-placeconcrete after the precast elements were installed Namelythe cast-in-pace connections and the PCSW composed theshear wall of CIPS The design parameters model materialsand loading program of MPCS were the same as those ofCIPS However the models had diverse construction tech-niques The details reinforcements of the cast-in-place con-nections or the edge components and PCSW are exhibited inFigure 2

22 Precast and Construction Details The structural mea-sures of CIPS were complying with provisions of the code[21] The CIPS model was constructed in the general con-struction technique including assembling reinforcementtemplate installation pouring concrete and maintenanceHowever the precast units are fabricated in factory trans-ported to the construction site and lifted by crane and com-bined together with the cast-in-place concrete later formingthe monolithic precast concrete structure in practice In thisproject a compromised construction method was adoptedThe model precast concrete shear walls were fabricated inthe laboratory The reinforcements were connected by thefollowing ways The upper protruding bar extended throughthe grouted sleeve to connect the next PCSW and the side onehooked the longitudinal reinforcements in CIP connectionand put on additional stirrups later Between the upper andlower PCSW the cast-in-place floor replaced the assemble-monolithic concrete composite slab with similar stiffness forconvenience The remaining CIP connection and floor werepoured after the PCSW was cured in 48 hours Thereforethe coupling beam was precast superposed one in MPCSand that in CIPS was integrity The model material wasmicroconcretemixed gypsumwith 894MPaTheproceduresof MPCS and CIPS are shown in Figure 3 They were curedat normal temperature for 28 days and tested by shake tableunder earthquake waves

Shock and Vibration 3

Table1Similitude

scalep

aram

eterso

fCIPSandMPC

S

Item

parameter

Geometry

leng

thPh

ysics

Load

Dyn

amiccharacteris

tics

Elastic

mod

ulus

Stress

Poiss

onrsquosratio

Strain

Mass

Massd

ensity

Frequency

Acceler

ation

Form

ula

119878 119871119878 119864

119878 120590=119878 119864

119878 ]119878 120576=119878 120590 119878 119864119878 119872=119878 119864119878 1198712

119878 119886119878 120588=119878 120590 119878 119871119878 119886

119878 119891=119878 11988612

119878 11987112

119878 119886=119878 119892

Relationship

02

02

02

11

0008

12236

1Re

mark

Con

trolsize

Con

trolm

aterial

Con

trolshaking

tabletest

Note(120588119892119897119864) 119898=(120588119892119897119864) 119901ldquo119898rdquom

eans

mod

elstr

uctureldquo119901rdquom

eans

prototypes

tructure

4 Shock and Vibration

1 2 3

A

B

C

Coupling beam 1

PCSW-1

CIP connection-T

X

Y

N

CIP connection-L

CIP connection-+

PCSW-2 PCSW-3

Coupling beam 2 Coupling beam 3

1 2 3

A

B

C

Coupling beam 1

SW-1

X

Y

N

SW-2 SW-3

Coupling beam 2 Coupling beam 3

Cast-in-place concrete partsPrecast concrete shear wall

Opening in the shear wall

Cast-in-place structure (CIPS) Monolithic precast concrete structure (MPCS)

Edge component Edge component

80 220 500 220 160 190 80 20

8022

030

022

016

022

030

022

080

2090

090

020

160 1901800

80 220 500 220 160 190 80 20

8022

030

022

016

022

030

022

080

2090

090

020

160 1901800

Figure 1 Plane view of CIPS and MPCS

In order to supplement the missing gravity and nonstruc-tural elements iron blockswere used as the artificialmass andevenly fixed on each floor of the twomodels about 156 tonesThe total mass of each model reached 136 tones includinggrade beams and the height of the two models was 756mwhich conformed to the capacity limitation of the shakingtable system

23 Test Procedure It is well known that the condition of soilat the site is one of the significant factors in choosing theseismic inputs for shaking table test The soil layer equivalentcutting wave speed and the thickness of overlay soil layerdetermine the site classification The type-II soil site wasdefined in the Code for Seismic Design of Buildings [21]which was the condition soil of this project In comparisonwith seismic design response spectra Superstition Hillsearthquake wave (B-WSM) Kocaeli (Turkey) earthquakewave (DZC) and El Centro earthquake wave (ELW) wereselected as the ground excitations and inputted by thementioned sequenceThewaves were chosen from the PacificEarthquake Engineering Research Center (PEER) The testswere performed with unidirectional and bidirectional earth-quake waves with the ratio of PGA 1 085 to evaluate overallseismic performance of CIPS andMPCSThe test program ofthe shake table test included eight phases and the peak groundacceleration (PGA) is 0035 g 007 g 014 g 022 g 040 g062 g 070 g and 080 g in each phase respectivelyThe PGAof 070 g and 080 g were inputted to observe the nonlinearbehavior of them After each teat phase the white noise

with PGA of 0035 g was inputted to detect the dynamiticcharacteristics of the models

To monitor the earthquake response of the two models32 single-axis accelerometers including two on the shaketable two on the grade beam twenty-four on each floor in119883 and 119884 directions and four on the diagonal points in the12th floor were set to record the horizontal acceleration Atotal of 12 displacement transducers and 12 velocity vibrationsensors were installed to each floor The arrangement of testinstruments is exhibited in Figure 4 Figure 5 displays themodels on the shake table

3 Failure Pattern and Mechanism

Cracks and damage to the models were traced along withthe increasing PGA With the PGA of 0035 g diagonalmicrocracks on the coupling beam were merely observedin main direction of CIPS and those of MPCS were foundin 119883 and 119884 directions of shear wall arranged from 1st to4th As the PGA increased from 007 g to 022 g cracks onthe CIPS were extended along the corner of coupling beamand some new diagonal microcrack occurred In the phasesthe existing cracks of MPCS penetrated the coupling beamand meanwhile new microcracks appeared The diagonalmicrocrackswere themain pattern of themodels in this stage

In the next stage the models presented different crackpatterns After the PGA of 040 g the cracks of CIPSconcentrated rapidly in the end of coupling beams suchas the biggest span-depth ratio one with decreased section

Shock and Vibration 5

ldquoLrdquo type

ldquoTrdquo type

Floor

Sleeve

Longitudinalreinforcement

Additional stirrup

Protruding bar

Grouting materials

Cast-in-place concreteconnection of ldquoLrdquo type Roughened surface

External precastconcrete shear wall

Additional stirrup

10040 60

5

4060

100

5

45

45

External precastconcrete shear wall

12(10)1430

Protruding bar

Longitudinalreinforcement

Cast-in-place concreteconnection of ldquoLrdquo type

External precastconcrete shear wall

Floor

Sleeve

Longitudinalreinforcement

Grouting materials

Additional stirrupProtruding bar

Cast-in-place concreteconnection of ldquoTrdquo type Roughened surface

External precastconcrete shear wall 604060

4060

45

45

5

5

16(10)1430

Additional stirrup Protruding bar

Longitudinalreinforcement

Cast-in-place concreteconnection of ldquoTrdquo type

External precastconcrete shear wall External precast

concrete shear wall

Internal precastconcrete shear wall

ldquo+rdquo type

Floor

Sleeve

Longitudinalreinforcement

Grouting materials

Additional stirrupProtruding bar

Cast-in-place concreteconnection of ldquo+rdquo type Roughened surface

Internal precastconcrete shear wall

60 40 60

45

45

5

5

6040

60

20(10)1430

Additional stirrup

Protruding bar

Longitudinalreinforcement

Cast-in-place concreteconnection of ldquo+rdquo type

Internal precastconcrete shear wall

Internal precastconcrete shear wall

(a) Reinforcement of CIP connections

Figure 2 Continued

6 Shock and Vibration

PCSW-1 PCSW -3

PCSW-2

30 60 60 50

27050 60 60 30

270

3030 30 30

90

570

250

6(10)1430

1440

14204(8)

120 740 120

10 10

Longitudinal reinforcement

Protruding bar

Longitudinal reinforcement

Protruding bar

Verticaland levelreinforcement

Coupling beam 1

Edge component

Protruding bar

90

570

30 60 60 50

230

3050 60 60 30

230

30

250

6(10)1430

1440

14204(8)

10 10

120 540 120

Longitudinal reinforcement

Protruding bar

Longitudinal reinforcement

Protruding bar

Verticaland levelreinforcement

Coupling beam 3

Protruding bar

Edge component

30 60 60 50270

50 60 60 30

270

30

90

570

30 30 30

250

6(10)1430

1440

14204(8)

10 10

120 940 120

Longitudinal reinforcement

Protruding bar

Longitudinal reinforcement

Protruding bar

Verticaland levelreinforcement

Coupling beam 2

Protruding bar

Edge component

8lowast60=

480

8lowast60=

480

8lowast60=

480

(b) Reinforcement of PCSW

Figure 2 Configuration and reinforcements of the components (unit mm)

for the plastic hinge and the smallest span-depth ratio onewith diagonal cracks On the contrary diagonal cracks ofMPCS extended slowly and the horizontal fracture betweenthe PCSW and the CIP floor as the special form occurredat the same time With the increasing seismic energy thenew cracks were successively added along the height of themodels The cracks of CIPS were like the diagonal cracksoccurring with the PGA of 0035 g and those of MPCS werehorizontal cracks occurring in horizontal connections Inorder to observe the nonlinear behavior of them the PGA of

080 gwas inputted In this phase the vertical cracks along theedge component and the group cracks in the shear wall in 4thstory occurred in CIPS We considered that the weak story ofCIPS was the 4th story The horizontal crack extended to thecast-in-place concrete and the vertical crack appeared in thecontact of precast and cast-in-place connection in 2nd and3rd stories in MPCSThe overall crack patterns of models areexhibited in Figure 6

In general the diagonal cracks were presented in the end-ing coupling beams of CIPS and MPCS The phenomenon is

Shock and Vibration 7

CIP connection

Coupling beam

PCSW elementCIP connection PCSW element

CIP floor

CIPS MPCS

Construction procedures of CIPS

Construction procedures of MPCS

Grouted sleeves

Figure 3 Construction procedures of CIPS and MPCS

Table 2 Natural frequency of CIPS and MPCS

PGAg

FrequencyHzFirst frequency Second frequency

119883-direction 119884-direction 119883-direction 119884-directionMPCS CIPS MPCS CIPS MPCS CIPS MPCS CIPS

Initial 619 513 7 569 2500 2075 2794 23250035 g 481 512 513 563 2056 2013 2244 235007 g 480 512 506 55 2006 1988 2200 2325014 g 469 456 494 537 1969 1988 2106 2206022 g 426 413 444 481 1825 1813 1981 2094040 g 279 281 375 375 1419 1356 1625 1656062 g 231 194 300 269 1163 1006 1363 1300080 g 186 155 288 254 938 942 1318 1288

the favored state They possessed different energy dissipationmechanisms under the intensity earthquake waves The cou-pling beam acted as the first line of energy dissipation Whilethe coupling beam formed a plastic hinge the shear wallchanged to a single-slice wall to dissipate energy as the secondline to forbid collapse in CIPS Besides the coupling beamthe relatively weak connections between the precast elementsplayed the new way to dissipate energy in MPCS whichwere horizontal cracks first and then the vertical cracksFurthermore they defended the lateral resisting system insome degree

4 Earthquake Response Analysis

41 Dynamic Characteristics The dynamic characteristics ofstructure include natural frequency stiffness damping ratioand mode shape They could be deduced from the whitenoise inputted after each test phase by the transfer functionThe first and second natural frequency in 119883 direction and119884 direction are shown in Table 2 Also the stiffness could becalculated with the frequency and it is presented in Figure 7[28] Initial frequency of MPCS was bigger than that ofCIPS and the same to the initial stiffness We assumed that

8 Shock and Vibration

D11DDDD11D11D11D11DDDD 11D

A1 D1

A2 V1

A3 D2

A4 V2

A5 D3

A6 V3

A7 D4

A8 V4

A9 D5

A10 V5

A11 D6

A12 D7

A13 D8 V6

A14

A15

A16

A17

A18

A19

A20

A21

A22

A23

A24

A25

A26

V7

V9

D9

V11

V8

V10

D10

V12

D12

Y

X

Figure 4 Arrangement of accelerometers displacement transduc-ers and velocity vibration sensors

CIPS MPCS

Figure 5 General view of CIPS and MPCS

the adjunction grouted sleeves and the additional stirrupsin the connections deduced the phenomenon As the firstearthquake excitation was inputted the natural frequency ofMPCS decreased about 20 which could be caused by theshrinkage and microcracks in connections as initial damage[29] With the increasing energy the influence of the initialdamage was not the main factor And then the two modelswith approximately equal frequency presented in failure stateThe curves of stiffness degradation of CIPS reduced graduallywith the increasing PGA In contrast withCIPS that ofMPCS

CIPS MPCS

Figure 6 Failure patterns of CIPS and MPCS

00

01

02

03

04

05

06

07

08

09

10

11

Stiff

ness

ratio

MPCS-XMPCS-Y

CIPS-XCIPS-Y

022 g014 g007 g 040 g0035 g 062 gPrimePGA (g)

Figure 7 Stiffness ratio versus frequency in119883 and 119884 directions

decreased evidently in the first phase and then diminishedslowly from PGA of 0035 g to 014 g Finally they hadsimilar residual stiffnesses The diverse tendency might beunderstood by the mentioned failure pattern

Damping ratio reflects the dissipation capacity of struc-ture As is shown in Figure 8 the damping ratio increasedgradually following the PGA In the first phase it was 42And then the damping ratio increased slowly before the PGAof 040 g which arranged from 42 to 50 The averagedamping ratio of each phase changed from 42 to 82in the loading process which was in the scope of cast-in-place concrete structure Nevertheless the energy dissipationmechanismofCIPS andMPCSwas different for the order anddistribution of cracks

The first and second mode shapes of the models aredescribed in Figure 9 In general the first-order mode shape

Shock and Vibration 9

000

001

002

003

004

005

006

007

008

009

010

011

Dam

ping

ratio

MPCS-XMPCS-Y

CIPS-XCIPS-Y

014 g 022 g007 g0035 g 040 g 062 gPrimePGA (g)

Figure 8 Damping ratios of CIPS and MPCS

exhibited bending deformation characteristics and the lateralstiffness was uniform distribution along the height of themodel The shapes of them were bending to 119909-axis littleby little The phenomenon would be caused by high-ordermodes And the trend of that of CIPSwasmore obvious in thePGAof 040 g and 062 g since the severe damage occurred inthe shear wall The second-order mode shapes of them weresimilar And the maximummode coefficient of second modeshape was in positon of 4th floor

42 Acceleration Response The ratio of measured accelera-tion to the corresponding input peak ground accelerationis called the acceleration amplifying factor It reflects thedynamic response of structure under earthquake excitationsThe acceleration amplifying factors along the height of themodels are described in Figure 10(a) under the B-WSMDZCand ELW ground motions for different seismic levels Obvi-ously that of the two models was gradually increased alongthe height of themodel in each test phaseWith the increasingPGA arranged from 007 g to 062 g the whole developmenttrend of that decreased gradually implying the progressivedegradation of structural stiffness In each test phase themodels presented different dynamic responses under thedifferent earthquake excitations In elastic stage with thePGA of 007 g and 014 g the CIPS had greater response toB-WSM and ELW than to DZC The MPCS had uniformresponse to the three excitations The earthquake waves withdifferent frequency-spectrum characteristics would lead tothis phenomenon The greatest response of CIPS occurredunder the ELW with PGA of 022 g and that of MPCSoccurred in the B-WSM with PGA of 040 g The diversitiesof twomodels about the acceleration amplifying factors couldbe caused by the connections between the precast elementsThe factors of CIPS and MPCS reduced from 546 to 319and from 523 to 308 in the final test phase respectivelyAs shown in Figure 10(b) the distribution feature was more

regular under the test phases with PGA from 0035 g to 014 gWhen the two models were subjected to severe damage theinfluence of high-order vibration modes increased graduallyand the acceleration amplifying factors at somemeasurementpoints no longer conformed to the distribution [30]

43 Earthquake Action The seismic force distribution char-acteristic of structure is very important reference for theaseismic design and application of the MPCS and CIPS Themaximum seismic force of the 119894th floor is derived as follows

119865119894max = 1198981198941003816100381610038161003816 119894 (119905) + 0 (119905) max

1003816100381610038161003816 (1)

where 119865119894max is the maximum seismic force 119898119894 is the lumpedmass of the 119894th floor 119894(119905) is the acceleration response of the119894th floor relative to the ground at the time of 119905 and 0(119905) is theground acceleration at the time of 119905

According to (1) the maximum seismic forces of CIPSand MPCS are presented and compared in Figure 11 Asillustrated in Figure 11(a) the results of them were presentedin linear behavior following the increasing earthquake waveswith PGA from 0035 g to 014 g The curves were increasedprogressively along the height of models They could reflectthe actual seismic force distribution in some extent of struc-ture in the elastic stage and the high-order modes could beignored in this time And then the cracks expanded graduallyin different ways of the twomodels and the influence of high-order modes worked The models entered into plastic stagewith obvious nonlinearity The maximum seismic forces ofthe CIPS emerged in the middle of model such as 5th 6thand 8th Those of MPCS were in 4th 5th and 8th In theplastic stage with PGA from 022 g to 062 g the distributionof seismic forces changed for the high-order modes

The interstory shear forceswere calculated by (1) and theyare illustrated along the height of models in Figure 11(b) Theinterstory shear force was increased gradually with the PGAand decreased along the height of model structure In elasticstage the distribution of the interstory shear of CIPS andMPCS showed similar rules like inverted triangle In plasticstage the vibration trend was effected by the high-ordermodes and interstory shear force was not strictlymeeting thedistribution pattern

44 Displacement Response The maximum story displace-ments of the CIPS and MPCS along the model heightobtained from the B-WSM DZC and ELW with thePGA from 0035 g to 062 g are compared and depicted inFigure 12(a) Since the shear wall structure is the effectivelateral resisting system [1] themaximum story displacementsof the models were small in 12th story arranged from 113mmto 756mm of CIPS and from 106mm to 699mm of MPCSin the elastic stageThe stage contained the PGA from 0035 gto 014 g With increasing the intensity of earthquake wavesthe two models demonstrated different damage phenomenadescribed in the above paragraph The concentrated damagein shear wall of 4th story of CIPS caused bigger displacementthan that of MPCS The bigger displacement formed thebigger torsion angle of floor and then brought about themuchbigger displacement in upside structure of CIPS And the

10 Shock and Vibration

0123456789

101112

Floo

r

01 02 03 04 05 06 07 08 09 10 1100Mode coefficient of first mode shape

0123456789

101112

Floo

r

01 02 03 04 05 06 07 08 09 10 1100Mode coefficient of first mode shape

minus08 minus06 minus04 minus02minus10 02 04 06 08 1000Mode coefficient of second mode shape

minus06 minus04 minus02minus08minus10 02 04 06 08 1000Mode coefficient of second mode shape

0123456789

101112

Floo

r

0123456789

101112

Floo

r

MPCS-initialMPCS-0035 gMPCS-007 gMPCS-014 g

MPCS-022 gMPCS-040 gMPCS-062 g

MPCS-initialMPCS-0035 gMPCS-007 gMPCS-014 g

MPCS-022 gMPCS-040 gMPCS-062 g

CIPS-initialCIPS-0035 gCIPS-007 gCIPS-014 g

CIPS-022 gCIPS-040 gCIPS-062 g

CIPS-initialCIPS-0035 gCIPS-007 gCIPS-014 g

CIPS-022 gCIPS-040 gCIPS-062 g

Figure 9 Mode shapes of CIPS and MPCS

maximumdestructive story driftwas in 4th floorMeanwhilethe two models showed obviously nonlinear behavior withthe PGA of 040 g and 062 g The shape of maximum storydisplacements of them was similar to the first mode shapeThe maximum story displacements were 4867mm in CIPSand 4198mm in MPCS

The maximum interstory drift ratios were calculated andpresented in Figure 12(b) The maximum value of story driftof CIPS was 11005 in 5th story under the DZC earthquakewave with the PGA of 070 g and that of MPCS was 11020in 5th story under the same wave They met the provisionabout the maximum story drift in elastic stage in area of 8-degree seismic fortification intensity in code of GB 50011-2010 [21] In the stage from PGA of 0035 g to 014 g thestory drift increased progressively The seismic design codeasks for the limiting value of plastic interstory drift ratio toforbid collapse We checked the story drift under the PGA of040 g The values of CIPS and MPCS were 1121 and 1127

which met the limiting value in rare earthquake action of 8-degree seismic regions Under the excitation of PGAof 062 gthe values exceeded the limiting value in rare earthquakeaction of 9-degree seismic regions And then the moreintensive earthquakewaves with PGAof 080 gwere inputtedthe models presented better seismic behavior withoutcollapse

5 Conclusion

A comparative seismic study between CIPS and MPCS wasconducted by shaking table test which contained a 15 scaled12-story CIPS model and MPCS model The experimentalresults of dynamic characteristic failure pattern and mecha-nism and seismic response of the models were discussed andcompared with each other to better understand their seismicbehavior Based on the intensive analysis of test results thefollowing conclusions are derived

Shock and Vibration 11

PGA of 062 gPGA of 040 gPGA of 022 gPGA of 014 gPGA of 007 g PGA of 062 gPGA of 040 gPGA of 022 gPGA of 014 gPGA of 007 g

EE E EB B B BDD DD

E(LW

)

D(Z

C)

B(-W

SM)

Peak ground acceleration of CIPS (g)

E EEEB B BB DDD D

E(LW

)

D(Z

C)

B(-W

SM)

Peak ground acceleration of MPCS (g)

12th floor11th floor10th floor9th floor8th floor

7th floor6th floor5th floor4th floor3rd floor

2nd floor1st floorFloor

12th floor11th floor10th floor9th floor8th floor

7th floor6th floor5th floor4th floor3rd floor

2nd floor1st floorFloor

081216202428323640444852566064

Acce

lera

tion

ampl

ifyin

g co

effici

ent

081216202428323640444852566064

Acce

lera

tion

ampl

ifyin

g co

effici

ent

(a) Acceleration amplifying factors changed along the test phases of CIPS and MPCS

0123456789

101112

Floo

r

15 20 25 30 35 40 45 50 5510Acceleration amplifying coefficient of B-WSM

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 g

MPCS-040 gMPCS-062 gCIPS-0035 gCIPS-007 g

CIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Acceleration amplifying factors of B-WSMwave along the height oftwo models

Figure 10 Acceleration amplifying factors of CIPS and MPCS

(1) The typical failure pattern of the CIPS was damageconcentration in coupling beams in first and thenthe cracks emerged in the shear wall in plasticstage However besides the concentrated damage incoupling beam the connections between the precastelement and the CIP floor were the weakness and

the vertical crack came after the horizontal crack inplastic stage The precast element with high qualitiesdid not present cracks in the test

(2) The initial natural frequency as the initial stiffnessof MPCS degraded more obviously than CIPS forthe fine cracks in the connections The damping

12 Shock and Vibration

1 2 3 4 5 6 7 8 9 10 11 120Seismic force of DZC (kN)

1 2 3 4 5 6 7 8 9 10 11 120Seismic force of B-WSM (kN)

1 2 3 4 5 6 7 8 9 10 11 12 130Seismic force of ELW (kN)

123456789

101112

Floo

r

123456789

101112

Floo

r

123456789

101112

Floo

r

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(a) Distribution of seismic forces

10 20 30 40 50 60 70 80 90 1000Story shear of ELW (kN)

0123456789

101112

Floo

r10 20 30 40 50 60 70 80 900

Story shear of B-WSM (kN)

0123456789

101112

Floo

r

0123456789

101112

Floo

r

10 20 30 40 50 60 70 800Story shear of DZC (kN)

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Distribution of story shear

Figure 11 Earthquake actions of CIPS and MPCS

ratio of them had similar parameter but the way ofenergy dissipation of them was different And theyhad similar mode shapes

(3) The acceleration amplifying factors of them increasedalong the height of models and decreased progres-sively with the increasing PGA Nevertheless theyhad different responses under the different earth-quake waves The maximum acceleration amplifyingfactor of CIPS was 546 under the ELW with PGA of022 g and that of MPCS was 523 under the B-WSMwith PGA of 040 g

(4) Distribution of seismic forces of them increasedlinearly along the height ofmodels and then presentednonlinearity for the influence of high-order modesThe distribution of the interstory shear of CIPS and

MPCS showed similar rules like inverted triangle inelastic stage

(5) The maximum story displacements of them werealmost equal in elastic stage However the maximumstory displacement of CIPS was greater than thatof MPCS in plastic stage which was caused bythe concentrated failure in 4th story of CIPS Themaximum interstory ratio under the seismic waveswith PGA of 007 g and 040 g was in accordancewith the provision of Chinese code Under the moreintensive earthquake wave of PGA of 080 g the twomodels had enough capacity to resist collapse

Conflicts of Interest

The authors declare that they have no conflicts of interest

Shock and Vibration 13

5 10 15 20 25 30 35 40 450Displacement of B-WSM (mm)

0123456789

101112

Floo

r

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of DZC (mm)

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of ELW (mm)

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(a) Maximum story displacements of CIPS and MPCS

0123456789

101112

Floo

r

000

1

001

0

000

8

000

4

001

2

000

0

000

7

000

5

000

9

000

3

001

1

000

2

000

6

Story drift of DZC

000

1

000

8

000

4

001

0

000

0

000

60

007

000

3

000

9

000

2

001

1

000

5

Story drift of B-WSM

0123456789

101112

Floo

r

0123456789

101112

Floo

r

000

1

001

2

001

0

000

4

000

8

001

3

000

0

000

6

000

9

000

5

001

1

000

30

002

000

7

Story drift of ELW

Elastic limit Plastic limitElastic limit Plastic limitElastic limit Plastic limit

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Story drift ratio of CIPS and MPCS

Figure 12 Displacement responses of CIPS and MPCS

Acknowledgments

This research has been supported by both Innovation TeamofXirsquoan University of Architecture and Technology and ShaanxiProvince Science and Technology Coordination InnovationProject (nos 2015KTZDSF03-05-01 2015KTZDSF03-04 and2014SZS04-Z01)The support of the National Natural ScienceFoundation of China (Grants nos 51408456 51578444 and51478381) and Ministry of Education Plan for Yangtze RiverScholar and Innovation TeamDevelopment (no IRT13089) isalso acknowledged

References

[1] R Park and T Paulay Reinforced Concrete Structures ohnWileyamp Sons 1975

[2] M Fintel ldquoPerformance of buildings with shear walls in earth-quakes of the last thirty yearsrdquo PCI journal vol 40 no 3 pp62ndash80 1995

[3] A A Yee ldquoSplice sleeve for reinforcing barsrdquo US Pat vol 35401968

[4] J H Ling A B Abd Rahman I S Ibrahim and Z AbdulHamid ldquoBehaviour of grouted pipe splice under incrementaltensile loadrdquo Construction and Building Materials vol 33 no8 pp 90ndash98 2012

[5] M T El-Sheikh R Sause S Pessiki and L-W Lu ldquoSeismicbehavior and design of unbonded post-tensioned precast con-crete framesrdquo PCI journal vol 44 no 3 pp 54ndash71 1999

[6] R S Henry S Sritharan and J M Ingham ldquoUnbonded tendonstresses in post-tensioned concrete walls at nominal flexuralstrengthrdquo ACI Structural Journal vol 109 no 2 pp 171ndash1822012

14 Shock and Vibration

[7] P Seifi R S Henry and JM Ingham ldquoPanel connection detailsin existing New Zealand precast concrete buildingsrdquo Bulletin ofthe New Zealand Society for Earthquake Engineering vol 49 no2 pp 190ndash199 2016

[8] K A Soudki S H Rizkalla and R W Daikiw ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformations part 2 prestressed connectionsrdquo PCI journal vol40 no 5 pp 82ndash96 1995

[9] K A Soudki S H Rizkalla and B LeBlanc ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformationsmdashpart 1 mild steel connectionsrdquo PCI journal vol40 no 4 pp 78ndash96 1995

[10] K A Soudki J S West S H Rizkalla and B BlackettldquoHorizontal connections for precast concrete shear wall panelsunder cyclic shear loadingrdquo PCI journal vol 41 no 3 pp 64ndash80 1996

[11] R Vaghei F Hejazi H Taheri M S Jaafar and A A Ali ldquoEval-uate performance of precast concrete wall to wall connectionrdquoAPCBEE Procedia vol 9 pp 285ndash290 2014

[12] H Jiang Z Chen J Zhang B Wu Y Tian andW Liu ldquoQuasi-static test of precast reinforced concrete shear wall structurerdquoJianzhu Jiegou XuebaoJournal of Building Structures vol 32 no6 pp 34ndash40 2011

[13] Z ChenH Jiang J Zhang BWu Y Tian andW Liu ldquoPseudo-dynamic substructure test on precast reinforced concrete shearwall structurerdquo Jianzhu Jiegou XuebaoJournal of Building Struc-tures vol 32 no 6 pp 41ndash50 2011

[14] J R QianW L Han Z Z Zhao and H Qin ldquoPseudo-dynamicSubstructure Test on a 3-story Full-scaleModel of PrefabricatedConcrete Shear Wall Structure with Rebars Splicing by GroutSleevesrdquo Journal of Building Structures vol 38 no 3 pp 26ndash382017

[15] P Negro D A Bournas and F J Molina ldquoPseudodynamictests on a full-scale 3-storey precast concrete building Globalresponserdquo Engineering Structures vol 57 pp 594ndash608 2013

[16] M Oliva P Gavrilovic and R W Clough ldquoSeismic testing oflarge panel precast walls Comparison of pseudostatic and shak-ing table testsrdquo Earthquake Engineering amp Structural Dynamicsvol 19 no 6 pp 859ndash875 1990

[17] L H Lee W H Yi and S Y Seo ldquoShaking Table Test of PrecastConcrete Wall Structurerdquo in Proceedings of the Eleventh WorldConference on Earthquake Engineering

[18] W Wang A Q Li et al ldquoShaking table test of precast concreteshear wall isolation structurerdquo Journal of Building Structuresvol 36 no 6 pp 35ndash43 2015

[19] EERI Preliminary Reconnaissance Report Earthquake Engi-neering Research Institute 1994

[20] EERI Armenia Earthquake Reconnaissance Report EarthquakeEngineering Research Institute 1989

[21] GB 50011-2010 Code for Seismic Design of Buildings ChinaArchitecture and Building Press Beijing China 2010

[22] JGJ 1-2014 Technical Specification for Precast Concrete Struc-tures China Architecture amp Building Press Beijing China2014

[23] 15J939-1 Example of Architectural Design for PrefabricatedConcrete Structures China Planning Press Beijing China 2015

[24] E Buckingham ldquoOn physically similar systems Illustrations ofthe use of dimensional equationsrdquo Physical Review A AtomicMolecular and Optical Physics vol 4 no 4 pp 345ndash376 1914

[25] Y Zhou andX L LuTheexperimentalmethod and technology ofshaking table test about building structure Science Press BeijingChina 2012

[26] JGJT 101-2015 Specification for Seismic Test of Buildings ChinaArchitecture amp Building Press Beijing China 2014

[27] C Y Shen F L Zhou et al ldquoExperimental research on micro-concrete used in dynamic test modelrdquo Journal of GuangzhouUniversity (Natural Science Edition) vol 4 no 3 pp 250ndash2532005

[28] J Xiao T L Pham and T Ding ldquoShake table test on seismicresponse of a precast frame with recycled aggregate concreterdquoAdvances in Structural Engineering vol 18 no 9 pp 1517ndash15342015

[29] S C Chakrabarti G C Nayak and D K Paul ldquoShear charac-teristics of cast-in-place vertical joints in story-high precast wallassemblyrdquo ACI Structural Journal vol 85 no 1 pp 30ndash45 1988

[30] CWang and J Xiao ldquoStudy of the seismic response of a recycledaggregate concrete frame structurerdquo Earthquake Engineeringand Engineering Vibration vol 12 no 4 pp 669ndash680 2013

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 2: Comparative Study of Seismic Behavior between Monolithic

2 Shock and Vibration

of welded and poorly constructed connections [19 20]Further experimental research is clearly needed to fill thegap in the knowledge of seismic behavior of precast concretestructure And large-scale shake table test is a reliablemethodto research the dynamic seismic response of precast concretestructure

This paper presents a comparative shaking table testprogram realized on two 1 5 scaledmodels of a 12-story shearwall structure to understand the dynamic seismic responseof precast concrete structure One is cast-in-place structure(CIPS) and the other ismonolithic precast concrete structure(MPCS) The prototype structure was designed in two baysand two spans in accordance with the provisions of code [21]and the splitting design of MPCS was according to the code[22] and design drawings sheets [23] Based on the testingresults the dynamic characteristics of the two models suchas frequency damping ratio and mode shape are evaluatedthrough the white-noise test In comparison the failurepattern and mechanism seismic force response story shearstory displacement and interstory drift will be intensivelyexplored analyzed and discussed Finally the comprehensiveunderstanding of seismic performance of MPCS and CIPSwill be revealed especially the earthquake response of theMPCS in its entirety

2 Experimental Program

21 Model Design

211 Similitude Relationship As the performance parametersof shake table and condition of lift in laboratory scaledmodel was applied in shake table test Prototype structurewas designed observing the provisions of Chinese code [21]And the reduced scale design for the models was establishedfrom Buckinghamrsquos Pi theorem [24] The similar constants ofgeometry stress and acceleration were first defined as 0202 and 1 respectively [25] And then the other parameterswere deduced by the similitude rules and summarized inTable 1 The CIPS and MPCS had the same similitude rulesFurthermore the models were designed as elastic-plasticmodels to observe the plastic behavior under the high-intensity earthquake waves [26]

212 Material Design As the scaled parameters of physicsthe stress and elastic modulus of the model materialdecreased to 20 of that of the concrete of prototypestructure The microconcrete was adopted as model materialfor the limitation of coarse aggregate Shen et al pro-posed ceramsite powder coal ash or pumice as blendingagent could reduce the elastic modulus of microconcrete[27] So the gypsum was mixed to the microconcreteAfter testing the model material was a compound withcement fine aggregate coarse aggregate water gypsum =1 364 364 093 05 (inweight ratio)Theultimate strengthof the microconcrete was 894MPa and the elastic modulusof that was 729GPa consistently with the similitude constant02 in contrast with C40 concrete Galvanized iron wire wasused to replace the reinforcements following the similar rulesof internal forces [25]

213 Plane Design Considering the architectural space thephysical dimension and connection plane of the precastelements of high-rise residence in practice the model wasa two-bay two-span twelve-story shear wall structure thatwas regular in plan and elevation The CIPS and MPCSscaled models were 1800mm times 1800mm in plane and witha constant story height of 600mm The distance of spanwas 1100mm and 700mm in 119883 direction and 900mm and900mm in 119884 direction The thickness of the shear wall andcoupling beam was 40mm and that of the slab was 30mm Itcontained three kinds of coupling beamswith different spans500mm 300mm and 160mm respectively Figure 1 displaysthe plane view of the models

The MPCS model included three kinds of cast-in-placeconnections binding precast elements ldquoLrdquo type ldquoTrdquo type andldquo+rdquo type to forman entirety part in each story and three kindsof precast concrete shear walls (PCSW) PCSW-1 PCSW-2and PCSW-3The three connections represented the externalPCSW connected in corner external and internal PCSWconnected in side and internal PCSW connected inside ofprecast structure Moreover the cast-in-pace connectionswere corresponding to the edge components of shear wallof the CIPS model which were separated from the precastelement in the factory and then made with cast-in-placeconcrete after the precast elements were installed Namelythe cast-in-pace connections and the PCSW composed theshear wall of CIPS The design parameters model materialsand loading program of MPCS were the same as those ofCIPS However the models had diverse construction tech-niques The details reinforcements of the cast-in-place con-nections or the edge components and PCSW are exhibited inFigure 2

22 Precast and Construction Details The structural mea-sures of CIPS were complying with provisions of the code[21] The CIPS model was constructed in the general con-struction technique including assembling reinforcementtemplate installation pouring concrete and maintenanceHowever the precast units are fabricated in factory trans-ported to the construction site and lifted by crane and com-bined together with the cast-in-place concrete later formingthe monolithic precast concrete structure in practice In thisproject a compromised construction method was adoptedThe model precast concrete shear walls were fabricated inthe laboratory The reinforcements were connected by thefollowing ways The upper protruding bar extended throughthe grouted sleeve to connect the next PCSW and the side onehooked the longitudinal reinforcements in CIP connectionand put on additional stirrups later Between the upper andlower PCSW the cast-in-place floor replaced the assemble-monolithic concrete composite slab with similar stiffness forconvenience The remaining CIP connection and floor werepoured after the PCSW was cured in 48 hours Thereforethe coupling beam was precast superposed one in MPCSand that in CIPS was integrity The model material wasmicroconcretemixed gypsumwith 894MPaTheproceduresof MPCS and CIPS are shown in Figure 3 They were curedat normal temperature for 28 days and tested by shake tableunder earthquake waves

Shock and Vibration 3

Table1Similitude

scalep

aram

eterso

fCIPSandMPC

S

Item

parameter

Geometry

leng

thPh

ysics

Load

Dyn

amiccharacteris

tics

Elastic

mod

ulus

Stress

Poiss

onrsquosratio

Strain

Mass

Massd

ensity

Frequency

Acceler

ation

Form

ula

119878 119871119878 119864

119878 120590=119878 119864

119878 ]119878 120576=119878 120590 119878 119864119878 119872=119878 119864119878 1198712

119878 119886119878 120588=119878 120590 119878 119871119878 119886

119878 119891=119878 11988612

119878 11987112

119878 119886=119878 119892

Relationship

02

02

02

11

0008

12236

1Re

mark

Con

trolsize

Con

trolm

aterial

Con

trolshaking

tabletest

Note(120588119892119897119864) 119898=(120588119892119897119864) 119901ldquo119898rdquom

eans

mod

elstr

uctureldquo119901rdquom

eans

prototypes

tructure

4 Shock and Vibration

1 2 3

A

B

C

Coupling beam 1

PCSW-1

CIP connection-T

X

Y

N

CIP connection-L

CIP connection-+

PCSW-2 PCSW-3

Coupling beam 2 Coupling beam 3

1 2 3

A

B

C

Coupling beam 1

SW-1

X

Y

N

SW-2 SW-3

Coupling beam 2 Coupling beam 3

Cast-in-place concrete partsPrecast concrete shear wall

Opening in the shear wall

Cast-in-place structure (CIPS) Monolithic precast concrete structure (MPCS)

Edge component Edge component

80 220 500 220 160 190 80 20

8022

030

022

016

022

030

022

080

2090

090

020

160 1901800

80 220 500 220 160 190 80 20

8022

030

022

016

022

030

022

080

2090

090

020

160 1901800

Figure 1 Plane view of CIPS and MPCS

In order to supplement the missing gravity and nonstruc-tural elements iron blockswere used as the artificialmass andevenly fixed on each floor of the twomodels about 156 tonesThe total mass of each model reached 136 tones includinggrade beams and the height of the two models was 756mwhich conformed to the capacity limitation of the shakingtable system

23 Test Procedure It is well known that the condition of soilat the site is one of the significant factors in choosing theseismic inputs for shaking table test The soil layer equivalentcutting wave speed and the thickness of overlay soil layerdetermine the site classification The type-II soil site wasdefined in the Code for Seismic Design of Buildings [21]which was the condition soil of this project In comparisonwith seismic design response spectra Superstition Hillsearthquake wave (B-WSM) Kocaeli (Turkey) earthquakewave (DZC) and El Centro earthquake wave (ELW) wereselected as the ground excitations and inputted by thementioned sequenceThewaves were chosen from the PacificEarthquake Engineering Research Center (PEER) The testswere performed with unidirectional and bidirectional earth-quake waves with the ratio of PGA 1 085 to evaluate overallseismic performance of CIPS andMPCSThe test program ofthe shake table test included eight phases and the peak groundacceleration (PGA) is 0035 g 007 g 014 g 022 g 040 g062 g 070 g and 080 g in each phase respectivelyThe PGAof 070 g and 080 g were inputted to observe the nonlinearbehavior of them After each teat phase the white noise

with PGA of 0035 g was inputted to detect the dynamiticcharacteristics of the models

To monitor the earthquake response of the two models32 single-axis accelerometers including two on the shaketable two on the grade beam twenty-four on each floor in119883 and 119884 directions and four on the diagonal points in the12th floor were set to record the horizontal acceleration Atotal of 12 displacement transducers and 12 velocity vibrationsensors were installed to each floor The arrangement of testinstruments is exhibited in Figure 4 Figure 5 displays themodels on the shake table

3 Failure Pattern and Mechanism

Cracks and damage to the models were traced along withthe increasing PGA With the PGA of 0035 g diagonalmicrocracks on the coupling beam were merely observedin main direction of CIPS and those of MPCS were foundin 119883 and 119884 directions of shear wall arranged from 1st to4th As the PGA increased from 007 g to 022 g cracks onthe CIPS were extended along the corner of coupling beamand some new diagonal microcrack occurred In the phasesthe existing cracks of MPCS penetrated the coupling beamand meanwhile new microcracks appeared The diagonalmicrocrackswere themain pattern of themodels in this stage

In the next stage the models presented different crackpatterns After the PGA of 040 g the cracks of CIPSconcentrated rapidly in the end of coupling beams suchas the biggest span-depth ratio one with decreased section

Shock and Vibration 5

ldquoLrdquo type

ldquoTrdquo type

Floor

Sleeve

Longitudinalreinforcement

Additional stirrup

Protruding bar

Grouting materials

Cast-in-place concreteconnection of ldquoLrdquo type Roughened surface

External precastconcrete shear wall

Additional stirrup

10040 60

5

4060

100

5

45

45

External precastconcrete shear wall

12(10)1430

Protruding bar

Longitudinalreinforcement

Cast-in-place concreteconnection of ldquoLrdquo type

External precastconcrete shear wall

Floor

Sleeve

Longitudinalreinforcement

Grouting materials

Additional stirrupProtruding bar

Cast-in-place concreteconnection of ldquoTrdquo type Roughened surface

External precastconcrete shear wall 604060

4060

45

45

5

5

16(10)1430

Additional stirrup Protruding bar

Longitudinalreinforcement

Cast-in-place concreteconnection of ldquoTrdquo type

External precastconcrete shear wall External precast

concrete shear wall

Internal precastconcrete shear wall

ldquo+rdquo type

Floor

Sleeve

Longitudinalreinforcement

Grouting materials

Additional stirrupProtruding bar

Cast-in-place concreteconnection of ldquo+rdquo type Roughened surface

Internal precastconcrete shear wall

60 40 60

45

45

5

5

6040

60

20(10)1430

Additional stirrup

Protruding bar

Longitudinalreinforcement

Cast-in-place concreteconnection of ldquo+rdquo type

Internal precastconcrete shear wall

Internal precastconcrete shear wall

(a) Reinforcement of CIP connections

Figure 2 Continued

6 Shock and Vibration

PCSW-1 PCSW -3

PCSW-2

30 60 60 50

27050 60 60 30

270

3030 30 30

90

570

250

6(10)1430

1440

14204(8)

120 740 120

10 10

Longitudinal reinforcement

Protruding bar

Longitudinal reinforcement

Protruding bar

Verticaland levelreinforcement

Coupling beam 1

Edge component

Protruding bar

90

570

30 60 60 50

230

3050 60 60 30

230

30

250

6(10)1430

1440

14204(8)

10 10

120 540 120

Longitudinal reinforcement

Protruding bar

Longitudinal reinforcement

Protruding bar

Verticaland levelreinforcement

Coupling beam 3

Protruding bar

Edge component

30 60 60 50270

50 60 60 30

270

30

90

570

30 30 30

250

6(10)1430

1440

14204(8)

10 10

120 940 120

Longitudinal reinforcement

Protruding bar

Longitudinal reinforcement

Protruding bar

Verticaland levelreinforcement

Coupling beam 2

Protruding bar

Edge component

8lowast60=

480

8lowast60=

480

8lowast60=

480

(b) Reinforcement of PCSW

Figure 2 Configuration and reinforcements of the components (unit mm)

for the plastic hinge and the smallest span-depth ratio onewith diagonal cracks On the contrary diagonal cracks ofMPCS extended slowly and the horizontal fracture betweenthe PCSW and the CIP floor as the special form occurredat the same time With the increasing seismic energy thenew cracks were successively added along the height of themodels The cracks of CIPS were like the diagonal cracksoccurring with the PGA of 0035 g and those of MPCS werehorizontal cracks occurring in horizontal connections Inorder to observe the nonlinear behavior of them the PGA of

080 gwas inputted In this phase the vertical cracks along theedge component and the group cracks in the shear wall in 4thstory occurred in CIPS We considered that the weak story ofCIPS was the 4th story The horizontal crack extended to thecast-in-place concrete and the vertical crack appeared in thecontact of precast and cast-in-place connection in 2nd and3rd stories in MPCSThe overall crack patterns of models areexhibited in Figure 6

In general the diagonal cracks were presented in the end-ing coupling beams of CIPS and MPCS The phenomenon is

Shock and Vibration 7

CIP connection

Coupling beam

PCSW elementCIP connection PCSW element

CIP floor

CIPS MPCS

Construction procedures of CIPS

Construction procedures of MPCS

Grouted sleeves

Figure 3 Construction procedures of CIPS and MPCS

Table 2 Natural frequency of CIPS and MPCS

PGAg

FrequencyHzFirst frequency Second frequency

119883-direction 119884-direction 119883-direction 119884-directionMPCS CIPS MPCS CIPS MPCS CIPS MPCS CIPS

Initial 619 513 7 569 2500 2075 2794 23250035 g 481 512 513 563 2056 2013 2244 235007 g 480 512 506 55 2006 1988 2200 2325014 g 469 456 494 537 1969 1988 2106 2206022 g 426 413 444 481 1825 1813 1981 2094040 g 279 281 375 375 1419 1356 1625 1656062 g 231 194 300 269 1163 1006 1363 1300080 g 186 155 288 254 938 942 1318 1288

the favored state They possessed different energy dissipationmechanisms under the intensity earthquake waves The cou-pling beam acted as the first line of energy dissipation Whilethe coupling beam formed a plastic hinge the shear wallchanged to a single-slice wall to dissipate energy as the secondline to forbid collapse in CIPS Besides the coupling beamthe relatively weak connections between the precast elementsplayed the new way to dissipate energy in MPCS whichwere horizontal cracks first and then the vertical cracksFurthermore they defended the lateral resisting system insome degree

4 Earthquake Response Analysis

41 Dynamic Characteristics The dynamic characteristics ofstructure include natural frequency stiffness damping ratioand mode shape They could be deduced from the whitenoise inputted after each test phase by the transfer functionThe first and second natural frequency in 119883 direction and119884 direction are shown in Table 2 Also the stiffness could becalculated with the frequency and it is presented in Figure 7[28] Initial frequency of MPCS was bigger than that ofCIPS and the same to the initial stiffness We assumed that

8 Shock and Vibration

D11DDDD11D11D11D11DDDD 11D

A1 D1

A2 V1

A3 D2

A4 V2

A5 D3

A6 V3

A7 D4

A8 V4

A9 D5

A10 V5

A11 D6

A12 D7

A13 D8 V6

A14

A15

A16

A17

A18

A19

A20

A21

A22

A23

A24

A25

A26

V7

V9

D9

V11

V8

V10

D10

V12

D12

Y

X

Figure 4 Arrangement of accelerometers displacement transduc-ers and velocity vibration sensors

CIPS MPCS

Figure 5 General view of CIPS and MPCS

the adjunction grouted sleeves and the additional stirrupsin the connections deduced the phenomenon As the firstearthquake excitation was inputted the natural frequency ofMPCS decreased about 20 which could be caused by theshrinkage and microcracks in connections as initial damage[29] With the increasing energy the influence of the initialdamage was not the main factor And then the two modelswith approximately equal frequency presented in failure stateThe curves of stiffness degradation of CIPS reduced graduallywith the increasing PGA In contrast withCIPS that ofMPCS

CIPS MPCS

Figure 6 Failure patterns of CIPS and MPCS

00

01

02

03

04

05

06

07

08

09

10

11

Stiff

ness

ratio

MPCS-XMPCS-Y

CIPS-XCIPS-Y

022 g014 g007 g 040 g0035 g 062 gPrimePGA (g)

Figure 7 Stiffness ratio versus frequency in119883 and 119884 directions

decreased evidently in the first phase and then diminishedslowly from PGA of 0035 g to 014 g Finally they hadsimilar residual stiffnesses The diverse tendency might beunderstood by the mentioned failure pattern

Damping ratio reflects the dissipation capacity of struc-ture As is shown in Figure 8 the damping ratio increasedgradually following the PGA In the first phase it was 42And then the damping ratio increased slowly before the PGAof 040 g which arranged from 42 to 50 The averagedamping ratio of each phase changed from 42 to 82in the loading process which was in the scope of cast-in-place concrete structure Nevertheless the energy dissipationmechanismofCIPS andMPCSwas different for the order anddistribution of cracks

The first and second mode shapes of the models aredescribed in Figure 9 In general the first-order mode shape

Shock and Vibration 9

000

001

002

003

004

005

006

007

008

009

010

011

Dam

ping

ratio

MPCS-XMPCS-Y

CIPS-XCIPS-Y

014 g 022 g007 g0035 g 040 g 062 gPrimePGA (g)

Figure 8 Damping ratios of CIPS and MPCS

exhibited bending deformation characteristics and the lateralstiffness was uniform distribution along the height of themodel The shapes of them were bending to 119909-axis littleby little The phenomenon would be caused by high-ordermodes And the trend of that of CIPSwasmore obvious in thePGAof 040 g and 062 g since the severe damage occurred inthe shear wall The second-order mode shapes of them weresimilar And the maximummode coefficient of second modeshape was in positon of 4th floor

42 Acceleration Response The ratio of measured accelera-tion to the corresponding input peak ground accelerationis called the acceleration amplifying factor It reflects thedynamic response of structure under earthquake excitationsThe acceleration amplifying factors along the height of themodels are described in Figure 10(a) under the B-WSMDZCand ELW ground motions for different seismic levels Obvi-ously that of the two models was gradually increased alongthe height of themodel in each test phaseWith the increasingPGA arranged from 007 g to 062 g the whole developmenttrend of that decreased gradually implying the progressivedegradation of structural stiffness In each test phase themodels presented different dynamic responses under thedifferent earthquake excitations In elastic stage with thePGA of 007 g and 014 g the CIPS had greater response toB-WSM and ELW than to DZC The MPCS had uniformresponse to the three excitations The earthquake waves withdifferent frequency-spectrum characteristics would lead tothis phenomenon The greatest response of CIPS occurredunder the ELW with PGA of 022 g and that of MPCSoccurred in the B-WSM with PGA of 040 g The diversitiesof twomodels about the acceleration amplifying factors couldbe caused by the connections between the precast elementsThe factors of CIPS and MPCS reduced from 546 to 319and from 523 to 308 in the final test phase respectivelyAs shown in Figure 10(b) the distribution feature was more

regular under the test phases with PGA from 0035 g to 014 gWhen the two models were subjected to severe damage theinfluence of high-order vibration modes increased graduallyand the acceleration amplifying factors at somemeasurementpoints no longer conformed to the distribution [30]

43 Earthquake Action The seismic force distribution char-acteristic of structure is very important reference for theaseismic design and application of the MPCS and CIPS Themaximum seismic force of the 119894th floor is derived as follows

119865119894max = 1198981198941003816100381610038161003816 119894 (119905) + 0 (119905) max

1003816100381610038161003816 (1)

where 119865119894max is the maximum seismic force 119898119894 is the lumpedmass of the 119894th floor 119894(119905) is the acceleration response of the119894th floor relative to the ground at the time of 119905 and 0(119905) is theground acceleration at the time of 119905

According to (1) the maximum seismic forces of CIPSand MPCS are presented and compared in Figure 11 Asillustrated in Figure 11(a) the results of them were presentedin linear behavior following the increasing earthquake waveswith PGA from 0035 g to 014 g The curves were increasedprogressively along the height of models They could reflectthe actual seismic force distribution in some extent of struc-ture in the elastic stage and the high-order modes could beignored in this time And then the cracks expanded graduallyin different ways of the twomodels and the influence of high-order modes worked The models entered into plastic stagewith obvious nonlinearity The maximum seismic forces ofthe CIPS emerged in the middle of model such as 5th 6thand 8th Those of MPCS were in 4th 5th and 8th In theplastic stage with PGA from 022 g to 062 g the distributionof seismic forces changed for the high-order modes

The interstory shear forceswere calculated by (1) and theyare illustrated along the height of models in Figure 11(b) Theinterstory shear force was increased gradually with the PGAand decreased along the height of model structure In elasticstage the distribution of the interstory shear of CIPS andMPCS showed similar rules like inverted triangle In plasticstage the vibration trend was effected by the high-ordermodes and interstory shear force was not strictlymeeting thedistribution pattern

44 Displacement Response The maximum story displace-ments of the CIPS and MPCS along the model heightobtained from the B-WSM DZC and ELW with thePGA from 0035 g to 062 g are compared and depicted inFigure 12(a) Since the shear wall structure is the effectivelateral resisting system [1] themaximum story displacementsof the models were small in 12th story arranged from 113mmto 756mm of CIPS and from 106mm to 699mm of MPCSin the elastic stageThe stage contained the PGA from 0035 gto 014 g With increasing the intensity of earthquake wavesthe two models demonstrated different damage phenomenadescribed in the above paragraph The concentrated damagein shear wall of 4th story of CIPS caused bigger displacementthan that of MPCS The bigger displacement formed thebigger torsion angle of floor and then brought about themuchbigger displacement in upside structure of CIPS And the

10 Shock and Vibration

0123456789

101112

Floo

r

01 02 03 04 05 06 07 08 09 10 1100Mode coefficient of first mode shape

0123456789

101112

Floo

r

01 02 03 04 05 06 07 08 09 10 1100Mode coefficient of first mode shape

minus08 minus06 minus04 minus02minus10 02 04 06 08 1000Mode coefficient of second mode shape

minus06 minus04 minus02minus08minus10 02 04 06 08 1000Mode coefficient of second mode shape

0123456789

101112

Floo

r

0123456789

101112

Floo

r

MPCS-initialMPCS-0035 gMPCS-007 gMPCS-014 g

MPCS-022 gMPCS-040 gMPCS-062 g

MPCS-initialMPCS-0035 gMPCS-007 gMPCS-014 g

MPCS-022 gMPCS-040 gMPCS-062 g

CIPS-initialCIPS-0035 gCIPS-007 gCIPS-014 g

CIPS-022 gCIPS-040 gCIPS-062 g

CIPS-initialCIPS-0035 gCIPS-007 gCIPS-014 g

CIPS-022 gCIPS-040 gCIPS-062 g

Figure 9 Mode shapes of CIPS and MPCS

maximumdestructive story driftwas in 4th floorMeanwhilethe two models showed obviously nonlinear behavior withthe PGA of 040 g and 062 g The shape of maximum storydisplacements of them was similar to the first mode shapeThe maximum story displacements were 4867mm in CIPSand 4198mm in MPCS

The maximum interstory drift ratios were calculated andpresented in Figure 12(b) The maximum value of story driftof CIPS was 11005 in 5th story under the DZC earthquakewave with the PGA of 070 g and that of MPCS was 11020in 5th story under the same wave They met the provisionabout the maximum story drift in elastic stage in area of 8-degree seismic fortification intensity in code of GB 50011-2010 [21] In the stage from PGA of 0035 g to 014 g thestory drift increased progressively The seismic design codeasks for the limiting value of plastic interstory drift ratio toforbid collapse We checked the story drift under the PGA of040 g The values of CIPS and MPCS were 1121 and 1127

which met the limiting value in rare earthquake action of 8-degree seismic regions Under the excitation of PGAof 062 gthe values exceeded the limiting value in rare earthquakeaction of 9-degree seismic regions And then the moreintensive earthquakewaves with PGAof 080 gwere inputtedthe models presented better seismic behavior withoutcollapse

5 Conclusion

A comparative seismic study between CIPS and MPCS wasconducted by shaking table test which contained a 15 scaled12-story CIPS model and MPCS model The experimentalresults of dynamic characteristic failure pattern and mecha-nism and seismic response of the models were discussed andcompared with each other to better understand their seismicbehavior Based on the intensive analysis of test results thefollowing conclusions are derived

Shock and Vibration 11

PGA of 062 gPGA of 040 gPGA of 022 gPGA of 014 gPGA of 007 g PGA of 062 gPGA of 040 gPGA of 022 gPGA of 014 gPGA of 007 g

EE E EB B B BDD DD

E(LW

)

D(Z

C)

B(-W

SM)

Peak ground acceleration of CIPS (g)

E EEEB B BB DDD D

E(LW

)

D(Z

C)

B(-W

SM)

Peak ground acceleration of MPCS (g)

12th floor11th floor10th floor9th floor8th floor

7th floor6th floor5th floor4th floor3rd floor

2nd floor1st floorFloor

12th floor11th floor10th floor9th floor8th floor

7th floor6th floor5th floor4th floor3rd floor

2nd floor1st floorFloor

081216202428323640444852566064

Acce

lera

tion

ampl

ifyin

g co

effici

ent

081216202428323640444852566064

Acce

lera

tion

ampl

ifyin

g co

effici

ent

(a) Acceleration amplifying factors changed along the test phases of CIPS and MPCS

0123456789

101112

Floo

r

15 20 25 30 35 40 45 50 5510Acceleration amplifying coefficient of B-WSM

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 g

MPCS-040 gMPCS-062 gCIPS-0035 gCIPS-007 g

CIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Acceleration amplifying factors of B-WSMwave along the height oftwo models

Figure 10 Acceleration amplifying factors of CIPS and MPCS

(1) The typical failure pattern of the CIPS was damageconcentration in coupling beams in first and thenthe cracks emerged in the shear wall in plasticstage However besides the concentrated damage incoupling beam the connections between the precastelement and the CIP floor were the weakness and

the vertical crack came after the horizontal crack inplastic stage The precast element with high qualitiesdid not present cracks in the test

(2) The initial natural frequency as the initial stiffnessof MPCS degraded more obviously than CIPS forthe fine cracks in the connections The damping

12 Shock and Vibration

1 2 3 4 5 6 7 8 9 10 11 120Seismic force of DZC (kN)

1 2 3 4 5 6 7 8 9 10 11 120Seismic force of B-WSM (kN)

1 2 3 4 5 6 7 8 9 10 11 12 130Seismic force of ELW (kN)

123456789

101112

Floo

r

123456789

101112

Floo

r

123456789

101112

Floo

r

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(a) Distribution of seismic forces

10 20 30 40 50 60 70 80 90 1000Story shear of ELW (kN)

0123456789

101112

Floo

r10 20 30 40 50 60 70 80 900

Story shear of B-WSM (kN)

0123456789

101112

Floo

r

0123456789

101112

Floo

r

10 20 30 40 50 60 70 800Story shear of DZC (kN)

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Distribution of story shear

Figure 11 Earthquake actions of CIPS and MPCS

ratio of them had similar parameter but the way ofenergy dissipation of them was different And theyhad similar mode shapes

(3) The acceleration amplifying factors of them increasedalong the height of models and decreased progres-sively with the increasing PGA Nevertheless theyhad different responses under the different earth-quake waves The maximum acceleration amplifyingfactor of CIPS was 546 under the ELW with PGA of022 g and that of MPCS was 523 under the B-WSMwith PGA of 040 g

(4) Distribution of seismic forces of them increasedlinearly along the height ofmodels and then presentednonlinearity for the influence of high-order modesThe distribution of the interstory shear of CIPS and

MPCS showed similar rules like inverted triangle inelastic stage

(5) The maximum story displacements of them werealmost equal in elastic stage However the maximumstory displacement of CIPS was greater than thatof MPCS in plastic stage which was caused bythe concentrated failure in 4th story of CIPS Themaximum interstory ratio under the seismic waveswith PGA of 007 g and 040 g was in accordancewith the provision of Chinese code Under the moreintensive earthquake wave of PGA of 080 g the twomodels had enough capacity to resist collapse

Conflicts of Interest

The authors declare that they have no conflicts of interest

Shock and Vibration 13

5 10 15 20 25 30 35 40 450Displacement of B-WSM (mm)

0123456789

101112

Floo

r

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of DZC (mm)

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of ELW (mm)

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(a) Maximum story displacements of CIPS and MPCS

0123456789

101112

Floo

r

000

1

001

0

000

8

000

4

001

2

000

0

000

7

000

5

000

9

000

3

001

1

000

2

000

6

Story drift of DZC

000

1

000

8

000

4

001

0

000

0

000

60

007

000

3

000

9

000

2

001

1

000

5

Story drift of B-WSM

0123456789

101112

Floo

r

0123456789

101112

Floo

r

000

1

001

2

001

0

000

4

000

8

001

3

000

0

000

6

000

9

000

5

001

1

000

30

002

000

7

Story drift of ELW

Elastic limit Plastic limitElastic limit Plastic limitElastic limit Plastic limit

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Story drift ratio of CIPS and MPCS

Figure 12 Displacement responses of CIPS and MPCS

Acknowledgments

This research has been supported by both Innovation TeamofXirsquoan University of Architecture and Technology and ShaanxiProvince Science and Technology Coordination InnovationProject (nos 2015KTZDSF03-05-01 2015KTZDSF03-04 and2014SZS04-Z01)The support of the National Natural ScienceFoundation of China (Grants nos 51408456 51578444 and51478381) and Ministry of Education Plan for Yangtze RiverScholar and Innovation TeamDevelopment (no IRT13089) isalso acknowledged

References

[1] R Park and T Paulay Reinforced Concrete Structures ohnWileyamp Sons 1975

[2] M Fintel ldquoPerformance of buildings with shear walls in earth-quakes of the last thirty yearsrdquo PCI journal vol 40 no 3 pp62ndash80 1995

[3] A A Yee ldquoSplice sleeve for reinforcing barsrdquo US Pat vol 35401968

[4] J H Ling A B Abd Rahman I S Ibrahim and Z AbdulHamid ldquoBehaviour of grouted pipe splice under incrementaltensile loadrdquo Construction and Building Materials vol 33 no8 pp 90ndash98 2012

[5] M T El-Sheikh R Sause S Pessiki and L-W Lu ldquoSeismicbehavior and design of unbonded post-tensioned precast con-crete framesrdquo PCI journal vol 44 no 3 pp 54ndash71 1999

[6] R S Henry S Sritharan and J M Ingham ldquoUnbonded tendonstresses in post-tensioned concrete walls at nominal flexuralstrengthrdquo ACI Structural Journal vol 109 no 2 pp 171ndash1822012

14 Shock and Vibration

[7] P Seifi R S Henry and JM Ingham ldquoPanel connection detailsin existing New Zealand precast concrete buildingsrdquo Bulletin ofthe New Zealand Society for Earthquake Engineering vol 49 no2 pp 190ndash199 2016

[8] K A Soudki S H Rizkalla and R W Daikiw ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformations part 2 prestressed connectionsrdquo PCI journal vol40 no 5 pp 82ndash96 1995

[9] K A Soudki S H Rizkalla and B LeBlanc ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformationsmdashpart 1 mild steel connectionsrdquo PCI journal vol40 no 4 pp 78ndash96 1995

[10] K A Soudki J S West S H Rizkalla and B BlackettldquoHorizontal connections for precast concrete shear wall panelsunder cyclic shear loadingrdquo PCI journal vol 41 no 3 pp 64ndash80 1996

[11] R Vaghei F Hejazi H Taheri M S Jaafar and A A Ali ldquoEval-uate performance of precast concrete wall to wall connectionrdquoAPCBEE Procedia vol 9 pp 285ndash290 2014

[12] H Jiang Z Chen J Zhang B Wu Y Tian andW Liu ldquoQuasi-static test of precast reinforced concrete shear wall structurerdquoJianzhu Jiegou XuebaoJournal of Building Structures vol 32 no6 pp 34ndash40 2011

[13] Z ChenH Jiang J Zhang BWu Y Tian andW Liu ldquoPseudo-dynamic substructure test on precast reinforced concrete shearwall structurerdquo Jianzhu Jiegou XuebaoJournal of Building Struc-tures vol 32 no 6 pp 41ndash50 2011

[14] J R QianW L Han Z Z Zhao and H Qin ldquoPseudo-dynamicSubstructure Test on a 3-story Full-scaleModel of PrefabricatedConcrete Shear Wall Structure with Rebars Splicing by GroutSleevesrdquo Journal of Building Structures vol 38 no 3 pp 26ndash382017

[15] P Negro D A Bournas and F J Molina ldquoPseudodynamictests on a full-scale 3-storey precast concrete building Globalresponserdquo Engineering Structures vol 57 pp 594ndash608 2013

[16] M Oliva P Gavrilovic and R W Clough ldquoSeismic testing oflarge panel precast walls Comparison of pseudostatic and shak-ing table testsrdquo Earthquake Engineering amp Structural Dynamicsvol 19 no 6 pp 859ndash875 1990

[17] L H Lee W H Yi and S Y Seo ldquoShaking Table Test of PrecastConcrete Wall Structurerdquo in Proceedings of the Eleventh WorldConference on Earthquake Engineering

[18] W Wang A Q Li et al ldquoShaking table test of precast concreteshear wall isolation structurerdquo Journal of Building Structuresvol 36 no 6 pp 35ndash43 2015

[19] EERI Preliminary Reconnaissance Report Earthquake Engi-neering Research Institute 1994

[20] EERI Armenia Earthquake Reconnaissance Report EarthquakeEngineering Research Institute 1989

[21] GB 50011-2010 Code for Seismic Design of Buildings ChinaArchitecture and Building Press Beijing China 2010

[22] JGJ 1-2014 Technical Specification for Precast Concrete Struc-tures China Architecture amp Building Press Beijing China2014

[23] 15J939-1 Example of Architectural Design for PrefabricatedConcrete Structures China Planning Press Beijing China 2015

[24] E Buckingham ldquoOn physically similar systems Illustrations ofthe use of dimensional equationsrdquo Physical Review A AtomicMolecular and Optical Physics vol 4 no 4 pp 345ndash376 1914

[25] Y Zhou andX L LuTheexperimentalmethod and technology ofshaking table test about building structure Science Press BeijingChina 2012

[26] JGJT 101-2015 Specification for Seismic Test of Buildings ChinaArchitecture amp Building Press Beijing China 2014

[27] C Y Shen F L Zhou et al ldquoExperimental research on micro-concrete used in dynamic test modelrdquo Journal of GuangzhouUniversity (Natural Science Edition) vol 4 no 3 pp 250ndash2532005

[28] J Xiao T L Pham and T Ding ldquoShake table test on seismicresponse of a precast frame with recycled aggregate concreterdquoAdvances in Structural Engineering vol 18 no 9 pp 1517ndash15342015

[29] S C Chakrabarti G C Nayak and D K Paul ldquoShear charac-teristics of cast-in-place vertical joints in story-high precast wallassemblyrdquo ACI Structural Journal vol 85 no 1 pp 30ndash45 1988

[30] CWang and J Xiao ldquoStudy of the seismic response of a recycledaggregate concrete frame structurerdquo Earthquake Engineeringand Engineering Vibration vol 12 no 4 pp 669ndash680 2013

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 3: Comparative Study of Seismic Behavior between Monolithic

Shock and Vibration 3

Table1Similitude

scalep

aram

eterso

fCIPSandMPC

S

Item

parameter

Geometry

leng

thPh

ysics

Load

Dyn

amiccharacteris

tics

Elastic

mod

ulus

Stress

Poiss

onrsquosratio

Strain

Mass

Massd

ensity

Frequency

Acceler

ation

Form

ula

119878 119871119878 119864

119878 120590=119878 119864

119878 ]119878 120576=119878 120590 119878 119864119878 119872=119878 119864119878 1198712

119878 119886119878 120588=119878 120590 119878 119871119878 119886

119878 119891=119878 11988612

119878 11987112

119878 119886=119878 119892

Relationship

02

02

02

11

0008

12236

1Re

mark

Con

trolsize

Con

trolm

aterial

Con

trolshaking

tabletest

Note(120588119892119897119864) 119898=(120588119892119897119864) 119901ldquo119898rdquom

eans

mod

elstr

uctureldquo119901rdquom

eans

prototypes

tructure

4 Shock and Vibration

1 2 3

A

B

C

Coupling beam 1

PCSW-1

CIP connection-T

X

Y

N

CIP connection-L

CIP connection-+

PCSW-2 PCSW-3

Coupling beam 2 Coupling beam 3

1 2 3

A

B

C

Coupling beam 1

SW-1

X

Y

N

SW-2 SW-3

Coupling beam 2 Coupling beam 3

Cast-in-place concrete partsPrecast concrete shear wall

Opening in the shear wall

Cast-in-place structure (CIPS) Monolithic precast concrete structure (MPCS)

Edge component Edge component

80 220 500 220 160 190 80 20

8022

030

022

016

022

030

022

080

2090

090

020

160 1901800

80 220 500 220 160 190 80 20

8022

030

022

016

022

030

022

080

2090

090

020

160 1901800

Figure 1 Plane view of CIPS and MPCS

In order to supplement the missing gravity and nonstruc-tural elements iron blockswere used as the artificialmass andevenly fixed on each floor of the twomodels about 156 tonesThe total mass of each model reached 136 tones includinggrade beams and the height of the two models was 756mwhich conformed to the capacity limitation of the shakingtable system

23 Test Procedure It is well known that the condition of soilat the site is one of the significant factors in choosing theseismic inputs for shaking table test The soil layer equivalentcutting wave speed and the thickness of overlay soil layerdetermine the site classification The type-II soil site wasdefined in the Code for Seismic Design of Buildings [21]which was the condition soil of this project In comparisonwith seismic design response spectra Superstition Hillsearthquake wave (B-WSM) Kocaeli (Turkey) earthquakewave (DZC) and El Centro earthquake wave (ELW) wereselected as the ground excitations and inputted by thementioned sequenceThewaves were chosen from the PacificEarthquake Engineering Research Center (PEER) The testswere performed with unidirectional and bidirectional earth-quake waves with the ratio of PGA 1 085 to evaluate overallseismic performance of CIPS andMPCSThe test program ofthe shake table test included eight phases and the peak groundacceleration (PGA) is 0035 g 007 g 014 g 022 g 040 g062 g 070 g and 080 g in each phase respectivelyThe PGAof 070 g and 080 g were inputted to observe the nonlinearbehavior of them After each teat phase the white noise

with PGA of 0035 g was inputted to detect the dynamiticcharacteristics of the models

To monitor the earthquake response of the two models32 single-axis accelerometers including two on the shaketable two on the grade beam twenty-four on each floor in119883 and 119884 directions and four on the diagonal points in the12th floor were set to record the horizontal acceleration Atotal of 12 displacement transducers and 12 velocity vibrationsensors were installed to each floor The arrangement of testinstruments is exhibited in Figure 4 Figure 5 displays themodels on the shake table

3 Failure Pattern and Mechanism

Cracks and damage to the models were traced along withthe increasing PGA With the PGA of 0035 g diagonalmicrocracks on the coupling beam were merely observedin main direction of CIPS and those of MPCS were foundin 119883 and 119884 directions of shear wall arranged from 1st to4th As the PGA increased from 007 g to 022 g cracks onthe CIPS were extended along the corner of coupling beamand some new diagonal microcrack occurred In the phasesthe existing cracks of MPCS penetrated the coupling beamand meanwhile new microcracks appeared The diagonalmicrocrackswere themain pattern of themodels in this stage

In the next stage the models presented different crackpatterns After the PGA of 040 g the cracks of CIPSconcentrated rapidly in the end of coupling beams suchas the biggest span-depth ratio one with decreased section

Shock and Vibration 5

ldquoLrdquo type

ldquoTrdquo type

Floor

Sleeve

Longitudinalreinforcement

Additional stirrup

Protruding bar

Grouting materials

Cast-in-place concreteconnection of ldquoLrdquo type Roughened surface

External precastconcrete shear wall

Additional stirrup

10040 60

5

4060

100

5

45

45

External precastconcrete shear wall

12(10)1430

Protruding bar

Longitudinalreinforcement

Cast-in-place concreteconnection of ldquoLrdquo type

External precastconcrete shear wall

Floor

Sleeve

Longitudinalreinforcement

Grouting materials

Additional stirrupProtruding bar

Cast-in-place concreteconnection of ldquoTrdquo type Roughened surface

External precastconcrete shear wall 604060

4060

45

45

5

5

16(10)1430

Additional stirrup Protruding bar

Longitudinalreinforcement

Cast-in-place concreteconnection of ldquoTrdquo type

External precastconcrete shear wall External precast

concrete shear wall

Internal precastconcrete shear wall

ldquo+rdquo type

Floor

Sleeve

Longitudinalreinforcement

Grouting materials

Additional stirrupProtruding bar

Cast-in-place concreteconnection of ldquo+rdquo type Roughened surface

Internal precastconcrete shear wall

60 40 60

45

45

5

5

6040

60

20(10)1430

Additional stirrup

Protruding bar

Longitudinalreinforcement

Cast-in-place concreteconnection of ldquo+rdquo type

Internal precastconcrete shear wall

Internal precastconcrete shear wall

(a) Reinforcement of CIP connections

Figure 2 Continued

6 Shock and Vibration

PCSW-1 PCSW -3

PCSW-2

30 60 60 50

27050 60 60 30

270

3030 30 30

90

570

250

6(10)1430

1440

14204(8)

120 740 120

10 10

Longitudinal reinforcement

Protruding bar

Longitudinal reinforcement

Protruding bar

Verticaland levelreinforcement

Coupling beam 1

Edge component

Protruding bar

90

570

30 60 60 50

230

3050 60 60 30

230

30

250

6(10)1430

1440

14204(8)

10 10

120 540 120

Longitudinal reinforcement

Protruding bar

Longitudinal reinforcement

Protruding bar

Verticaland levelreinforcement

Coupling beam 3

Protruding bar

Edge component

30 60 60 50270

50 60 60 30

270

30

90

570

30 30 30

250

6(10)1430

1440

14204(8)

10 10

120 940 120

Longitudinal reinforcement

Protruding bar

Longitudinal reinforcement

Protruding bar

Verticaland levelreinforcement

Coupling beam 2

Protruding bar

Edge component

8lowast60=

480

8lowast60=

480

8lowast60=

480

(b) Reinforcement of PCSW

Figure 2 Configuration and reinforcements of the components (unit mm)

for the plastic hinge and the smallest span-depth ratio onewith diagonal cracks On the contrary diagonal cracks ofMPCS extended slowly and the horizontal fracture betweenthe PCSW and the CIP floor as the special form occurredat the same time With the increasing seismic energy thenew cracks were successively added along the height of themodels The cracks of CIPS were like the diagonal cracksoccurring with the PGA of 0035 g and those of MPCS werehorizontal cracks occurring in horizontal connections Inorder to observe the nonlinear behavior of them the PGA of

080 gwas inputted In this phase the vertical cracks along theedge component and the group cracks in the shear wall in 4thstory occurred in CIPS We considered that the weak story ofCIPS was the 4th story The horizontal crack extended to thecast-in-place concrete and the vertical crack appeared in thecontact of precast and cast-in-place connection in 2nd and3rd stories in MPCSThe overall crack patterns of models areexhibited in Figure 6

In general the diagonal cracks were presented in the end-ing coupling beams of CIPS and MPCS The phenomenon is

Shock and Vibration 7

CIP connection

Coupling beam

PCSW elementCIP connection PCSW element

CIP floor

CIPS MPCS

Construction procedures of CIPS

Construction procedures of MPCS

Grouted sleeves

Figure 3 Construction procedures of CIPS and MPCS

Table 2 Natural frequency of CIPS and MPCS

PGAg

FrequencyHzFirst frequency Second frequency

119883-direction 119884-direction 119883-direction 119884-directionMPCS CIPS MPCS CIPS MPCS CIPS MPCS CIPS

Initial 619 513 7 569 2500 2075 2794 23250035 g 481 512 513 563 2056 2013 2244 235007 g 480 512 506 55 2006 1988 2200 2325014 g 469 456 494 537 1969 1988 2106 2206022 g 426 413 444 481 1825 1813 1981 2094040 g 279 281 375 375 1419 1356 1625 1656062 g 231 194 300 269 1163 1006 1363 1300080 g 186 155 288 254 938 942 1318 1288

the favored state They possessed different energy dissipationmechanisms under the intensity earthquake waves The cou-pling beam acted as the first line of energy dissipation Whilethe coupling beam formed a plastic hinge the shear wallchanged to a single-slice wall to dissipate energy as the secondline to forbid collapse in CIPS Besides the coupling beamthe relatively weak connections between the precast elementsplayed the new way to dissipate energy in MPCS whichwere horizontal cracks first and then the vertical cracksFurthermore they defended the lateral resisting system insome degree

4 Earthquake Response Analysis

41 Dynamic Characteristics The dynamic characteristics ofstructure include natural frequency stiffness damping ratioand mode shape They could be deduced from the whitenoise inputted after each test phase by the transfer functionThe first and second natural frequency in 119883 direction and119884 direction are shown in Table 2 Also the stiffness could becalculated with the frequency and it is presented in Figure 7[28] Initial frequency of MPCS was bigger than that ofCIPS and the same to the initial stiffness We assumed that

8 Shock and Vibration

D11DDDD11D11D11D11DDDD 11D

A1 D1

A2 V1

A3 D2

A4 V2

A5 D3

A6 V3

A7 D4

A8 V4

A9 D5

A10 V5

A11 D6

A12 D7

A13 D8 V6

A14

A15

A16

A17

A18

A19

A20

A21

A22

A23

A24

A25

A26

V7

V9

D9

V11

V8

V10

D10

V12

D12

Y

X

Figure 4 Arrangement of accelerometers displacement transduc-ers and velocity vibration sensors

CIPS MPCS

Figure 5 General view of CIPS and MPCS

the adjunction grouted sleeves and the additional stirrupsin the connections deduced the phenomenon As the firstearthquake excitation was inputted the natural frequency ofMPCS decreased about 20 which could be caused by theshrinkage and microcracks in connections as initial damage[29] With the increasing energy the influence of the initialdamage was not the main factor And then the two modelswith approximately equal frequency presented in failure stateThe curves of stiffness degradation of CIPS reduced graduallywith the increasing PGA In contrast withCIPS that ofMPCS

CIPS MPCS

Figure 6 Failure patterns of CIPS and MPCS

00

01

02

03

04

05

06

07

08

09

10

11

Stiff

ness

ratio

MPCS-XMPCS-Y

CIPS-XCIPS-Y

022 g014 g007 g 040 g0035 g 062 gPrimePGA (g)

Figure 7 Stiffness ratio versus frequency in119883 and 119884 directions

decreased evidently in the first phase and then diminishedslowly from PGA of 0035 g to 014 g Finally they hadsimilar residual stiffnesses The diverse tendency might beunderstood by the mentioned failure pattern

Damping ratio reflects the dissipation capacity of struc-ture As is shown in Figure 8 the damping ratio increasedgradually following the PGA In the first phase it was 42And then the damping ratio increased slowly before the PGAof 040 g which arranged from 42 to 50 The averagedamping ratio of each phase changed from 42 to 82in the loading process which was in the scope of cast-in-place concrete structure Nevertheless the energy dissipationmechanismofCIPS andMPCSwas different for the order anddistribution of cracks

The first and second mode shapes of the models aredescribed in Figure 9 In general the first-order mode shape

Shock and Vibration 9

000

001

002

003

004

005

006

007

008

009

010

011

Dam

ping

ratio

MPCS-XMPCS-Y

CIPS-XCIPS-Y

014 g 022 g007 g0035 g 040 g 062 gPrimePGA (g)

Figure 8 Damping ratios of CIPS and MPCS

exhibited bending deformation characteristics and the lateralstiffness was uniform distribution along the height of themodel The shapes of them were bending to 119909-axis littleby little The phenomenon would be caused by high-ordermodes And the trend of that of CIPSwasmore obvious in thePGAof 040 g and 062 g since the severe damage occurred inthe shear wall The second-order mode shapes of them weresimilar And the maximummode coefficient of second modeshape was in positon of 4th floor

42 Acceleration Response The ratio of measured accelera-tion to the corresponding input peak ground accelerationis called the acceleration amplifying factor It reflects thedynamic response of structure under earthquake excitationsThe acceleration amplifying factors along the height of themodels are described in Figure 10(a) under the B-WSMDZCand ELW ground motions for different seismic levels Obvi-ously that of the two models was gradually increased alongthe height of themodel in each test phaseWith the increasingPGA arranged from 007 g to 062 g the whole developmenttrend of that decreased gradually implying the progressivedegradation of structural stiffness In each test phase themodels presented different dynamic responses under thedifferent earthquake excitations In elastic stage with thePGA of 007 g and 014 g the CIPS had greater response toB-WSM and ELW than to DZC The MPCS had uniformresponse to the three excitations The earthquake waves withdifferent frequency-spectrum characteristics would lead tothis phenomenon The greatest response of CIPS occurredunder the ELW with PGA of 022 g and that of MPCSoccurred in the B-WSM with PGA of 040 g The diversitiesof twomodels about the acceleration amplifying factors couldbe caused by the connections between the precast elementsThe factors of CIPS and MPCS reduced from 546 to 319and from 523 to 308 in the final test phase respectivelyAs shown in Figure 10(b) the distribution feature was more

regular under the test phases with PGA from 0035 g to 014 gWhen the two models were subjected to severe damage theinfluence of high-order vibration modes increased graduallyand the acceleration amplifying factors at somemeasurementpoints no longer conformed to the distribution [30]

43 Earthquake Action The seismic force distribution char-acteristic of structure is very important reference for theaseismic design and application of the MPCS and CIPS Themaximum seismic force of the 119894th floor is derived as follows

119865119894max = 1198981198941003816100381610038161003816 119894 (119905) + 0 (119905) max

1003816100381610038161003816 (1)

where 119865119894max is the maximum seismic force 119898119894 is the lumpedmass of the 119894th floor 119894(119905) is the acceleration response of the119894th floor relative to the ground at the time of 119905 and 0(119905) is theground acceleration at the time of 119905

According to (1) the maximum seismic forces of CIPSand MPCS are presented and compared in Figure 11 Asillustrated in Figure 11(a) the results of them were presentedin linear behavior following the increasing earthquake waveswith PGA from 0035 g to 014 g The curves were increasedprogressively along the height of models They could reflectthe actual seismic force distribution in some extent of struc-ture in the elastic stage and the high-order modes could beignored in this time And then the cracks expanded graduallyin different ways of the twomodels and the influence of high-order modes worked The models entered into plastic stagewith obvious nonlinearity The maximum seismic forces ofthe CIPS emerged in the middle of model such as 5th 6thand 8th Those of MPCS were in 4th 5th and 8th In theplastic stage with PGA from 022 g to 062 g the distributionof seismic forces changed for the high-order modes

The interstory shear forceswere calculated by (1) and theyare illustrated along the height of models in Figure 11(b) Theinterstory shear force was increased gradually with the PGAand decreased along the height of model structure In elasticstage the distribution of the interstory shear of CIPS andMPCS showed similar rules like inverted triangle In plasticstage the vibration trend was effected by the high-ordermodes and interstory shear force was not strictlymeeting thedistribution pattern

44 Displacement Response The maximum story displace-ments of the CIPS and MPCS along the model heightobtained from the B-WSM DZC and ELW with thePGA from 0035 g to 062 g are compared and depicted inFigure 12(a) Since the shear wall structure is the effectivelateral resisting system [1] themaximum story displacementsof the models were small in 12th story arranged from 113mmto 756mm of CIPS and from 106mm to 699mm of MPCSin the elastic stageThe stage contained the PGA from 0035 gto 014 g With increasing the intensity of earthquake wavesthe two models demonstrated different damage phenomenadescribed in the above paragraph The concentrated damagein shear wall of 4th story of CIPS caused bigger displacementthan that of MPCS The bigger displacement formed thebigger torsion angle of floor and then brought about themuchbigger displacement in upside structure of CIPS And the

10 Shock and Vibration

0123456789

101112

Floo

r

01 02 03 04 05 06 07 08 09 10 1100Mode coefficient of first mode shape

0123456789

101112

Floo

r

01 02 03 04 05 06 07 08 09 10 1100Mode coefficient of first mode shape

minus08 minus06 minus04 minus02minus10 02 04 06 08 1000Mode coefficient of second mode shape

minus06 minus04 minus02minus08minus10 02 04 06 08 1000Mode coefficient of second mode shape

0123456789

101112

Floo

r

0123456789

101112

Floo

r

MPCS-initialMPCS-0035 gMPCS-007 gMPCS-014 g

MPCS-022 gMPCS-040 gMPCS-062 g

MPCS-initialMPCS-0035 gMPCS-007 gMPCS-014 g

MPCS-022 gMPCS-040 gMPCS-062 g

CIPS-initialCIPS-0035 gCIPS-007 gCIPS-014 g

CIPS-022 gCIPS-040 gCIPS-062 g

CIPS-initialCIPS-0035 gCIPS-007 gCIPS-014 g

CIPS-022 gCIPS-040 gCIPS-062 g

Figure 9 Mode shapes of CIPS and MPCS

maximumdestructive story driftwas in 4th floorMeanwhilethe two models showed obviously nonlinear behavior withthe PGA of 040 g and 062 g The shape of maximum storydisplacements of them was similar to the first mode shapeThe maximum story displacements were 4867mm in CIPSand 4198mm in MPCS

The maximum interstory drift ratios were calculated andpresented in Figure 12(b) The maximum value of story driftof CIPS was 11005 in 5th story under the DZC earthquakewave with the PGA of 070 g and that of MPCS was 11020in 5th story under the same wave They met the provisionabout the maximum story drift in elastic stage in area of 8-degree seismic fortification intensity in code of GB 50011-2010 [21] In the stage from PGA of 0035 g to 014 g thestory drift increased progressively The seismic design codeasks for the limiting value of plastic interstory drift ratio toforbid collapse We checked the story drift under the PGA of040 g The values of CIPS and MPCS were 1121 and 1127

which met the limiting value in rare earthquake action of 8-degree seismic regions Under the excitation of PGAof 062 gthe values exceeded the limiting value in rare earthquakeaction of 9-degree seismic regions And then the moreintensive earthquakewaves with PGAof 080 gwere inputtedthe models presented better seismic behavior withoutcollapse

5 Conclusion

A comparative seismic study between CIPS and MPCS wasconducted by shaking table test which contained a 15 scaled12-story CIPS model and MPCS model The experimentalresults of dynamic characteristic failure pattern and mecha-nism and seismic response of the models were discussed andcompared with each other to better understand their seismicbehavior Based on the intensive analysis of test results thefollowing conclusions are derived

Shock and Vibration 11

PGA of 062 gPGA of 040 gPGA of 022 gPGA of 014 gPGA of 007 g PGA of 062 gPGA of 040 gPGA of 022 gPGA of 014 gPGA of 007 g

EE E EB B B BDD DD

E(LW

)

D(Z

C)

B(-W

SM)

Peak ground acceleration of CIPS (g)

E EEEB B BB DDD D

E(LW

)

D(Z

C)

B(-W

SM)

Peak ground acceleration of MPCS (g)

12th floor11th floor10th floor9th floor8th floor

7th floor6th floor5th floor4th floor3rd floor

2nd floor1st floorFloor

12th floor11th floor10th floor9th floor8th floor

7th floor6th floor5th floor4th floor3rd floor

2nd floor1st floorFloor

081216202428323640444852566064

Acce

lera

tion

ampl

ifyin

g co

effici

ent

081216202428323640444852566064

Acce

lera

tion

ampl

ifyin

g co

effici

ent

(a) Acceleration amplifying factors changed along the test phases of CIPS and MPCS

0123456789

101112

Floo

r

15 20 25 30 35 40 45 50 5510Acceleration amplifying coefficient of B-WSM

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 g

MPCS-040 gMPCS-062 gCIPS-0035 gCIPS-007 g

CIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Acceleration amplifying factors of B-WSMwave along the height oftwo models

Figure 10 Acceleration amplifying factors of CIPS and MPCS

(1) The typical failure pattern of the CIPS was damageconcentration in coupling beams in first and thenthe cracks emerged in the shear wall in plasticstage However besides the concentrated damage incoupling beam the connections between the precastelement and the CIP floor were the weakness and

the vertical crack came after the horizontal crack inplastic stage The precast element with high qualitiesdid not present cracks in the test

(2) The initial natural frequency as the initial stiffnessof MPCS degraded more obviously than CIPS forthe fine cracks in the connections The damping

12 Shock and Vibration

1 2 3 4 5 6 7 8 9 10 11 120Seismic force of DZC (kN)

1 2 3 4 5 6 7 8 9 10 11 120Seismic force of B-WSM (kN)

1 2 3 4 5 6 7 8 9 10 11 12 130Seismic force of ELW (kN)

123456789

101112

Floo

r

123456789

101112

Floo

r

123456789

101112

Floo

r

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(a) Distribution of seismic forces

10 20 30 40 50 60 70 80 90 1000Story shear of ELW (kN)

0123456789

101112

Floo

r10 20 30 40 50 60 70 80 900

Story shear of B-WSM (kN)

0123456789

101112

Floo

r

0123456789

101112

Floo

r

10 20 30 40 50 60 70 800Story shear of DZC (kN)

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Distribution of story shear

Figure 11 Earthquake actions of CIPS and MPCS

ratio of them had similar parameter but the way ofenergy dissipation of them was different And theyhad similar mode shapes

(3) The acceleration amplifying factors of them increasedalong the height of models and decreased progres-sively with the increasing PGA Nevertheless theyhad different responses under the different earth-quake waves The maximum acceleration amplifyingfactor of CIPS was 546 under the ELW with PGA of022 g and that of MPCS was 523 under the B-WSMwith PGA of 040 g

(4) Distribution of seismic forces of them increasedlinearly along the height ofmodels and then presentednonlinearity for the influence of high-order modesThe distribution of the interstory shear of CIPS and

MPCS showed similar rules like inverted triangle inelastic stage

(5) The maximum story displacements of them werealmost equal in elastic stage However the maximumstory displacement of CIPS was greater than thatof MPCS in plastic stage which was caused bythe concentrated failure in 4th story of CIPS Themaximum interstory ratio under the seismic waveswith PGA of 007 g and 040 g was in accordancewith the provision of Chinese code Under the moreintensive earthquake wave of PGA of 080 g the twomodels had enough capacity to resist collapse

Conflicts of Interest

The authors declare that they have no conflicts of interest

Shock and Vibration 13

5 10 15 20 25 30 35 40 450Displacement of B-WSM (mm)

0123456789

101112

Floo

r

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of DZC (mm)

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of ELW (mm)

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(a) Maximum story displacements of CIPS and MPCS

0123456789

101112

Floo

r

000

1

001

0

000

8

000

4

001

2

000

0

000

7

000

5

000

9

000

3

001

1

000

2

000

6

Story drift of DZC

000

1

000

8

000

4

001

0

000

0

000

60

007

000

3

000

9

000

2

001

1

000

5

Story drift of B-WSM

0123456789

101112

Floo

r

0123456789

101112

Floo

r

000

1

001

2

001

0

000

4

000

8

001

3

000

0

000

6

000

9

000

5

001

1

000

30

002

000

7

Story drift of ELW

Elastic limit Plastic limitElastic limit Plastic limitElastic limit Plastic limit

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Story drift ratio of CIPS and MPCS

Figure 12 Displacement responses of CIPS and MPCS

Acknowledgments

This research has been supported by both Innovation TeamofXirsquoan University of Architecture and Technology and ShaanxiProvince Science and Technology Coordination InnovationProject (nos 2015KTZDSF03-05-01 2015KTZDSF03-04 and2014SZS04-Z01)The support of the National Natural ScienceFoundation of China (Grants nos 51408456 51578444 and51478381) and Ministry of Education Plan for Yangtze RiverScholar and Innovation TeamDevelopment (no IRT13089) isalso acknowledged

References

[1] R Park and T Paulay Reinforced Concrete Structures ohnWileyamp Sons 1975

[2] M Fintel ldquoPerformance of buildings with shear walls in earth-quakes of the last thirty yearsrdquo PCI journal vol 40 no 3 pp62ndash80 1995

[3] A A Yee ldquoSplice sleeve for reinforcing barsrdquo US Pat vol 35401968

[4] J H Ling A B Abd Rahman I S Ibrahim and Z AbdulHamid ldquoBehaviour of grouted pipe splice under incrementaltensile loadrdquo Construction and Building Materials vol 33 no8 pp 90ndash98 2012

[5] M T El-Sheikh R Sause S Pessiki and L-W Lu ldquoSeismicbehavior and design of unbonded post-tensioned precast con-crete framesrdquo PCI journal vol 44 no 3 pp 54ndash71 1999

[6] R S Henry S Sritharan and J M Ingham ldquoUnbonded tendonstresses in post-tensioned concrete walls at nominal flexuralstrengthrdquo ACI Structural Journal vol 109 no 2 pp 171ndash1822012

14 Shock and Vibration

[7] P Seifi R S Henry and JM Ingham ldquoPanel connection detailsin existing New Zealand precast concrete buildingsrdquo Bulletin ofthe New Zealand Society for Earthquake Engineering vol 49 no2 pp 190ndash199 2016

[8] K A Soudki S H Rizkalla and R W Daikiw ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformations part 2 prestressed connectionsrdquo PCI journal vol40 no 5 pp 82ndash96 1995

[9] K A Soudki S H Rizkalla and B LeBlanc ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformationsmdashpart 1 mild steel connectionsrdquo PCI journal vol40 no 4 pp 78ndash96 1995

[10] K A Soudki J S West S H Rizkalla and B BlackettldquoHorizontal connections for precast concrete shear wall panelsunder cyclic shear loadingrdquo PCI journal vol 41 no 3 pp 64ndash80 1996

[11] R Vaghei F Hejazi H Taheri M S Jaafar and A A Ali ldquoEval-uate performance of precast concrete wall to wall connectionrdquoAPCBEE Procedia vol 9 pp 285ndash290 2014

[12] H Jiang Z Chen J Zhang B Wu Y Tian andW Liu ldquoQuasi-static test of precast reinforced concrete shear wall structurerdquoJianzhu Jiegou XuebaoJournal of Building Structures vol 32 no6 pp 34ndash40 2011

[13] Z ChenH Jiang J Zhang BWu Y Tian andW Liu ldquoPseudo-dynamic substructure test on precast reinforced concrete shearwall structurerdquo Jianzhu Jiegou XuebaoJournal of Building Struc-tures vol 32 no 6 pp 41ndash50 2011

[14] J R QianW L Han Z Z Zhao and H Qin ldquoPseudo-dynamicSubstructure Test on a 3-story Full-scaleModel of PrefabricatedConcrete Shear Wall Structure with Rebars Splicing by GroutSleevesrdquo Journal of Building Structures vol 38 no 3 pp 26ndash382017

[15] P Negro D A Bournas and F J Molina ldquoPseudodynamictests on a full-scale 3-storey precast concrete building Globalresponserdquo Engineering Structures vol 57 pp 594ndash608 2013

[16] M Oliva P Gavrilovic and R W Clough ldquoSeismic testing oflarge panel precast walls Comparison of pseudostatic and shak-ing table testsrdquo Earthquake Engineering amp Structural Dynamicsvol 19 no 6 pp 859ndash875 1990

[17] L H Lee W H Yi and S Y Seo ldquoShaking Table Test of PrecastConcrete Wall Structurerdquo in Proceedings of the Eleventh WorldConference on Earthquake Engineering

[18] W Wang A Q Li et al ldquoShaking table test of precast concreteshear wall isolation structurerdquo Journal of Building Structuresvol 36 no 6 pp 35ndash43 2015

[19] EERI Preliminary Reconnaissance Report Earthquake Engi-neering Research Institute 1994

[20] EERI Armenia Earthquake Reconnaissance Report EarthquakeEngineering Research Institute 1989

[21] GB 50011-2010 Code for Seismic Design of Buildings ChinaArchitecture and Building Press Beijing China 2010

[22] JGJ 1-2014 Technical Specification for Precast Concrete Struc-tures China Architecture amp Building Press Beijing China2014

[23] 15J939-1 Example of Architectural Design for PrefabricatedConcrete Structures China Planning Press Beijing China 2015

[24] E Buckingham ldquoOn physically similar systems Illustrations ofthe use of dimensional equationsrdquo Physical Review A AtomicMolecular and Optical Physics vol 4 no 4 pp 345ndash376 1914

[25] Y Zhou andX L LuTheexperimentalmethod and technology ofshaking table test about building structure Science Press BeijingChina 2012

[26] JGJT 101-2015 Specification for Seismic Test of Buildings ChinaArchitecture amp Building Press Beijing China 2014

[27] C Y Shen F L Zhou et al ldquoExperimental research on micro-concrete used in dynamic test modelrdquo Journal of GuangzhouUniversity (Natural Science Edition) vol 4 no 3 pp 250ndash2532005

[28] J Xiao T L Pham and T Ding ldquoShake table test on seismicresponse of a precast frame with recycled aggregate concreterdquoAdvances in Structural Engineering vol 18 no 9 pp 1517ndash15342015

[29] S C Chakrabarti G C Nayak and D K Paul ldquoShear charac-teristics of cast-in-place vertical joints in story-high precast wallassemblyrdquo ACI Structural Journal vol 85 no 1 pp 30ndash45 1988

[30] CWang and J Xiao ldquoStudy of the seismic response of a recycledaggregate concrete frame structurerdquo Earthquake Engineeringand Engineering Vibration vol 12 no 4 pp 669ndash680 2013

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 4: Comparative Study of Seismic Behavior between Monolithic

4 Shock and Vibration

1 2 3

A

B

C

Coupling beam 1

PCSW-1

CIP connection-T

X

Y

N

CIP connection-L

CIP connection-+

PCSW-2 PCSW-3

Coupling beam 2 Coupling beam 3

1 2 3

A

B

C

Coupling beam 1

SW-1

X

Y

N

SW-2 SW-3

Coupling beam 2 Coupling beam 3

Cast-in-place concrete partsPrecast concrete shear wall

Opening in the shear wall

Cast-in-place structure (CIPS) Monolithic precast concrete structure (MPCS)

Edge component Edge component

80 220 500 220 160 190 80 20

8022

030

022

016

022

030

022

080

2090

090

020

160 1901800

80 220 500 220 160 190 80 20

8022

030

022

016

022

030

022

080

2090

090

020

160 1901800

Figure 1 Plane view of CIPS and MPCS

In order to supplement the missing gravity and nonstruc-tural elements iron blockswere used as the artificialmass andevenly fixed on each floor of the twomodels about 156 tonesThe total mass of each model reached 136 tones includinggrade beams and the height of the two models was 756mwhich conformed to the capacity limitation of the shakingtable system

23 Test Procedure It is well known that the condition of soilat the site is one of the significant factors in choosing theseismic inputs for shaking table test The soil layer equivalentcutting wave speed and the thickness of overlay soil layerdetermine the site classification The type-II soil site wasdefined in the Code for Seismic Design of Buildings [21]which was the condition soil of this project In comparisonwith seismic design response spectra Superstition Hillsearthquake wave (B-WSM) Kocaeli (Turkey) earthquakewave (DZC) and El Centro earthquake wave (ELW) wereselected as the ground excitations and inputted by thementioned sequenceThewaves were chosen from the PacificEarthquake Engineering Research Center (PEER) The testswere performed with unidirectional and bidirectional earth-quake waves with the ratio of PGA 1 085 to evaluate overallseismic performance of CIPS andMPCSThe test program ofthe shake table test included eight phases and the peak groundacceleration (PGA) is 0035 g 007 g 014 g 022 g 040 g062 g 070 g and 080 g in each phase respectivelyThe PGAof 070 g and 080 g were inputted to observe the nonlinearbehavior of them After each teat phase the white noise

with PGA of 0035 g was inputted to detect the dynamiticcharacteristics of the models

To monitor the earthquake response of the two models32 single-axis accelerometers including two on the shaketable two on the grade beam twenty-four on each floor in119883 and 119884 directions and four on the diagonal points in the12th floor were set to record the horizontal acceleration Atotal of 12 displacement transducers and 12 velocity vibrationsensors were installed to each floor The arrangement of testinstruments is exhibited in Figure 4 Figure 5 displays themodels on the shake table

3 Failure Pattern and Mechanism

Cracks and damage to the models were traced along withthe increasing PGA With the PGA of 0035 g diagonalmicrocracks on the coupling beam were merely observedin main direction of CIPS and those of MPCS were foundin 119883 and 119884 directions of shear wall arranged from 1st to4th As the PGA increased from 007 g to 022 g cracks onthe CIPS were extended along the corner of coupling beamand some new diagonal microcrack occurred In the phasesthe existing cracks of MPCS penetrated the coupling beamand meanwhile new microcracks appeared The diagonalmicrocrackswere themain pattern of themodels in this stage

In the next stage the models presented different crackpatterns After the PGA of 040 g the cracks of CIPSconcentrated rapidly in the end of coupling beams suchas the biggest span-depth ratio one with decreased section

Shock and Vibration 5

ldquoLrdquo type

ldquoTrdquo type

Floor

Sleeve

Longitudinalreinforcement

Additional stirrup

Protruding bar

Grouting materials

Cast-in-place concreteconnection of ldquoLrdquo type Roughened surface

External precastconcrete shear wall

Additional stirrup

10040 60

5

4060

100

5

45

45

External precastconcrete shear wall

12(10)1430

Protruding bar

Longitudinalreinforcement

Cast-in-place concreteconnection of ldquoLrdquo type

External precastconcrete shear wall

Floor

Sleeve

Longitudinalreinforcement

Grouting materials

Additional stirrupProtruding bar

Cast-in-place concreteconnection of ldquoTrdquo type Roughened surface

External precastconcrete shear wall 604060

4060

45

45

5

5

16(10)1430

Additional stirrup Protruding bar

Longitudinalreinforcement

Cast-in-place concreteconnection of ldquoTrdquo type

External precastconcrete shear wall External precast

concrete shear wall

Internal precastconcrete shear wall

ldquo+rdquo type

Floor

Sleeve

Longitudinalreinforcement

Grouting materials

Additional stirrupProtruding bar

Cast-in-place concreteconnection of ldquo+rdquo type Roughened surface

Internal precastconcrete shear wall

60 40 60

45

45

5

5

6040

60

20(10)1430

Additional stirrup

Protruding bar

Longitudinalreinforcement

Cast-in-place concreteconnection of ldquo+rdquo type

Internal precastconcrete shear wall

Internal precastconcrete shear wall

(a) Reinforcement of CIP connections

Figure 2 Continued

6 Shock and Vibration

PCSW-1 PCSW -3

PCSW-2

30 60 60 50

27050 60 60 30

270

3030 30 30

90

570

250

6(10)1430

1440

14204(8)

120 740 120

10 10

Longitudinal reinforcement

Protruding bar

Longitudinal reinforcement

Protruding bar

Verticaland levelreinforcement

Coupling beam 1

Edge component

Protruding bar

90

570

30 60 60 50

230

3050 60 60 30

230

30

250

6(10)1430

1440

14204(8)

10 10

120 540 120

Longitudinal reinforcement

Protruding bar

Longitudinal reinforcement

Protruding bar

Verticaland levelreinforcement

Coupling beam 3

Protruding bar

Edge component

30 60 60 50270

50 60 60 30

270

30

90

570

30 30 30

250

6(10)1430

1440

14204(8)

10 10

120 940 120

Longitudinal reinforcement

Protruding bar

Longitudinal reinforcement

Protruding bar

Verticaland levelreinforcement

Coupling beam 2

Protruding bar

Edge component

8lowast60=

480

8lowast60=

480

8lowast60=

480

(b) Reinforcement of PCSW

Figure 2 Configuration and reinforcements of the components (unit mm)

for the plastic hinge and the smallest span-depth ratio onewith diagonal cracks On the contrary diagonal cracks ofMPCS extended slowly and the horizontal fracture betweenthe PCSW and the CIP floor as the special form occurredat the same time With the increasing seismic energy thenew cracks were successively added along the height of themodels The cracks of CIPS were like the diagonal cracksoccurring with the PGA of 0035 g and those of MPCS werehorizontal cracks occurring in horizontal connections Inorder to observe the nonlinear behavior of them the PGA of

080 gwas inputted In this phase the vertical cracks along theedge component and the group cracks in the shear wall in 4thstory occurred in CIPS We considered that the weak story ofCIPS was the 4th story The horizontal crack extended to thecast-in-place concrete and the vertical crack appeared in thecontact of precast and cast-in-place connection in 2nd and3rd stories in MPCSThe overall crack patterns of models areexhibited in Figure 6

In general the diagonal cracks were presented in the end-ing coupling beams of CIPS and MPCS The phenomenon is

Shock and Vibration 7

CIP connection

Coupling beam

PCSW elementCIP connection PCSW element

CIP floor

CIPS MPCS

Construction procedures of CIPS

Construction procedures of MPCS

Grouted sleeves

Figure 3 Construction procedures of CIPS and MPCS

Table 2 Natural frequency of CIPS and MPCS

PGAg

FrequencyHzFirst frequency Second frequency

119883-direction 119884-direction 119883-direction 119884-directionMPCS CIPS MPCS CIPS MPCS CIPS MPCS CIPS

Initial 619 513 7 569 2500 2075 2794 23250035 g 481 512 513 563 2056 2013 2244 235007 g 480 512 506 55 2006 1988 2200 2325014 g 469 456 494 537 1969 1988 2106 2206022 g 426 413 444 481 1825 1813 1981 2094040 g 279 281 375 375 1419 1356 1625 1656062 g 231 194 300 269 1163 1006 1363 1300080 g 186 155 288 254 938 942 1318 1288

the favored state They possessed different energy dissipationmechanisms under the intensity earthquake waves The cou-pling beam acted as the first line of energy dissipation Whilethe coupling beam formed a plastic hinge the shear wallchanged to a single-slice wall to dissipate energy as the secondline to forbid collapse in CIPS Besides the coupling beamthe relatively weak connections between the precast elementsplayed the new way to dissipate energy in MPCS whichwere horizontal cracks first and then the vertical cracksFurthermore they defended the lateral resisting system insome degree

4 Earthquake Response Analysis

41 Dynamic Characteristics The dynamic characteristics ofstructure include natural frequency stiffness damping ratioand mode shape They could be deduced from the whitenoise inputted after each test phase by the transfer functionThe first and second natural frequency in 119883 direction and119884 direction are shown in Table 2 Also the stiffness could becalculated with the frequency and it is presented in Figure 7[28] Initial frequency of MPCS was bigger than that ofCIPS and the same to the initial stiffness We assumed that

8 Shock and Vibration

D11DDDD11D11D11D11DDDD 11D

A1 D1

A2 V1

A3 D2

A4 V2

A5 D3

A6 V3

A7 D4

A8 V4

A9 D5

A10 V5

A11 D6

A12 D7

A13 D8 V6

A14

A15

A16

A17

A18

A19

A20

A21

A22

A23

A24

A25

A26

V7

V9

D9

V11

V8

V10

D10

V12

D12

Y

X

Figure 4 Arrangement of accelerometers displacement transduc-ers and velocity vibration sensors

CIPS MPCS

Figure 5 General view of CIPS and MPCS

the adjunction grouted sleeves and the additional stirrupsin the connections deduced the phenomenon As the firstearthquake excitation was inputted the natural frequency ofMPCS decreased about 20 which could be caused by theshrinkage and microcracks in connections as initial damage[29] With the increasing energy the influence of the initialdamage was not the main factor And then the two modelswith approximately equal frequency presented in failure stateThe curves of stiffness degradation of CIPS reduced graduallywith the increasing PGA In contrast withCIPS that ofMPCS

CIPS MPCS

Figure 6 Failure patterns of CIPS and MPCS

00

01

02

03

04

05

06

07

08

09

10

11

Stiff

ness

ratio

MPCS-XMPCS-Y

CIPS-XCIPS-Y

022 g014 g007 g 040 g0035 g 062 gPrimePGA (g)

Figure 7 Stiffness ratio versus frequency in119883 and 119884 directions

decreased evidently in the first phase and then diminishedslowly from PGA of 0035 g to 014 g Finally they hadsimilar residual stiffnesses The diverse tendency might beunderstood by the mentioned failure pattern

Damping ratio reflects the dissipation capacity of struc-ture As is shown in Figure 8 the damping ratio increasedgradually following the PGA In the first phase it was 42And then the damping ratio increased slowly before the PGAof 040 g which arranged from 42 to 50 The averagedamping ratio of each phase changed from 42 to 82in the loading process which was in the scope of cast-in-place concrete structure Nevertheless the energy dissipationmechanismofCIPS andMPCSwas different for the order anddistribution of cracks

The first and second mode shapes of the models aredescribed in Figure 9 In general the first-order mode shape

Shock and Vibration 9

000

001

002

003

004

005

006

007

008

009

010

011

Dam

ping

ratio

MPCS-XMPCS-Y

CIPS-XCIPS-Y

014 g 022 g007 g0035 g 040 g 062 gPrimePGA (g)

Figure 8 Damping ratios of CIPS and MPCS

exhibited bending deformation characteristics and the lateralstiffness was uniform distribution along the height of themodel The shapes of them were bending to 119909-axis littleby little The phenomenon would be caused by high-ordermodes And the trend of that of CIPSwasmore obvious in thePGAof 040 g and 062 g since the severe damage occurred inthe shear wall The second-order mode shapes of them weresimilar And the maximummode coefficient of second modeshape was in positon of 4th floor

42 Acceleration Response The ratio of measured accelera-tion to the corresponding input peak ground accelerationis called the acceleration amplifying factor It reflects thedynamic response of structure under earthquake excitationsThe acceleration amplifying factors along the height of themodels are described in Figure 10(a) under the B-WSMDZCand ELW ground motions for different seismic levels Obvi-ously that of the two models was gradually increased alongthe height of themodel in each test phaseWith the increasingPGA arranged from 007 g to 062 g the whole developmenttrend of that decreased gradually implying the progressivedegradation of structural stiffness In each test phase themodels presented different dynamic responses under thedifferent earthquake excitations In elastic stage with thePGA of 007 g and 014 g the CIPS had greater response toB-WSM and ELW than to DZC The MPCS had uniformresponse to the three excitations The earthquake waves withdifferent frequency-spectrum characteristics would lead tothis phenomenon The greatest response of CIPS occurredunder the ELW with PGA of 022 g and that of MPCSoccurred in the B-WSM with PGA of 040 g The diversitiesof twomodels about the acceleration amplifying factors couldbe caused by the connections between the precast elementsThe factors of CIPS and MPCS reduced from 546 to 319and from 523 to 308 in the final test phase respectivelyAs shown in Figure 10(b) the distribution feature was more

regular under the test phases with PGA from 0035 g to 014 gWhen the two models were subjected to severe damage theinfluence of high-order vibration modes increased graduallyand the acceleration amplifying factors at somemeasurementpoints no longer conformed to the distribution [30]

43 Earthquake Action The seismic force distribution char-acteristic of structure is very important reference for theaseismic design and application of the MPCS and CIPS Themaximum seismic force of the 119894th floor is derived as follows

119865119894max = 1198981198941003816100381610038161003816 119894 (119905) + 0 (119905) max

1003816100381610038161003816 (1)

where 119865119894max is the maximum seismic force 119898119894 is the lumpedmass of the 119894th floor 119894(119905) is the acceleration response of the119894th floor relative to the ground at the time of 119905 and 0(119905) is theground acceleration at the time of 119905

According to (1) the maximum seismic forces of CIPSand MPCS are presented and compared in Figure 11 Asillustrated in Figure 11(a) the results of them were presentedin linear behavior following the increasing earthquake waveswith PGA from 0035 g to 014 g The curves were increasedprogressively along the height of models They could reflectthe actual seismic force distribution in some extent of struc-ture in the elastic stage and the high-order modes could beignored in this time And then the cracks expanded graduallyin different ways of the twomodels and the influence of high-order modes worked The models entered into plastic stagewith obvious nonlinearity The maximum seismic forces ofthe CIPS emerged in the middle of model such as 5th 6thand 8th Those of MPCS were in 4th 5th and 8th In theplastic stage with PGA from 022 g to 062 g the distributionof seismic forces changed for the high-order modes

The interstory shear forceswere calculated by (1) and theyare illustrated along the height of models in Figure 11(b) Theinterstory shear force was increased gradually with the PGAand decreased along the height of model structure In elasticstage the distribution of the interstory shear of CIPS andMPCS showed similar rules like inverted triangle In plasticstage the vibration trend was effected by the high-ordermodes and interstory shear force was not strictlymeeting thedistribution pattern

44 Displacement Response The maximum story displace-ments of the CIPS and MPCS along the model heightobtained from the B-WSM DZC and ELW with thePGA from 0035 g to 062 g are compared and depicted inFigure 12(a) Since the shear wall structure is the effectivelateral resisting system [1] themaximum story displacementsof the models were small in 12th story arranged from 113mmto 756mm of CIPS and from 106mm to 699mm of MPCSin the elastic stageThe stage contained the PGA from 0035 gto 014 g With increasing the intensity of earthquake wavesthe two models demonstrated different damage phenomenadescribed in the above paragraph The concentrated damagein shear wall of 4th story of CIPS caused bigger displacementthan that of MPCS The bigger displacement formed thebigger torsion angle of floor and then brought about themuchbigger displacement in upside structure of CIPS And the

10 Shock and Vibration

0123456789

101112

Floo

r

01 02 03 04 05 06 07 08 09 10 1100Mode coefficient of first mode shape

0123456789

101112

Floo

r

01 02 03 04 05 06 07 08 09 10 1100Mode coefficient of first mode shape

minus08 minus06 minus04 minus02minus10 02 04 06 08 1000Mode coefficient of second mode shape

minus06 minus04 minus02minus08minus10 02 04 06 08 1000Mode coefficient of second mode shape

0123456789

101112

Floo

r

0123456789

101112

Floo

r

MPCS-initialMPCS-0035 gMPCS-007 gMPCS-014 g

MPCS-022 gMPCS-040 gMPCS-062 g

MPCS-initialMPCS-0035 gMPCS-007 gMPCS-014 g

MPCS-022 gMPCS-040 gMPCS-062 g

CIPS-initialCIPS-0035 gCIPS-007 gCIPS-014 g

CIPS-022 gCIPS-040 gCIPS-062 g

CIPS-initialCIPS-0035 gCIPS-007 gCIPS-014 g

CIPS-022 gCIPS-040 gCIPS-062 g

Figure 9 Mode shapes of CIPS and MPCS

maximumdestructive story driftwas in 4th floorMeanwhilethe two models showed obviously nonlinear behavior withthe PGA of 040 g and 062 g The shape of maximum storydisplacements of them was similar to the first mode shapeThe maximum story displacements were 4867mm in CIPSand 4198mm in MPCS

The maximum interstory drift ratios were calculated andpresented in Figure 12(b) The maximum value of story driftof CIPS was 11005 in 5th story under the DZC earthquakewave with the PGA of 070 g and that of MPCS was 11020in 5th story under the same wave They met the provisionabout the maximum story drift in elastic stage in area of 8-degree seismic fortification intensity in code of GB 50011-2010 [21] In the stage from PGA of 0035 g to 014 g thestory drift increased progressively The seismic design codeasks for the limiting value of plastic interstory drift ratio toforbid collapse We checked the story drift under the PGA of040 g The values of CIPS and MPCS were 1121 and 1127

which met the limiting value in rare earthquake action of 8-degree seismic regions Under the excitation of PGAof 062 gthe values exceeded the limiting value in rare earthquakeaction of 9-degree seismic regions And then the moreintensive earthquakewaves with PGAof 080 gwere inputtedthe models presented better seismic behavior withoutcollapse

5 Conclusion

A comparative seismic study between CIPS and MPCS wasconducted by shaking table test which contained a 15 scaled12-story CIPS model and MPCS model The experimentalresults of dynamic characteristic failure pattern and mecha-nism and seismic response of the models were discussed andcompared with each other to better understand their seismicbehavior Based on the intensive analysis of test results thefollowing conclusions are derived

Shock and Vibration 11

PGA of 062 gPGA of 040 gPGA of 022 gPGA of 014 gPGA of 007 g PGA of 062 gPGA of 040 gPGA of 022 gPGA of 014 gPGA of 007 g

EE E EB B B BDD DD

E(LW

)

D(Z

C)

B(-W

SM)

Peak ground acceleration of CIPS (g)

E EEEB B BB DDD D

E(LW

)

D(Z

C)

B(-W

SM)

Peak ground acceleration of MPCS (g)

12th floor11th floor10th floor9th floor8th floor

7th floor6th floor5th floor4th floor3rd floor

2nd floor1st floorFloor

12th floor11th floor10th floor9th floor8th floor

7th floor6th floor5th floor4th floor3rd floor

2nd floor1st floorFloor

081216202428323640444852566064

Acce

lera

tion

ampl

ifyin

g co

effici

ent

081216202428323640444852566064

Acce

lera

tion

ampl

ifyin

g co

effici

ent

(a) Acceleration amplifying factors changed along the test phases of CIPS and MPCS

0123456789

101112

Floo

r

15 20 25 30 35 40 45 50 5510Acceleration amplifying coefficient of B-WSM

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 g

MPCS-040 gMPCS-062 gCIPS-0035 gCIPS-007 g

CIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Acceleration amplifying factors of B-WSMwave along the height oftwo models

Figure 10 Acceleration amplifying factors of CIPS and MPCS

(1) The typical failure pattern of the CIPS was damageconcentration in coupling beams in first and thenthe cracks emerged in the shear wall in plasticstage However besides the concentrated damage incoupling beam the connections between the precastelement and the CIP floor were the weakness and

the vertical crack came after the horizontal crack inplastic stage The precast element with high qualitiesdid not present cracks in the test

(2) The initial natural frequency as the initial stiffnessof MPCS degraded more obviously than CIPS forthe fine cracks in the connections The damping

12 Shock and Vibration

1 2 3 4 5 6 7 8 9 10 11 120Seismic force of DZC (kN)

1 2 3 4 5 6 7 8 9 10 11 120Seismic force of B-WSM (kN)

1 2 3 4 5 6 7 8 9 10 11 12 130Seismic force of ELW (kN)

123456789

101112

Floo

r

123456789

101112

Floo

r

123456789

101112

Floo

r

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(a) Distribution of seismic forces

10 20 30 40 50 60 70 80 90 1000Story shear of ELW (kN)

0123456789

101112

Floo

r10 20 30 40 50 60 70 80 900

Story shear of B-WSM (kN)

0123456789

101112

Floo

r

0123456789

101112

Floo

r

10 20 30 40 50 60 70 800Story shear of DZC (kN)

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Distribution of story shear

Figure 11 Earthquake actions of CIPS and MPCS

ratio of them had similar parameter but the way ofenergy dissipation of them was different And theyhad similar mode shapes

(3) The acceleration amplifying factors of them increasedalong the height of models and decreased progres-sively with the increasing PGA Nevertheless theyhad different responses under the different earth-quake waves The maximum acceleration amplifyingfactor of CIPS was 546 under the ELW with PGA of022 g and that of MPCS was 523 under the B-WSMwith PGA of 040 g

(4) Distribution of seismic forces of them increasedlinearly along the height ofmodels and then presentednonlinearity for the influence of high-order modesThe distribution of the interstory shear of CIPS and

MPCS showed similar rules like inverted triangle inelastic stage

(5) The maximum story displacements of them werealmost equal in elastic stage However the maximumstory displacement of CIPS was greater than thatof MPCS in plastic stage which was caused bythe concentrated failure in 4th story of CIPS Themaximum interstory ratio under the seismic waveswith PGA of 007 g and 040 g was in accordancewith the provision of Chinese code Under the moreintensive earthquake wave of PGA of 080 g the twomodels had enough capacity to resist collapse

Conflicts of Interest

The authors declare that they have no conflicts of interest

Shock and Vibration 13

5 10 15 20 25 30 35 40 450Displacement of B-WSM (mm)

0123456789

101112

Floo

r

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of DZC (mm)

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of ELW (mm)

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(a) Maximum story displacements of CIPS and MPCS

0123456789

101112

Floo

r

000

1

001

0

000

8

000

4

001

2

000

0

000

7

000

5

000

9

000

3

001

1

000

2

000

6

Story drift of DZC

000

1

000

8

000

4

001

0

000

0

000

60

007

000

3

000

9

000

2

001

1

000

5

Story drift of B-WSM

0123456789

101112

Floo

r

0123456789

101112

Floo

r

000

1

001

2

001

0

000

4

000

8

001

3

000

0

000

6

000

9

000

5

001

1

000

30

002

000

7

Story drift of ELW

Elastic limit Plastic limitElastic limit Plastic limitElastic limit Plastic limit

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Story drift ratio of CIPS and MPCS

Figure 12 Displacement responses of CIPS and MPCS

Acknowledgments

This research has been supported by both Innovation TeamofXirsquoan University of Architecture and Technology and ShaanxiProvince Science and Technology Coordination InnovationProject (nos 2015KTZDSF03-05-01 2015KTZDSF03-04 and2014SZS04-Z01)The support of the National Natural ScienceFoundation of China (Grants nos 51408456 51578444 and51478381) and Ministry of Education Plan for Yangtze RiverScholar and Innovation TeamDevelopment (no IRT13089) isalso acknowledged

References

[1] R Park and T Paulay Reinforced Concrete Structures ohnWileyamp Sons 1975

[2] M Fintel ldquoPerformance of buildings with shear walls in earth-quakes of the last thirty yearsrdquo PCI journal vol 40 no 3 pp62ndash80 1995

[3] A A Yee ldquoSplice sleeve for reinforcing barsrdquo US Pat vol 35401968

[4] J H Ling A B Abd Rahman I S Ibrahim and Z AbdulHamid ldquoBehaviour of grouted pipe splice under incrementaltensile loadrdquo Construction and Building Materials vol 33 no8 pp 90ndash98 2012

[5] M T El-Sheikh R Sause S Pessiki and L-W Lu ldquoSeismicbehavior and design of unbonded post-tensioned precast con-crete framesrdquo PCI journal vol 44 no 3 pp 54ndash71 1999

[6] R S Henry S Sritharan and J M Ingham ldquoUnbonded tendonstresses in post-tensioned concrete walls at nominal flexuralstrengthrdquo ACI Structural Journal vol 109 no 2 pp 171ndash1822012

14 Shock and Vibration

[7] P Seifi R S Henry and JM Ingham ldquoPanel connection detailsin existing New Zealand precast concrete buildingsrdquo Bulletin ofthe New Zealand Society for Earthquake Engineering vol 49 no2 pp 190ndash199 2016

[8] K A Soudki S H Rizkalla and R W Daikiw ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformations part 2 prestressed connectionsrdquo PCI journal vol40 no 5 pp 82ndash96 1995

[9] K A Soudki S H Rizkalla and B LeBlanc ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformationsmdashpart 1 mild steel connectionsrdquo PCI journal vol40 no 4 pp 78ndash96 1995

[10] K A Soudki J S West S H Rizkalla and B BlackettldquoHorizontal connections for precast concrete shear wall panelsunder cyclic shear loadingrdquo PCI journal vol 41 no 3 pp 64ndash80 1996

[11] R Vaghei F Hejazi H Taheri M S Jaafar and A A Ali ldquoEval-uate performance of precast concrete wall to wall connectionrdquoAPCBEE Procedia vol 9 pp 285ndash290 2014

[12] H Jiang Z Chen J Zhang B Wu Y Tian andW Liu ldquoQuasi-static test of precast reinforced concrete shear wall structurerdquoJianzhu Jiegou XuebaoJournal of Building Structures vol 32 no6 pp 34ndash40 2011

[13] Z ChenH Jiang J Zhang BWu Y Tian andW Liu ldquoPseudo-dynamic substructure test on precast reinforced concrete shearwall structurerdquo Jianzhu Jiegou XuebaoJournal of Building Struc-tures vol 32 no 6 pp 41ndash50 2011

[14] J R QianW L Han Z Z Zhao and H Qin ldquoPseudo-dynamicSubstructure Test on a 3-story Full-scaleModel of PrefabricatedConcrete Shear Wall Structure with Rebars Splicing by GroutSleevesrdquo Journal of Building Structures vol 38 no 3 pp 26ndash382017

[15] P Negro D A Bournas and F J Molina ldquoPseudodynamictests on a full-scale 3-storey precast concrete building Globalresponserdquo Engineering Structures vol 57 pp 594ndash608 2013

[16] M Oliva P Gavrilovic and R W Clough ldquoSeismic testing oflarge panel precast walls Comparison of pseudostatic and shak-ing table testsrdquo Earthquake Engineering amp Structural Dynamicsvol 19 no 6 pp 859ndash875 1990

[17] L H Lee W H Yi and S Y Seo ldquoShaking Table Test of PrecastConcrete Wall Structurerdquo in Proceedings of the Eleventh WorldConference on Earthquake Engineering

[18] W Wang A Q Li et al ldquoShaking table test of precast concreteshear wall isolation structurerdquo Journal of Building Structuresvol 36 no 6 pp 35ndash43 2015

[19] EERI Preliminary Reconnaissance Report Earthquake Engi-neering Research Institute 1994

[20] EERI Armenia Earthquake Reconnaissance Report EarthquakeEngineering Research Institute 1989

[21] GB 50011-2010 Code for Seismic Design of Buildings ChinaArchitecture and Building Press Beijing China 2010

[22] JGJ 1-2014 Technical Specification for Precast Concrete Struc-tures China Architecture amp Building Press Beijing China2014

[23] 15J939-1 Example of Architectural Design for PrefabricatedConcrete Structures China Planning Press Beijing China 2015

[24] E Buckingham ldquoOn physically similar systems Illustrations ofthe use of dimensional equationsrdquo Physical Review A AtomicMolecular and Optical Physics vol 4 no 4 pp 345ndash376 1914

[25] Y Zhou andX L LuTheexperimentalmethod and technology ofshaking table test about building structure Science Press BeijingChina 2012

[26] JGJT 101-2015 Specification for Seismic Test of Buildings ChinaArchitecture amp Building Press Beijing China 2014

[27] C Y Shen F L Zhou et al ldquoExperimental research on micro-concrete used in dynamic test modelrdquo Journal of GuangzhouUniversity (Natural Science Edition) vol 4 no 3 pp 250ndash2532005

[28] J Xiao T L Pham and T Ding ldquoShake table test on seismicresponse of a precast frame with recycled aggregate concreterdquoAdvances in Structural Engineering vol 18 no 9 pp 1517ndash15342015

[29] S C Chakrabarti G C Nayak and D K Paul ldquoShear charac-teristics of cast-in-place vertical joints in story-high precast wallassemblyrdquo ACI Structural Journal vol 85 no 1 pp 30ndash45 1988

[30] CWang and J Xiao ldquoStudy of the seismic response of a recycledaggregate concrete frame structurerdquo Earthquake Engineeringand Engineering Vibration vol 12 no 4 pp 669ndash680 2013

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 5: Comparative Study of Seismic Behavior between Monolithic

Shock and Vibration 5

ldquoLrdquo type

ldquoTrdquo type

Floor

Sleeve

Longitudinalreinforcement

Additional stirrup

Protruding bar

Grouting materials

Cast-in-place concreteconnection of ldquoLrdquo type Roughened surface

External precastconcrete shear wall

Additional stirrup

10040 60

5

4060

100

5

45

45

External precastconcrete shear wall

12(10)1430

Protruding bar

Longitudinalreinforcement

Cast-in-place concreteconnection of ldquoLrdquo type

External precastconcrete shear wall

Floor

Sleeve

Longitudinalreinforcement

Grouting materials

Additional stirrupProtruding bar

Cast-in-place concreteconnection of ldquoTrdquo type Roughened surface

External precastconcrete shear wall 604060

4060

45

45

5

5

16(10)1430

Additional stirrup Protruding bar

Longitudinalreinforcement

Cast-in-place concreteconnection of ldquoTrdquo type

External precastconcrete shear wall External precast

concrete shear wall

Internal precastconcrete shear wall

ldquo+rdquo type

Floor

Sleeve

Longitudinalreinforcement

Grouting materials

Additional stirrupProtruding bar

Cast-in-place concreteconnection of ldquo+rdquo type Roughened surface

Internal precastconcrete shear wall

60 40 60

45

45

5

5

6040

60

20(10)1430

Additional stirrup

Protruding bar

Longitudinalreinforcement

Cast-in-place concreteconnection of ldquo+rdquo type

Internal precastconcrete shear wall

Internal precastconcrete shear wall

(a) Reinforcement of CIP connections

Figure 2 Continued

6 Shock and Vibration

PCSW-1 PCSW -3

PCSW-2

30 60 60 50

27050 60 60 30

270

3030 30 30

90

570

250

6(10)1430

1440

14204(8)

120 740 120

10 10

Longitudinal reinforcement

Protruding bar

Longitudinal reinforcement

Protruding bar

Verticaland levelreinforcement

Coupling beam 1

Edge component

Protruding bar

90

570

30 60 60 50

230

3050 60 60 30

230

30

250

6(10)1430

1440

14204(8)

10 10

120 540 120

Longitudinal reinforcement

Protruding bar

Longitudinal reinforcement

Protruding bar

Verticaland levelreinforcement

Coupling beam 3

Protruding bar

Edge component

30 60 60 50270

50 60 60 30

270

30

90

570

30 30 30

250

6(10)1430

1440

14204(8)

10 10

120 940 120

Longitudinal reinforcement

Protruding bar

Longitudinal reinforcement

Protruding bar

Verticaland levelreinforcement

Coupling beam 2

Protruding bar

Edge component

8lowast60=

480

8lowast60=

480

8lowast60=

480

(b) Reinforcement of PCSW

Figure 2 Configuration and reinforcements of the components (unit mm)

for the plastic hinge and the smallest span-depth ratio onewith diagonal cracks On the contrary diagonal cracks ofMPCS extended slowly and the horizontal fracture betweenthe PCSW and the CIP floor as the special form occurredat the same time With the increasing seismic energy thenew cracks were successively added along the height of themodels The cracks of CIPS were like the diagonal cracksoccurring with the PGA of 0035 g and those of MPCS werehorizontal cracks occurring in horizontal connections Inorder to observe the nonlinear behavior of them the PGA of

080 gwas inputted In this phase the vertical cracks along theedge component and the group cracks in the shear wall in 4thstory occurred in CIPS We considered that the weak story ofCIPS was the 4th story The horizontal crack extended to thecast-in-place concrete and the vertical crack appeared in thecontact of precast and cast-in-place connection in 2nd and3rd stories in MPCSThe overall crack patterns of models areexhibited in Figure 6

In general the diagonal cracks were presented in the end-ing coupling beams of CIPS and MPCS The phenomenon is

Shock and Vibration 7

CIP connection

Coupling beam

PCSW elementCIP connection PCSW element

CIP floor

CIPS MPCS

Construction procedures of CIPS

Construction procedures of MPCS

Grouted sleeves

Figure 3 Construction procedures of CIPS and MPCS

Table 2 Natural frequency of CIPS and MPCS

PGAg

FrequencyHzFirst frequency Second frequency

119883-direction 119884-direction 119883-direction 119884-directionMPCS CIPS MPCS CIPS MPCS CIPS MPCS CIPS

Initial 619 513 7 569 2500 2075 2794 23250035 g 481 512 513 563 2056 2013 2244 235007 g 480 512 506 55 2006 1988 2200 2325014 g 469 456 494 537 1969 1988 2106 2206022 g 426 413 444 481 1825 1813 1981 2094040 g 279 281 375 375 1419 1356 1625 1656062 g 231 194 300 269 1163 1006 1363 1300080 g 186 155 288 254 938 942 1318 1288

the favored state They possessed different energy dissipationmechanisms under the intensity earthquake waves The cou-pling beam acted as the first line of energy dissipation Whilethe coupling beam formed a plastic hinge the shear wallchanged to a single-slice wall to dissipate energy as the secondline to forbid collapse in CIPS Besides the coupling beamthe relatively weak connections between the precast elementsplayed the new way to dissipate energy in MPCS whichwere horizontal cracks first and then the vertical cracksFurthermore they defended the lateral resisting system insome degree

4 Earthquake Response Analysis

41 Dynamic Characteristics The dynamic characteristics ofstructure include natural frequency stiffness damping ratioand mode shape They could be deduced from the whitenoise inputted after each test phase by the transfer functionThe first and second natural frequency in 119883 direction and119884 direction are shown in Table 2 Also the stiffness could becalculated with the frequency and it is presented in Figure 7[28] Initial frequency of MPCS was bigger than that ofCIPS and the same to the initial stiffness We assumed that

8 Shock and Vibration

D11DDDD11D11D11D11DDDD 11D

A1 D1

A2 V1

A3 D2

A4 V2

A5 D3

A6 V3

A7 D4

A8 V4

A9 D5

A10 V5

A11 D6

A12 D7

A13 D8 V6

A14

A15

A16

A17

A18

A19

A20

A21

A22

A23

A24

A25

A26

V7

V9

D9

V11

V8

V10

D10

V12

D12

Y

X

Figure 4 Arrangement of accelerometers displacement transduc-ers and velocity vibration sensors

CIPS MPCS

Figure 5 General view of CIPS and MPCS

the adjunction grouted sleeves and the additional stirrupsin the connections deduced the phenomenon As the firstearthquake excitation was inputted the natural frequency ofMPCS decreased about 20 which could be caused by theshrinkage and microcracks in connections as initial damage[29] With the increasing energy the influence of the initialdamage was not the main factor And then the two modelswith approximately equal frequency presented in failure stateThe curves of stiffness degradation of CIPS reduced graduallywith the increasing PGA In contrast withCIPS that ofMPCS

CIPS MPCS

Figure 6 Failure patterns of CIPS and MPCS

00

01

02

03

04

05

06

07

08

09

10

11

Stiff

ness

ratio

MPCS-XMPCS-Y

CIPS-XCIPS-Y

022 g014 g007 g 040 g0035 g 062 gPrimePGA (g)

Figure 7 Stiffness ratio versus frequency in119883 and 119884 directions

decreased evidently in the first phase and then diminishedslowly from PGA of 0035 g to 014 g Finally they hadsimilar residual stiffnesses The diverse tendency might beunderstood by the mentioned failure pattern

Damping ratio reflects the dissipation capacity of struc-ture As is shown in Figure 8 the damping ratio increasedgradually following the PGA In the first phase it was 42And then the damping ratio increased slowly before the PGAof 040 g which arranged from 42 to 50 The averagedamping ratio of each phase changed from 42 to 82in the loading process which was in the scope of cast-in-place concrete structure Nevertheless the energy dissipationmechanismofCIPS andMPCSwas different for the order anddistribution of cracks

The first and second mode shapes of the models aredescribed in Figure 9 In general the first-order mode shape

Shock and Vibration 9

000

001

002

003

004

005

006

007

008

009

010

011

Dam

ping

ratio

MPCS-XMPCS-Y

CIPS-XCIPS-Y

014 g 022 g007 g0035 g 040 g 062 gPrimePGA (g)

Figure 8 Damping ratios of CIPS and MPCS

exhibited bending deformation characteristics and the lateralstiffness was uniform distribution along the height of themodel The shapes of them were bending to 119909-axis littleby little The phenomenon would be caused by high-ordermodes And the trend of that of CIPSwasmore obvious in thePGAof 040 g and 062 g since the severe damage occurred inthe shear wall The second-order mode shapes of them weresimilar And the maximummode coefficient of second modeshape was in positon of 4th floor

42 Acceleration Response The ratio of measured accelera-tion to the corresponding input peak ground accelerationis called the acceleration amplifying factor It reflects thedynamic response of structure under earthquake excitationsThe acceleration amplifying factors along the height of themodels are described in Figure 10(a) under the B-WSMDZCand ELW ground motions for different seismic levels Obvi-ously that of the two models was gradually increased alongthe height of themodel in each test phaseWith the increasingPGA arranged from 007 g to 062 g the whole developmenttrend of that decreased gradually implying the progressivedegradation of structural stiffness In each test phase themodels presented different dynamic responses under thedifferent earthquake excitations In elastic stage with thePGA of 007 g and 014 g the CIPS had greater response toB-WSM and ELW than to DZC The MPCS had uniformresponse to the three excitations The earthquake waves withdifferent frequency-spectrum characteristics would lead tothis phenomenon The greatest response of CIPS occurredunder the ELW with PGA of 022 g and that of MPCSoccurred in the B-WSM with PGA of 040 g The diversitiesof twomodels about the acceleration amplifying factors couldbe caused by the connections between the precast elementsThe factors of CIPS and MPCS reduced from 546 to 319and from 523 to 308 in the final test phase respectivelyAs shown in Figure 10(b) the distribution feature was more

regular under the test phases with PGA from 0035 g to 014 gWhen the two models were subjected to severe damage theinfluence of high-order vibration modes increased graduallyand the acceleration amplifying factors at somemeasurementpoints no longer conformed to the distribution [30]

43 Earthquake Action The seismic force distribution char-acteristic of structure is very important reference for theaseismic design and application of the MPCS and CIPS Themaximum seismic force of the 119894th floor is derived as follows

119865119894max = 1198981198941003816100381610038161003816 119894 (119905) + 0 (119905) max

1003816100381610038161003816 (1)

where 119865119894max is the maximum seismic force 119898119894 is the lumpedmass of the 119894th floor 119894(119905) is the acceleration response of the119894th floor relative to the ground at the time of 119905 and 0(119905) is theground acceleration at the time of 119905

According to (1) the maximum seismic forces of CIPSand MPCS are presented and compared in Figure 11 Asillustrated in Figure 11(a) the results of them were presentedin linear behavior following the increasing earthquake waveswith PGA from 0035 g to 014 g The curves were increasedprogressively along the height of models They could reflectthe actual seismic force distribution in some extent of struc-ture in the elastic stage and the high-order modes could beignored in this time And then the cracks expanded graduallyin different ways of the twomodels and the influence of high-order modes worked The models entered into plastic stagewith obvious nonlinearity The maximum seismic forces ofthe CIPS emerged in the middle of model such as 5th 6thand 8th Those of MPCS were in 4th 5th and 8th In theplastic stage with PGA from 022 g to 062 g the distributionof seismic forces changed for the high-order modes

The interstory shear forceswere calculated by (1) and theyare illustrated along the height of models in Figure 11(b) Theinterstory shear force was increased gradually with the PGAand decreased along the height of model structure In elasticstage the distribution of the interstory shear of CIPS andMPCS showed similar rules like inverted triangle In plasticstage the vibration trend was effected by the high-ordermodes and interstory shear force was not strictlymeeting thedistribution pattern

44 Displacement Response The maximum story displace-ments of the CIPS and MPCS along the model heightobtained from the B-WSM DZC and ELW with thePGA from 0035 g to 062 g are compared and depicted inFigure 12(a) Since the shear wall structure is the effectivelateral resisting system [1] themaximum story displacementsof the models were small in 12th story arranged from 113mmto 756mm of CIPS and from 106mm to 699mm of MPCSin the elastic stageThe stage contained the PGA from 0035 gto 014 g With increasing the intensity of earthquake wavesthe two models demonstrated different damage phenomenadescribed in the above paragraph The concentrated damagein shear wall of 4th story of CIPS caused bigger displacementthan that of MPCS The bigger displacement formed thebigger torsion angle of floor and then brought about themuchbigger displacement in upside structure of CIPS And the

10 Shock and Vibration

0123456789

101112

Floo

r

01 02 03 04 05 06 07 08 09 10 1100Mode coefficient of first mode shape

0123456789

101112

Floo

r

01 02 03 04 05 06 07 08 09 10 1100Mode coefficient of first mode shape

minus08 minus06 minus04 minus02minus10 02 04 06 08 1000Mode coefficient of second mode shape

minus06 minus04 minus02minus08minus10 02 04 06 08 1000Mode coefficient of second mode shape

0123456789

101112

Floo

r

0123456789

101112

Floo

r

MPCS-initialMPCS-0035 gMPCS-007 gMPCS-014 g

MPCS-022 gMPCS-040 gMPCS-062 g

MPCS-initialMPCS-0035 gMPCS-007 gMPCS-014 g

MPCS-022 gMPCS-040 gMPCS-062 g

CIPS-initialCIPS-0035 gCIPS-007 gCIPS-014 g

CIPS-022 gCIPS-040 gCIPS-062 g

CIPS-initialCIPS-0035 gCIPS-007 gCIPS-014 g

CIPS-022 gCIPS-040 gCIPS-062 g

Figure 9 Mode shapes of CIPS and MPCS

maximumdestructive story driftwas in 4th floorMeanwhilethe two models showed obviously nonlinear behavior withthe PGA of 040 g and 062 g The shape of maximum storydisplacements of them was similar to the first mode shapeThe maximum story displacements were 4867mm in CIPSand 4198mm in MPCS

The maximum interstory drift ratios were calculated andpresented in Figure 12(b) The maximum value of story driftof CIPS was 11005 in 5th story under the DZC earthquakewave with the PGA of 070 g and that of MPCS was 11020in 5th story under the same wave They met the provisionabout the maximum story drift in elastic stage in area of 8-degree seismic fortification intensity in code of GB 50011-2010 [21] In the stage from PGA of 0035 g to 014 g thestory drift increased progressively The seismic design codeasks for the limiting value of plastic interstory drift ratio toforbid collapse We checked the story drift under the PGA of040 g The values of CIPS and MPCS were 1121 and 1127

which met the limiting value in rare earthquake action of 8-degree seismic regions Under the excitation of PGAof 062 gthe values exceeded the limiting value in rare earthquakeaction of 9-degree seismic regions And then the moreintensive earthquakewaves with PGAof 080 gwere inputtedthe models presented better seismic behavior withoutcollapse

5 Conclusion

A comparative seismic study between CIPS and MPCS wasconducted by shaking table test which contained a 15 scaled12-story CIPS model and MPCS model The experimentalresults of dynamic characteristic failure pattern and mecha-nism and seismic response of the models were discussed andcompared with each other to better understand their seismicbehavior Based on the intensive analysis of test results thefollowing conclusions are derived

Shock and Vibration 11

PGA of 062 gPGA of 040 gPGA of 022 gPGA of 014 gPGA of 007 g PGA of 062 gPGA of 040 gPGA of 022 gPGA of 014 gPGA of 007 g

EE E EB B B BDD DD

E(LW

)

D(Z

C)

B(-W

SM)

Peak ground acceleration of CIPS (g)

E EEEB B BB DDD D

E(LW

)

D(Z

C)

B(-W

SM)

Peak ground acceleration of MPCS (g)

12th floor11th floor10th floor9th floor8th floor

7th floor6th floor5th floor4th floor3rd floor

2nd floor1st floorFloor

12th floor11th floor10th floor9th floor8th floor

7th floor6th floor5th floor4th floor3rd floor

2nd floor1st floorFloor

081216202428323640444852566064

Acce

lera

tion

ampl

ifyin

g co

effici

ent

081216202428323640444852566064

Acce

lera

tion

ampl

ifyin

g co

effici

ent

(a) Acceleration amplifying factors changed along the test phases of CIPS and MPCS

0123456789

101112

Floo

r

15 20 25 30 35 40 45 50 5510Acceleration amplifying coefficient of B-WSM

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 g

MPCS-040 gMPCS-062 gCIPS-0035 gCIPS-007 g

CIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Acceleration amplifying factors of B-WSMwave along the height oftwo models

Figure 10 Acceleration amplifying factors of CIPS and MPCS

(1) The typical failure pattern of the CIPS was damageconcentration in coupling beams in first and thenthe cracks emerged in the shear wall in plasticstage However besides the concentrated damage incoupling beam the connections between the precastelement and the CIP floor were the weakness and

the vertical crack came after the horizontal crack inplastic stage The precast element with high qualitiesdid not present cracks in the test

(2) The initial natural frequency as the initial stiffnessof MPCS degraded more obviously than CIPS forthe fine cracks in the connections The damping

12 Shock and Vibration

1 2 3 4 5 6 7 8 9 10 11 120Seismic force of DZC (kN)

1 2 3 4 5 6 7 8 9 10 11 120Seismic force of B-WSM (kN)

1 2 3 4 5 6 7 8 9 10 11 12 130Seismic force of ELW (kN)

123456789

101112

Floo

r

123456789

101112

Floo

r

123456789

101112

Floo

r

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(a) Distribution of seismic forces

10 20 30 40 50 60 70 80 90 1000Story shear of ELW (kN)

0123456789

101112

Floo

r10 20 30 40 50 60 70 80 900

Story shear of B-WSM (kN)

0123456789

101112

Floo

r

0123456789

101112

Floo

r

10 20 30 40 50 60 70 800Story shear of DZC (kN)

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Distribution of story shear

Figure 11 Earthquake actions of CIPS and MPCS

ratio of them had similar parameter but the way ofenergy dissipation of them was different And theyhad similar mode shapes

(3) The acceleration amplifying factors of them increasedalong the height of models and decreased progres-sively with the increasing PGA Nevertheless theyhad different responses under the different earth-quake waves The maximum acceleration amplifyingfactor of CIPS was 546 under the ELW with PGA of022 g and that of MPCS was 523 under the B-WSMwith PGA of 040 g

(4) Distribution of seismic forces of them increasedlinearly along the height ofmodels and then presentednonlinearity for the influence of high-order modesThe distribution of the interstory shear of CIPS and

MPCS showed similar rules like inverted triangle inelastic stage

(5) The maximum story displacements of them werealmost equal in elastic stage However the maximumstory displacement of CIPS was greater than thatof MPCS in plastic stage which was caused bythe concentrated failure in 4th story of CIPS Themaximum interstory ratio under the seismic waveswith PGA of 007 g and 040 g was in accordancewith the provision of Chinese code Under the moreintensive earthquake wave of PGA of 080 g the twomodels had enough capacity to resist collapse

Conflicts of Interest

The authors declare that they have no conflicts of interest

Shock and Vibration 13

5 10 15 20 25 30 35 40 450Displacement of B-WSM (mm)

0123456789

101112

Floo

r

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of DZC (mm)

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of ELW (mm)

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(a) Maximum story displacements of CIPS and MPCS

0123456789

101112

Floo

r

000

1

001

0

000

8

000

4

001

2

000

0

000

7

000

5

000

9

000

3

001

1

000

2

000

6

Story drift of DZC

000

1

000

8

000

4

001

0

000

0

000

60

007

000

3

000

9

000

2

001

1

000

5

Story drift of B-WSM

0123456789

101112

Floo

r

0123456789

101112

Floo

r

000

1

001

2

001

0

000

4

000

8

001

3

000

0

000

6

000

9

000

5

001

1

000

30

002

000

7

Story drift of ELW

Elastic limit Plastic limitElastic limit Plastic limitElastic limit Plastic limit

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Story drift ratio of CIPS and MPCS

Figure 12 Displacement responses of CIPS and MPCS

Acknowledgments

This research has been supported by both Innovation TeamofXirsquoan University of Architecture and Technology and ShaanxiProvince Science and Technology Coordination InnovationProject (nos 2015KTZDSF03-05-01 2015KTZDSF03-04 and2014SZS04-Z01)The support of the National Natural ScienceFoundation of China (Grants nos 51408456 51578444 and51478381) and Ministry of Education Plan for Yangtze RiverScholar and Innovation TeamDevelopment (no IRT13089) isalso acknowledged

References

[1] R Park and T Paulay Reinforced Concrete Structures ohnWileyamp Sons 1975

[2] M Fintel ldquoPerformance of buildings with shear walls in earth-quakes of the last thirty yearsrdquo PCI journal vol 40 no 3 pp62ndash80 1995

[3] A A Yee ldquoSplice sleeve for reinforcing barsrdquo US Pat vol 35401968

[4] J H Ling A B Abd Rahman I S Ibrahim and Z AbdulHamid ldquoBehaviour of grouted pipe splice under incrementaltensile loadrdquo Construction and Building Materials vol 33 no8 pp 90ndash98 2012

[5] M T El-Sheikh R Sause S Pessiki and L-W Lu ldquoSeismicbehavior and design of unbonded post-tensioned precast con-crete framesrdquo PCI journal vol 44 no 3 pp 54ndash71 1999

[6] R S Henry S Sritharan and J M Ingham ldquoUnbonded tendonstresses in post-tensioned concrete walls at nominal flexuralstrengthrdquo ACI Structural Journal vol 109 no 2 pp 171ndash1822012

14 Shock and Vibration

[7] P Seifi R S Henry and JM Ingham ldquoPanel connection detailsin existing New Zealand precast concrete buildingsrdquo Bulletin ofthe New Zealand Society for Earthquake Engineering vol 49 no2 pp 190ndash199 2016

[8] K A Soudki S H Rizkalla and R W Daikiw ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformations part 2 prestressed connectionsrdquo PCI journal vol40 no 5 pp 82ndash96 1995

[9] K A Soudki S H Rizkalla and B LeBlanc ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformationsmdashpart 1 mild steel connectionsrdquo PCI journal vol40 no 4 pp 78ndash96 1995

[10] K A Soudki J S West S H Rizkalla and B BlackettldquoHorizontal connections for precast concrete shear wall panelsunder cyclic shear loadingrdquo PCI journal vol 41 no 3 pp 64ndash80 1996

[11] R Vaghei F Hejazi H Taheri M S Jaafar and A A Ali ldquoEval-uate performance of precast concrete wall to wall connectionrdquoAPCBEE Procedia vol 9 pp 285ndash290 2014

[12] H Jiang Z Chen J Zhang B Wu Y Tian andW Liu ldquoQuasi-static test of precast reinforced concrete shear wall structurerdquoJianzhu Jiegou XuebaoJournal of Building Structures vol 32 no6 pp 34ndash40 2011

[13] Z ChenH Jiang J Zhang BWu Y Tian andW Liu ldquoPseudo-dynamic substructure test on precast reinforced concrete shearwall structurerdquo Jianzhu Jiegou XuebaoJournal of Building Struc-tures vol 32 no 6 pp 41ndash50 2011

[14] J R QianW L Han Z Z Zhao and H Qin ldquoPseudo-dynamicSubstructure Test on a 3-story Full-scaleModel of PrefabricatedConcrete Shear Wall Structure with Rebars Splicing by GroutSleevesrdquo Journal of Building Structures vol 38 no 3 pp 26ndash382017

[15] P Negro D A Bournas and F J Molina ldquoPseudodynamictests on a full-scale 3-storey precast concrete building Globalresponserdquo Engineering Structures vol 57 pp 594ndash608 2013

[16] M Oliva P Gavrilovic and R W Clough ldquoSeismic testing oflarge panel precast walls Comparison of pseudostatic and shak-ing table testsrdquo Earthquake Engineering amp Structural Dynamicsvol 19 no 6 pp 859ndash875 1990

[17] L H Lee W H Yi and S Y Seo ldquoShaking Table Test of PrecastConcrete Wall Structurerdquo in Proceedings of the Eleventh WorldConference on Earthquake Engineering

[18] W Wang A Q Li et al ldquoShaking table test of precast concreteshear wall isolation structurerdquo Journal of Building Structuresvol 36 no 6 pp 35ndash43 2015

[19] EERI Preliminary Reconnaissance Report Earthquake Engi-neering Research Institute 1994

[20] EERI Armenia Earthquake Reconnaissance Report EarthquakeEngineering Research Institute 1989

[21] GB 50011-2010 Code for Seismic Design of Buildings ChinaArchitecture and Building Press Beijing China 2010

[22] JGJ 1-2014 Technical Specification for Precast Concrete Struc-tures China Architecture amp Building Press Beijing China2014

[23] 15J939-1 Example of Architectural Design for PrefabricatedConcrete Structures China Planning Press Beijing China 2015

[24] E Buckingham ldquoOn physically similar systems Illustrations ofthe use of dimensional equationsrdquo Physical Review A AtomicMolecular and Optical Physics vol 4 no 4 pp 345ndash376 1914

[25] Y Zhou andX L LuTheexperimentalmethod and technology ofshaking table test about building structure Science Press BeijingChina 2012

[26] JGJT 101-2015 Specification for Seismic Test of Buildings ChinaArchitecture amp Building Press Beijing China 2014

[27] C Y Shen F L Zhou et al ldquoExperimental research on micro-concrete used in dynamic test modelrdquo Journal of GuangzhouUniversity (Natural Science Edition) vol 4 no 3 pp 250ndash2532005

[28] J Xiao T L Pham and T Ding ldquoShake table test on seismicresponse of a precast frame with recycled aggregate concreterdquoAdvances in Structural Engineering vol 18 no 9 pp 1517ndash15342015

[29] S C Chakrabarti G C Nayak and D K Paul ldquoShear charac-teristics of cast-in-place vertical joints in story-high precast wallassemblyrdquo ACI Structural Journal vol 85 no 1 pp 30ndash45 1988

[30] CWang and J Xiao ldquoStudy of the seismic response of a recycledaggregate concrete frame structurerdquo Earthquake Engineeringand Engineering Vibration vol 12 no 4 pp 669ndash680 2013

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 6: Comparative Study of Seismic Behavior between Monolithic

6 Shock and Vibration

PCSW-1 PCSW -3

PCSW-2

30 60 60 50

27050 60 60 30

270

3030 30 30

90

570

250

6(10)1430

1440

14204(8)

120 740 120

10 10

Longitudinal reinforcement

Protruding bar

Longitudinal reinforcement

Protruding bar

Verticaland levelreinforcement

Coupling beam 1

Edge component

Protruding bar

90

570

30 60 60 50

230

3050 60 60 30

230

30

250

6(10)1430

1440

14204(8)

10 10

120 540 120

Longitudinal reinforcement

Protruding bar

Longitudinal reinforcement

Protruding bar

Verticaland levelreinforcement

Coupling beam 3

Protruding bar

Edge component

30 60 60 50270

50 60 60 30

270

30

90

570

30 30 30

250

6(10)1430

1440

14204(8)

10 10

120 940 120

Longitudinal reinforcement

Protruding bar

Longitudinal reinforcement

Protruding bar

Verticaland levelreinforcement

Coupling beam 2

Protruding bar

Edge component

8lowast60=

480

8lowast60=

480

8lowast60=

480

(b) Reinforcement of PCSW

Figure 2 Configuration and reinforcements of the components (unit mm)

for the plastic hinge and the smallest span-depth ratio onewith diagonal cracks On the contrary diagonal cracks ofMPCS extended slowly and the horizontal fracture betweenthe PCSW and the CIP floor as the special form occurredat the same time With the increasing seismic energy thenew cracks were successively added along the height of themodels The cracks of CIPS were like the diagonal cracksoccurring with the PGA of 0035 g and those of MPCS werehorizontal cracks occurring in horizontal connections Inorder to observe the nonlinear behavior of them the PGA of

080 gwas inputted In this phase the vertical cracks along theedge component and the group cracks in the shear wall in 4thstory occurred in CIPS We considered that the weak story ofCIPS was the 4th story The horizontal crack extended to thecast-in-place concrete and the vertical crack appeared in thecontact of precast and cast-in-place connection in 2nd and3rd stories in MPCSThe overall crack patterns of models areexhibited in Figure 6

In general the diagonal cracks were presented in the end-ing coupling beams of CIPS and MPCS The phenomenon is

Shock and Vibration 7

CIP connection

Coupling beam

PCSW elementCIP connection PCSW element

CIP floor

CIPS MPCS

Construction procedures of CIPS

Construction procedures of MPCS

Grouted sleeves

Figure 3 Construction procedures of CIPS and MPCS

Table 2 Natural frequency of CIPS and MPCS

PGAg

FrequencyHzFirst frequency Second frequency

119883-direction 119884-direction 119883-direction 119884-directionMPCS CIPS MPCS CIPS MPCS CIPS MPCS CIPS

Initial 619 513 7 569 2500 2075 2794 23250035 g 481 512 513 563 2056 2013 2244 235007 g 480 512 506 55 2006 1988 2200 2325014 g 469 456 494 537 1969 1988 2106 2206022 g 426 413 444 481 1825 1813 1981 2094040 g 279 281 375 375 1419 1356 1625 1656062 g 231 194 300 269 1163 1006 1363 1300080 g 186 155 288 254 938 942 1318 1288

the favored state They possessed different energy dissipationmechanisms under the intensity earthquake waves The cou-pling beam acted as the first line of energy dissipation Whilethe coupling beam formed a plastic hinge the shear wallchanged to a single-slice wall to dissipate energy as the secondline to forbid collapse in CIPS Besides the coupling beamthe relatively weak connections between the precast elementsplayed the new way to dissipate energy in MPCS whichwere horizontal cracks first and then the vertical cracksFurthermore they defended the lateral resisting system insome degree

4 Earthquake Response Analysis

41 Dynamic Characteristics The dynamic characteristics ofstructure include natural frequency stiffness damping ratioand mode shape They could be deduced from the whitenoise inputted after each test phase by the transfer functionThe first and second natural frequency in 119883 direction and119884 direction are shown in Table 2 Also the stiffness could becalculated with the frequency and it is presented in Figure 7[28] Initial frequency of MPCS was bigger than that ofCIPS and the same to the initial stiffness We assumed that

8 Shock and Vibration

D11DDDD11D11D11D11DDDD 11D

A1 D1

A2 V1

A3 D2

A4 V2

A5 D3

A6 V3

A7 D4

A8 V4

A9 D5

A10 V5

A11 D6

A12 D7

A13 D8 V6

A14

A15

A16

A17

A18

A19

A20

A21

A22

A23

A24

A25

A26

V7

V9

D9

V11

V8

V10

D10

V12

D12

Y

X

Figure 4 Arrangement of accelerometers displacement transduc-ers and velocity vibration sensors

CIPS MPCS

Figure 5 General view of CIPS and MPCS

the adjunction grouted sleeves and the additional stirrupsin the connections deduced the phenomenon As the firstearthquake excitation was inputted the natural frequency ofMPCS decreased about 20 which could be caused by theshrinkage and microcracks in connections as initial damage[29] With the increasing energy the influence of the initialdamage was not the main factor And then the two modelswith approximately equal frequency presented in failure stateThe curves of stiffness degradation of CIPS reduced graduallywith the increasing PGA In contrast withCIPS that ofMPCS

CIPS MPCS

Figure 6 Failure patterns of CIPS and MPCS

00

01

02

03

04

05

06

07

08

09

10

11

Stiff

ness

ratio

MPCS-XMPCS-Y

CIPS-XCIPS-Y

022 g014 g007 g 040 g0035 g 062 gPrimePGA (g)

Figure 7 Stiffness ratio versus frequency in119883 and 119884 directions

decreased evidently in the first phase and then diminishedslowly from PGA of 0035 g to 014 g Finally they hadsimilar residual stiffnesses The diverse tendency might beunderstood by the mentioned failure pattern

Damping ratio reflects the dissipation capacity of struc-ture As is shown in Figure 8 the damping ratio increasedgradually following the PGA In the first phase it was 42And then the damping ratio increased slowly before the PGAof 040 g which arranged from 42 to 50 The averagedamping ratio of each phase changed from 42 to 82in the loading process which was in the scope of cast-in-place concrete structure Nevertheless the energy dissipationmechanismofCIPS andMPCSwas different for the order anddistribution of cracks

The first and second mode shapes of the models aredescribed in Figure 9 In general the first-order mode shape

Shock and Vibration 9

000

001

002

003

004

005

006

007

008

009

010

011

Dam

ping

ratio

MPCS-XMPCS-Y

CIPS-XCIPS-Y

014 g 022 g007 g0035 g 040 g 062 gPrimePGA (g)

Figure 8 Damping ratios of CIPS and MPCS

exhibited bending deformation characteristics and the lateralstiffness was uniform distribution along the height of themodel The shapes of them were bending to 119909-axis littleby little The phenomenon would be caused by high-ordermodes And the trend of that of CIPSwasmore obvious in thePGAof 040 g and 062 g since the severe damage occurred inthe shear wall The second-order mode shapes of them weresimilar And the maximummode coefficient of second modeshape was in positon of 4th floor

42 Acceleration Response The ratio of measured accelera-tion to the corresponding input peak ground accelerationis called the acceleration amplifying factor It reflects thedynamic response of structure under earthquake excitationsThe acceleration amplifying factors along the height of themodels are described in Figure 10(a) under the B-WSMDZCand ELW ground motions for different seismic levels Obvi-ously that of the two models was gradually increased alongthe height of themodel in each test phaseWith the increasingPGA arranged from 007 g to 062 g the whole developmenttrend of that decreased gradually implying the progressivedegradation of structural stiffness In each test phase themodels presented different dynamic responses under thedifferent earthquake excitations In elastic stage with thePGA of 007 g and 014 g the CIPS had greater response toB-WSM and ELW than to DZC The MPCS had uniformresponse to the three excitations The earthquake waves withdifferent frequency-spectrum characteristics would lead tothis phenomenon The greatest response of CIPS occurredunder the ELW with PGA of 022 g and that of MPCSoccurred in the B-WSM with PGA of 040 g The diversitiesof twomodels about the acceleration amplifying factors couldbe caused by the connections between the precast elementsThe factors of CIPS and MPCS reduced from 546 to 319and from 523 to 308 in the final test phase respectivelyAs shown in Figure 10(b) the distribution feature was more

regular under the test phases with PGA from 0035 g to 014 gWhen the two models were subjected to severe damage theinfluence of high-order vibration modes increased graduallyand the acceleration amplifying factors at somemeasurementpoints no longer conformed to the distribution [30]

43 Earthquake Action The seismic force distribution char-acteristic of structure is very important reference for theaseismic design and application of the MPCS and CIPS Themaximum seismic force of the 119894th floor is derived as follows

119865119894max = 1198981198941003816100381610038161003816 119894 (119905) + 0 (119905) max

1003816100381610038161003816 (1)

where 119865119894max is the maximum seismic force 119898119894 is the lumpedmass of the 119894th floor 119894(119905) is the acceleration response of the119894th floor relative to the ground at the time of 119905 and 0(119905) is theground acceleration at the time of 119905

According to (1) the maximum seismic forces of CIPSand MPCS are presented and compared in Figure 11 Asillustrated in Figure 11(a) the results of them were presentedin linear behavior following the increasing earthquake waveswith PGA from 0035 g to 014 g The curves were increasedprogressively along the height of models They could reflectthe actual seismic force distribution in some extent of struc-ture in the elastic stage and the high-order modes could beignored in this time And then the cracks expanded graduallyin different ways of the twomodels and the influence of high-order modes worked The models entered into plastic stagewith obvious nonlinearity The maximum seismic forces ofthe CIPS emerged in the middle of model such as 5th 6thand 8th Those of MPCS were in 4th 5th and 8th In theplastic stage with PGA from 022 g to 062 g the distributionof seismic forces changed for the high-order modes

The interstory shear forceswere calculated by (1) and theyare illustrated along the height of models in Figure 11(b) Theinterstory shear force was increased gradually with the PGAand decreased along the height of model structure In elasticstage the distribution of the interstory shear of CIPS andMPCS showed similar rules like inverted triangle In plasticstage the vibration trend was effected by the high-ordermodes and interstory shear force was not strictlymeeting thedistribution pattern

44 Displacement Response The maximum story displace-ments of the CIPS and MPCS along the model heightobtained from the B-WSM DZC and ELW with thePGA from 0035 g to 062 g are compared and depicted inFigure 12(a) Since the shear wall structure is the effectivelateral resisting system [1] themaximum story displacementsof the models were small in 12th story arranged from 113mmto 756mm of CIPS and from 106mm to 699mm of MPCSin the elastic stageThe stage contained the PGA from 0035 gto 014 g With increasing the intensity of earthquake wavesthe two models demonstrated different damage phenomenadescribed in the above paragraph The concentrated damagein shear wall of 4th story of CIPS caused bigger displacementthan that of MPCS The bigger displacement formed thebigger torsion angle of floor and then brought about themuchbigger displacement in upside structure of CIPS And the

10 Shock and Vibration

0123456789

101112

Floo

r

01 02 03 04 05 06 07 08 09 10 1100Mode coefficient of first mode shape

0123456789

101112

Floo

r

01 02 03 04 05 06 07 08 09 10 1100Mode coefficient of first mode shape

minus08 minus06 minus04 minus02minus10 02 04 06 08 1000Mode coefficient of second mode shape

minus06 minus04 minus02minus08minus10 02 04 06 08 1000Mode coefficient of second mode shape

0123456789

101112

Floo

r

0123456789

101112

Floo

r

MPCS-initialMPCS-0035 gMPCS-007 gMPCS-014 g

MPCS-022 gMPCS-040 gMPCS-062 g

MPCS-initialMPCS-0035 gMPCS-007 gMPCS-014 g

MPCS-022 gMPCS-040 gMPCS-062 g

CIPS-initialCIPS-0035 gCIPS-007 gCIPS-014 g

CIPS-022 gCIPS-040 gCIPS-062 g

CIPS-initialCIPS-0035 gCIPS-007 gCIPS-014 g

CIPS-022 gCIPS-040 gCIPS-062 g

Figure 9 Mode shapes of CIPS and MPCS

maximumdestructive story driftwas in 4th floorMeanwhilethe two models showed obviously nonlinear behavior withthe PGA of 040 g and 062 g The shape of maximum storydisplacements of them was similar to the first mode shapeThe maximum story displacements were 4867mm in CIPSand 4198mm in MPCS

The maximum interstory drift ratios were calculated andpresented in Figure 12(b) The maximum value of story driftof CIPS was 11005 in 5th story under the DZC earthquakewave with the PGA of 070 g and that of MPCS was 11020in 5th story under the same wave They met the provisionabout the maximum story drift in elastic stage in area of 8-degree seismic fortification intensity in code of GB 50011-2010 [21] In the stage from PGA of 0035 g to 014 g thestory drift increased progressively The seismic design codeasks for the limiting value of plastic interstory drift ratio toforbid collapse We checked the story drift under the PGA of040 g The values of CIPS and MPCS were 1121 and 1127

which met the limiting value in rare earthquake action of 8-degree seismic regions Under the excitation of PGAof 062 gthe values exceeded the limiting value in rare earthquakeaction of 9-degree seismic regions And then the moreintensive earthquakewaves with PGAof 080 gwere inputtedthe models presented better seismic behavior withoutcollapse

5 Conclusion

A comparative seismic study between CIPS and MPCS wasconducted by shaking table test which contained a 15 scaled12-story CIPS model and MPCS model The experimentalresults of dynamic characteristic failure pattern and mecha-nism and seismic response of the models were discussed andcompared with each other to better understand their seismicbehavior Based on the intensive analysis of test results thefollowing conclusions are derived

Shock and Vibration 11

PGA of 062 gPGA of 040 gPGA of 022 gPGA of 014 gPGA of 007 g PGA of 062 gPGA of 040 gPGA of 022 gPGA of 014 gPGA of 007 g

EE E EB B B BDD DD

E(LW

)

D(Z

C)

B(-W

SM)

Peak ground acceleration of CIPS (g)

E EEEB B BB DDD D

E(LW

)

D(Z

C)

B(-W

SM)

Peak ground acceleration of MPCS (g)

12th floor11th floor10th floor9th floor8th floor

7th floor6th floor5th floor4th floor3rd floor

2nd floor1st floorFloor

12th floor11th floor10th floor9th floor8th floor

7th floor6th floor5th floor4th floor3rd floor

2nd floor1st floorFloor

081216202428323640444852566064

Acce

lera

tion

ampl

ifyin

g co

effici

ent

081216202428323640444852566064

Acce

lera

tion

ampl

ifyin

g co

effici

ent

(a) Acceleration amplifying factors changed along the test phases of CIPS and MPCS

0123456789

101112

Floo

r

15 20 25 30 35 40 45 50 5510Acceleration amplifying coefficient of B-WSM

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 g

MPCS-040 gMPCS-062 gCIPS-0035 gCIPS-007 g

CIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Acceleration amplifying factors of B-WSMwave along the height oftwo models

Figure 10 Acceleration amplifying factors of CIPS and MPCS

(1) The typical failure pattern of the CIPS was damageconcentration in coupling beams in first and thenthe cracks emerged in the shear wall in plasticstage However besides the concentrated damage incoupling beam the connections between the precastelement and the CIP floor were the weakness and

the vertical crack came after the horizontal crack inplastic stage The precast element with high qualitiesdid not present cracks in the test

(2) The initial natural frequency as the initial stiffnessof MPCS degraded more obviously than CIPS forthe fine cracks in the connections The damping

12 Shock and Vibration

1 2 3 4 5 6 7 8 9 10 11 120Seismic force of DZC (kN)

1 2 3 4 5 6 7 8 9 10 11 120Seismic force of B-WSM (kN)

1 2 3 4 5 6 7 8 9 10 11 12 130Seismic force of ELW (kN)

123456789

101112

Floo

r

123456789

101112

Floo

r

123456789

101112

Floo

r

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(a) Distribution of seismic forces

10 20 30 40 50 60 70 80 90 1000Story shear of ELW (kN)

0123456789

101112

Floo

r10 20 30 40 50 60 70 80 900

Story shear of B-WSM (kN)

0123456789

101112

Floo

r

0123456789

101112

Floo

r

10 20 30 40 50 60 70 800Story shear of DZC (kN)

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Distribution of story shear

Figure 11 Earthquake actions of CIPS and MPCS

ratio of them had similar parameter but the way ofenergy dissipation of them was different And theyhad similar mode shapes

(3) The acceleration amplifying factors of them increasedalong the height of models and decreased progres-sively with the increasing PGA Nevertheless theyhad different responses under the different earth-quake waves The maximum acceleration amplifyingfactor of CIPS was 546 under the ELW with PGA of022 g and that of MPCS was 523 under the B-WSMwith PGA of 040 g

(4) Distribution of seismic forces of them increasedlinearly along the height ofmodels and then presentednonlinearity for the influence of high-order modesThe distribution of the interstory shear of CIPS and

MPCS showed similar rules like inverted triangle inelastic stage

(5) The maximum story displacements of them werealmost equal in elastic stage However the maximumstory displacement of CIPS was greater than thatof MPCS in plastic stage which was caused bythe concentrated failure in 4th story of CIPS Themaximum interstory ratio under the seismic waveswith PGA of 007 g and 040 g was in accordancewith the provision of Chinese code Under the moreintensive earthquake wave of PGA of 080 g the twomodels had enough capacity to resist collapse

Conflicts of Interest

The authors declare that they have no conflicts of interest

Shock and Vibration 13

5 10 15 20 25 30 35 40 450Displacement of B-WSM (mm)

0123456789

101112

Floo

r

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of DZC (mm)

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of ELW (mm)

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(a) Maximum story displacements of CIPS and MPCS

0123456789

101112

Floo

r

000

1

001

0

000

8

000

4

001

2

000

0

000

7

000

5

000

9

000

3

001

1

000

2

000

6

Story drift of DZC

000

1

000

8

000

4

001

0

000

0

000

60

007

000

3

000

9

000

2

001

1

000

5

Story drift of B-WSM

0123456789

101112

Floo

r

0123456789

101112

Floo

r

000

1

001

2

001

0

000

4

000

8

001

3

000

0

000

6

000

9

000

5

001

1

000

30

002

000

7

Story drift of ELW

Elastic limit Plastic limitElastic limit Plastic limitElastic limit Plastic limit

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Story drift ratio of CIPS and MPCS

Figure 12 Displacement responses of CIPS and MPCS

Acknowledgments

This research has been supported by both Innovation TeamofXirsquoan University of Architecture and Technology and ShaanxiProvince Science and Technology Coordination InnovationProject (nos 2015KTZDSF03-05-01 2015KTZDSF03-04 and2014SZS04-Z01)The support of the National Natural ScienceFoundation of China (Grants nos 51408456 51578444 and51478381) and Ministry of Education Plan for Yangtze RiverScholar and Innovation TeamDevelopment (no IRT13089) isalso acknowledged

References

[1] R Park and T Paulay Reinforced Concrete Structures ohnWileyamp Sons 1975

[2] M Fintel ldquoPerformance of buildings with shear walls in earth-quakes of the last thirty yearsrdquo PCI journal vol 40 no 3 pp62ndash80 1995

[3] A A Yee ldquoSplice sleeve for reinforcing barsrdquo US Pat vol 35401968

[4] J H Ling A B Abd Rahman I S Ibrahim and Z AbdulHamid ldquoBehaviour of grouted pipe splice under incrementaltensile loadrdquo Construction and Building Materials vol 33 no8 pp 90ndash98 2012

[5] M T El-Sheikh R Sause S Pessiki and L-W Lu ldquoSeismicbehavior and design of unbonded post-tensioned precast con-crete framesrdquo PCI journal vol 44 no 3 pp 54ndash71 1999

[6] R S Henry S Sritharan and J M Ingham ldquoUnbonded tendonstresses in post-tensioned concrete walls at nominal flexuralstrengthrdquo ACI Structural Journal vol 109 no 2 pp 171ndash1822012

14 Shock and Vibration

[7] P Seifi R S Henry and JM Ingham ldquoPanel connection detailsin existing New Zealand precast concrete buildingsrdquo Bulletin ofthe New Zealand Society for Earthquake Engineering vol 49 no2 pp 190ndash199 2016

[8] K A Soudki S H Rizkalla and R W Daikiw ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformations part 2 prestressed connectionsrdquo PCI journal vol40 no 5 pp 82ndash96 1995

[9] K A Soudki S H Rizkalla and B LeBlanc ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformationsmdashpart 1 mild steel connectionsrdquo PCI journal vol40 no 4 pp 78ndash96 1995

[10] K A Soudki J S West S H Rizkalla and B BlackettldquoHorizontal connections for precast concrete shear wall panelsunder cyclic shear loadingrdquo PCI journal vol 41 no 3 pp 64ndash80 1996

[11] R Vaghei F Hejazi H Taheri M S Jaafar and A A Ali ldquoEval-uate performance of precast concrete wall to wall connectionrdquoAPCBEE Procedia vol 9 pp 285ndash290 2014

[12] H Jiang Z Chen J Zhang B Wu Y Tian andW Liu ldquoQuasi-static test of precast reinforced concrete shear wall structurerdquoJianzhu Jiegou XuebaoJournal of Building Structures vol 32 no6 pp 34ndash40 2011

[13] Z ChenH Jiang J Zhang BWu Y Tian andW Liu ldquoPseudo-dynamic substructure test on precast reinforced concrete shearwall structurerdquo Jianzhu Jiegou XuebaoJournal of Building Struc-tures vol 32 no 6 pp 41ndash50 2011

[14] J R QianW L Han Z Z Zhao and H Qin ldquoPseudo-dynamicSubstructure Test on a 3-story Full-scaleModel of PrefabricatedConcrete Shear Wall Structure with Rebars Splicing by GroutSleevesrdquo Journal of Building Structures vol 38 no 3 pp 26ndash382017

[15] P Negro D A Bournas and F J Molina ldquoPseudodynamictests on a full-scale 3-storey precast concrete building Globalresponserdquo Engineering Structures vol 57 pp 594ndash608 2013

[16] M Oliva P Gavrilovic and R W Clough ldquoSeismic testing oflarge panel precast walls Comparison of pseudostatic and shak-ing table testsrdquo Earthquake Engineering amp Structural Dynamicsvol 19 no 6 pp 859ndash875 1990

[17] L H Lee W H Yi and S Y Seo ldquoShaking Table Test of PrecastConcrete Wall Structurerdquo in Proceedings of the Eleventh WorldConference on Earthquake Engineering

[18] W Wang A Q Li et al ldquoShaking table test of precast concreteshear wall isolation structurerdquo Journal of Building Structuresvol 36 no 6 pp 35ndash43 2015

[19] EERI Preliminary Reconnaissance Report Earthquake Engi-neering Research Institute 1994

[20] EERI Armenia Earthquake Reconnaissance Report EarthquakeEngineering Research Institute 1989

[21] GB 50011-2010 Code for Seismic Design of Buildings ChinaArchitecture and Building Press Beijing China 2010

[22] JGJ 1-2014 Technical Specification for Precast Concrete Struc-tures China Architecture amp Building Press Beijing China2014

[23] 15J939-1 Example of Architectural Design for PrefabricatedConcrete Structures China Planning Press Beijing China 2015

[24] E Buckingham ldquoOn physically similar systems Illustrations ofthe use of dimensional equationsrdquo Physical Review A AtomicMolecular and Optical Physics vol 4 no 4 pp 345ndash376 1914

[25] Y Zhou andX L LuTheexperimentalmethod and technology ofshaking table test about building structure Science Press BeijingChina 2012

[26] JGJT 101-2015 Specification for Seismic Test of Buildings ChinaArchitecture amp Building Press Beijing China 2014

[27] C Y Shen F L Zhou et al ldquoExperimental research on micro-concrete used in dynamic test modelrdquo Journal of GuangzhouUniversity (Natural Science Edition) vol 4 no 3 pp 250ndash2532005

[28] J Xiao T L Pham and T Ding ldquoShake table test on seismicresponse of a precast frame with recycled aggregate concreterdquoAdvances in Structural Engineering vol 18 no 9 pp 1517ndash15342015

[29] S C Chakrabarti G C Nayak and D K Paul ldquoShear charac-teristics of cast-in-place vertical joints in story-high precast wallassemblyrdquo ACI Structural Journal vol 85 no 1 pp 30ndash45 1988

[30] CWang and J Xiao ldquoStudy of the seismic response of a recycledaggregate concrete frame structurerdquo Earthquake Engineeringand Engineering Vibration vol 12 no 4 pp 669ndash680 2013

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 7: Comparative Study of Seismic Behavior between Monolithic

Shock and Vibration 7

CIP connection

Coupling beam

PCSW elementCIP connection PCSW element

CIP floor

CIPS MPCS

Construction procedures of CIPS

Construction procedures of MPCS

Grouted sleeves

Figure 3 Construction procedures of CIPS and MPCS

Table 2 Natural frequency of CIPS and MPCS

PGAg

FrequencyHzFirst frequency Second frequency

119883-direction 119884-direction 119883-direction 119884-directionMPCS CIPS MPCS CIPS MPCS CIPS MPCS CIPS

Initial 619 513 7 569 2500 2075 2794 23250035 g 481 512 513 563 2056 2013 2244 235007 g 480 512 506 55 2006 1988 2200 2325014 g 469 456 494 537 1969 1988 2106 2206022 g 426 413 444 481 1825 1813 1981 2094040 g 279 281 375 375 1419 1356 1625 1656062 g 231 194 300 269 1163 1006 1363 1300080 g 186 155 288 254 938 942 1318 1288

the favored state They possessed different energy dissipationmechanisms under the intensity earthquake waves The cou-pling beam acted as the first line of energy dissipation Whilethe coupling beam formed a plastic hinge the shear wallchanged to a single-slice wall to dissipate energy as the secondline to forbid collapse in CIPS Besides the coupling beamthe relatively weak connections between the precast elementsplayed the new way to dissipate energy in MPCS whichwere horizontal cracks first and then the vertical cracksFurthermore they defended the lateral resisting system insome degree

4 Earthquake Response Analysis

41 Dynamic Characteristics The dynamic characteristics ofstructure include natural frequency stiffness damping ratioand mode shape They could be deduced from the whitenoise inputted after each test phase by the transfer functionThe first and second natural frequency in 119883 direction and119884 direction are shown in Table 2 Also the stiffness could becalculated with the frequency and it is presented in Figure 7[28] Initial frequency of MPCS was bigger than that ofCIPS and the same to the initial stiffness We assumed that

8 Shock and Vibration

D11DDDD11D11D11D11DDDD 11D

A1 D1

A2 V1

A3 D2

A4 V2

A5 D3

A6 V3

A7 D4

A8 V4

A9 D5

A10 V5

A11 D6

A12 D7

A13 D8 V6

A14

A15

A16

A17

A18

A19

A20

A21

A22

A23

A24

A25

A26

V7

V9

D9

V11

V8

V10

D10

V12

D12

Y

X

Figure 4 Arrangement of accelerometers displacement transduc-ers and velocity vibration sensors

CIPS MPCS

Figure 5 General view of CIPS and MPCS

the adjunction grouted sleeves and the additional stirrupsin the connections deduced the phenomenon As the firstearthquake excitation was inputted the natural frequency ofMPCS decreased about 20 which could be caused by theshrinkage and microcracks in connections as initial damage[29] With the increasing energy the influence of the initialdamage was not the main factor And then the two modelswith approximately equal frequency presented in failure stateThe curves of stiffness degradation of CIPS reduced graduallywith the increasing PGA In contrast withCIPS that ofMPCS

CIPS MPCS

Figure 6 Failure patterns of CIPS and MPCS

00

01

02

03

04

05

06

07

08

09

10

11

Stiff

ness

ratio

MPCS-XMPCS-Y

CIPS-XCIPS-Y

022 g014 g007 g 040 g0035 g 062 gPrimePGA (g)

Figure 7 Stiffness ratio versus frequency in119883 and 119884 directions

decreased evidently in the first phase and then diminishedslowly from PGA of 0035 g to 014 g Finally they hadsimilar residual stiffnesses The diverse tendency might beunderstood by the mentioned failure pattern

Damping ratio reflects the dissipation capacity of struc-ture As is shown in Figure 8 the damping ratio increasedgradually following the PGA In the first phase it was 42And then the damping ratio increased slowly before the PGAof 040 g which arranged from 42 to 50 The averagedamping ratio of each phase changed from 42 to 82in the loading process which was in the scope of cast-in-place concrete structure Nevertheless the energy dissipationmechanismofCIPS andMPCSwas different for the order anddistribution of cracks

The first and second mode shapes of the models aredescribed in Figure 9 In general the first-order mode shape

Shock and Vibration 9

000

001

002

003

004

005

006

007

008

009

010

011

Dam

ping

ratio

MPCS-XMPCS-Y

CIPS-XCIPS-Y

014 g 022 g007 g0035 g 040 g 062 gPrimePGA (g)

Figure 8 Damping ratios of CIPS and MPCS

exhibited bending deformation characteristics and the lateralstiffness was uniform distribution along the height of themodel The shapes of them were bending to 119909-axis littleby little The phenomenon would be caused by high-ordermodes And the trend of that of CIPSwasmore obvious in thePGAof 040 g and 062 g since the severe damage occurred inthe shear wall The second-order mode shapes of them weresimilar And the maximummode coefficient of second modeshape was in positon of 4th floor

42 Acceleration Response The ratio of measured accelera-tion to the corresponding input peak ground accelerationis called the acceleration amplifying factor It reflects thedynamic response of structure under earthquake excitationsThe acceleration amplifying factors along the height of themodels are described in Figure 10(a) under the B-WSMDZCand ELW ground motions for different seismic levels Obvi-ously that of the two models was gradually increased alongthe height of themodel in each test phaseWith the increasingPGA arranged from 007 g to 062 g the whole developmenttrend of that decreased gradually implying the progressivedegradation of structural stiffness In each test phase themodels presented different dynamic responses under thedifferent earthquake excitations In elastic stage with thePGA of 007 g and 014 g the CIPS had greater response toB-WSM and ELW than to DZC The MPCS had uniformresponse to the three excitations The earthquake waves withdifferent frequency-spectrum characteristics would lead tothis phenomenon The greatest response of CIPS occurredunder the ELW with PGA of 022 g and that of MPCSoccurred in the B-WSM with PGA of 040 g The diversitiesof twomodels about the acceleration amplifying factors couldbe caused by the connections between the precast elementsThe factors of CIPS and MPCS reduced from 546 to 319and from 523 to 308 in the final test phase respectivelyAs shown in Figure 10(b) the distribution feature was more

regular under the test phases with PGA from 0035 g to 014 gWhen the two models were subjected to severe damage theinfluence of high-order vibration modes increased graduallyand the acceleration amplifying factors at somemeasurementpoints no longer conformed to the distribution [30]

43 Earthquake Action The seismic force distribution char-acteristic of structure is very important reference for theaseismic design and application of the MPCS and CIPS Themaximum seismic force of the 119894th floor is derived as follows

119865119894max = 1198981198941003816100381610038161003816 119894 (119905) + 0 (119905) max

1003816100381610038161003816 (1)

where 119865119894max is the maximum seismic force 119898119894 is the lumpedmass of the 119894th floor 119894(119905) is the acceleration response of the119894th floor relative to the ground at the time of 119905 and 0(119905) is theground acceleration at the time of 119905

According to (1) the maximum seismic forces of CIPSand MPCS are presented and compared in Figure 11 Asillustrated in Figure 11(a) the results of them were presentedin linear behavior following the increasing earthquake waveswith PGA from 0035 g to 014 g The curves were increasedprogressively along the height of models They could reflectthe actual seismic force distribution in some extent of struc-ture in the elastic stage and the high-order modes could beignored in this time And then the cracks expanded graduallyin different ways of the twomodels and the influence of high-order modes worked The models entered into plastic stagewith obvious nonlinearity The maximum seismic forces ofthe CIPS emerged in the middle of model such as 5th 6thand 8th Those of MPCS were in 4th 5th and 8th In theplastic stage with PGA from 022 g to 062 g the distributionof seismic forces changed for the high-order modes

The interstory shear forceswere calculated by (1) and theyare illustrated along the height of models in Figure 11(b) Theinterstory shear force was increased gradually with the PGAand decreased along the height of model structure In elasticstage the distribution of the interstory shear of CIPS andMPCS showed similar rules like inverted triangle In plasticstage the vibration trend was effected by the high-ordermodes and interstory shear force was not strictlymeeting thedistribution pattern

44 Displacement Response The maximum story displace-ments of the CIPS and MPCS along the model heightobtained from the B-WSM DZC and ELW with thePGA from 0035 g to 062 g are compared and depicted inFigure 12(a) Since the shear wall structure is the effectivelateral resisting system [1] themaximum story displacementsof the models were small in 12th story arranged from 113mmto 756mm of CIPS and from 106mm to 699mm of MPCSin the elastic stageThe stage contained the PGA from 0035 gto 014 g With increasing the intensity of earthquake wavesthe two models demonstrated different damage phenomenadescribed in the above paragraph The concentrated damagein shear wall of 4th story of CIPS caused bigger displacementthan that of MPCS The bigger displacement formed thebigger torsion angle of floor and then brought about themuchbigger displacement in upside structure of CIPS And the

10 Shock and Vibration

0123456789

101112

Floo

r

01 02 03 04 05 06 07 08 09 10 1100Mode coefficient of first mode shape

0123456789

101112

Floo

r

01 02 03 04 05 06 07 08 09 10 1100Mode coefficient of first mode shape

minus08 minus06 minus04 minus02minus10 02 04 06 08 1000Mode coefficient of second mode shape

minus06 minus04 minus02minus08minus10 02 04 06 08 1000Mode coefficient of second mode shape

0123456789

101112

Floo

r

0123456789

101112

Floo

r

MPCS-initialMPCS-0035 gMPCS-007 gMPCS-014 g

MPCS-022 gMPCS-040 gMPCS-062 g

MPCS-initialMPCS-0035 gMPCS-007 gMPCS-014 g

MPCS-022 gMPCS-040 gMPCS-062 g

CIPS-initialCIPS-0035 gCIPS-007 gCIPS-014 g

CIPS-022 gCIPS-040 gCIPS-062 g

CIPS-initialCIPS-0035 gCIPS-007 gCIPS-014 g

CIPS-022 gCIPS-040 gCIPS-062 g

Figure 9 Mode shapes of CIPS and MPCS

maximumdestructive story driftwas in 4th floorMeanwhilethe two models showed obviously nonlinear behavior withthe PGA of 040 g and 062 g The shape of maximum storydisplacements of them was similar to the first mode shapeThe maximum story displacements were 4867mm in CIPSand 4198mm in MPCS

The maximum interstory drift ratios were calculated andpresented in Figure 12(b) The maximum value of story driftof CIPS was 11005 in 5th story under the DZC earthquakewave with the PGA of 070 g and that of MPCS was 11020in 5th story under the same wave They met the provisionabout the maximum story drift in elastic stage in area of 8-degree seismic fortification intensity in code of GB 50011-2010 [21] In the stage from PGA of 0035 g to 014 g thestory drift increased progressively The seismic design codeasks for the limiting value of plastic interstory drift ratio toforbid collapse We checked the story drift under the PGA of040 g The values of CIPS and MPCS were 1121 and 1127

which met the limiting value in rare earthquake action of 8-degree seismic regions Under the excitation of PGAof 062 gthe values exceeded the limiting value in rare earthquakeaction of 9-degree seismic regions And then the moreintensive earthquakewaves with PGAof 080 gwere inputtedthe models presented better seismic behavior withoutcollapse

5 Conclusion

A comparative seismic study between CIPS and MPCS wasconducted by shaking table test which contained a 15 scaled12-story CIPS model and MPCS model The experimentalresults of dynamic characteristic failure pattern and mecha-nism and seismic response of the models were discussed andcompared with each other to better understand their seismicbehavior Based on the intensive analysis of test results thefollowing conclusions are derived

Shock and Vibration 11

PGA of 062 gPGA of 040 gPGA of 022 gPGA of 014 gPGA of 007 g PGA of 062 gPGA of 040 gPGA of 022 gPGA of 014 gPGA of 007 g

EE E EB B B BDD DD

E(LW

)

D(Z

C)

B(-W

SM)

Peak ground acceleration of CIPS (g)

E EEEB B BB DDD D

E(LW

)

D(Z

C)

B(-W

SM)

Peak ground acceleration of MPCS (g)

12th floor11th floor10th floor9th floor8th floor

7th floor6th floor5th floor4th floor3rd floor

2nd floor1st floorFloor

12th floor11th floor10th floor9th floor8th floor

7th floor6th floor5th floor4th floor3rd floor

2nd floor1st floorFloor

081216202428323640444852566064

Acce

lera

tion

ampl

ifyin

g co

effici

ent

081216202428323640444852566064

Acce

lera

tion

ampl

ifyin

g co

effici

ent

(a) Acceleration amplifying factors changed along the test phases of CIPS and MPCS

0123456789

101112

Floo

r

15 20 25 30 35 40 45 50 5510Acceleration amplifying coefficient of B-WSM

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 g

MPCS-040 gMPCS-062 gCIPS-0035 gCIPS-007 g

CIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Acceleration amplifying factors of B-WSMwave along the height oftwo models

Figure 10 Acceleration amplifying factors of CIPS and MPCS

(1) The typical failure pattern of the CIPS was damageconcentration in coupling beams in first and thenthe cracks emerged in the shear wall in plasticstage However besides the concentrated damage incoupling beam the connections between the precastelement and the CIP floor were the weakness and

the vertical crack came after the horizontal crack inplastic stage The precast element with high qualitiesdid not present cracks in the test

(2) The initial natural frequency as the initial stiffnessof MPCS degraded more obviously than CIPS forthe fine cracks in the connections The damping

12 Shock and Vibration

1 2 3 4 5 6 7 8 9 10 11 120Seismic force of DZC (kN)

1 2 3 4 5 6 7 8 9 10 11 120Seismic force of B-WSM (kN)

1 2 3 4 5 6 7 8 9 10 11 12 130Seismic force of ELW (kN)

123456789

101112

Floo

r

123456789

101112

Floo

r

123456789

101112

Floo

r

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(a) Distribution of seismic forces

10 20 30 40 50 60 70 80 90 1000Story shear of ELW (kN)

0123456789

101112

Floo

r10 20 30 40 50 60 70 80 900

Story shear of B-WSM (kN)

0123456789

101112

Floo

r

0123456789

101112

Floo

r

10 20 30 40 50 60 70 800Story shear of DZC (kN)

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Distribution of story shear

Figure 11 Earthquake actions of CIPS and MPCS

ratio of them had similar parameter but the way ofenergy dissipation of them was different And theyhad similar mode shapes

(3) The acceleration amplifying factors of them increasedalong the height of models and decreased progres-sively with the increasing PGA Nevertheless theyhad different responses under the different earth-quake waves The maximum acceleration amplifyingfactor of CIPS was 546 under the ELW with PGA of022 g and that of MPCS was 523 under the B-WSMwith PGA of 040 g

(4) Distribution of seismic forces of them increasedlinearly along the height ofmodels and then presentednonlinearity for the influence of high-order modesThe distribution of the interstory shear of CIPS and

MPCS showed similar rules like inverted triangle inelastic stage

(5) The maximum story displacements of them werealmost equal in elastic stage However the maximumstory displacement of CIPS was greater than thatof MPCS in plastic stage which was caused bythe concentrated failure in 4th story of CIPS Themaximum interstory ratio under the seismic waveswith PGA of 007 g and 040 g was in accordancewith the provision of Chinese code Under the moreintensive earthquake wave of PGA of 080 g the twomodels had enough capacity to resist collapse

Conflicts of Interest

The authors declare that they have no conflicts of interest

Shock and Vibration 13

5 10 15 20 25 30 35 40 450Displacement of B-WSM (mm)

0123456789

101112

Floo

r

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of DZC (mm)

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of ELW (mm)

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(a) Maximum story displacements of CIPS and MPCS

0123456789

101112

Floo

r

000

1

001

0

000

8

000

4

001

2

000

0

000

7

000

5

000

9

000

3

001

1

000

2

000

6

Story drift of DZC

000

1

000

8

000

4

001

0

000

0

000

60

007

000

3

000

9

000

2

001

1

000

5

Story drift of B-WSM

0123456789

101112

Floo

r

0123456789

101112

Floo

r

000

1

001

2

001

0

000

4

000

8

001

3

000

0

000

6

000

9

000

5

001

1

000

30

002

000

7

Story drift of ELW

Elastic limit Plastic limitElastic limit Plastic limitElastic limit Plastic limit

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Story drift ratio of CIPS and MPCS

Figure 12 Displacement responses of CIPS and MPCS

Acknowledgments

This research has been supported by both Innovation TeamofXirsquoan University of Architecture and Technology and ShaanxiProvince Science and Technology Coordination InnovationProject (nos 2015KTZDSF03-05-01 2015KTZDSF03-04 and2014SZS04-Z01)The support of the National Natural ScienceFoundation of China (Grants nos 51408456 51578444 and51478381) and Ministry of Education Plan for Yangtze RiverScholar and Innovation TeamDevelopment (no IRT13089) isalso acknowledged

References

[1] R Park and T Paulay Reinforced Concrete Structures ohnWileyamp Sons 1975

[2] M Fintel ldquoPerformance of buildings with shear walls in earth-quakes of the last thirty yearsrdquo PCI journal vol 40 no 3 pp62ndash80 1995

[3] A A Yee ldquoSplice sleeve for reinforcing barsrdquo US Pat vol 35401968

[4] J H Ling A B Abd Rahman I S Ibrahim and Z AbdulHamid ldquoBehaviour of grouted pipe splice under incrementaltensile loadrdquo Construction and Building Materials vol 33 no8 pp 90ndash98 2012

[5] M T El-Sheikh R Sause S Pessiki and L-W Lu ldquoSeismicbehavior and design of unbonded post-tensioned precast con-crete framesrdquo PCI journal vol 44 no 3 pp 54ndash71 1999

[6] R S Henry S Sritharan and J M Ingham ldquoUnbonded tendonstresses in post-tensioned concrete walls at nominal flexuralstrengthrdquo ACI Structural Journal vol 109 no 2 pp 171ndash1822012

14 Shock and Vibration

[7] P Seifi R S Henry and JM Ingham ldquoPanel connection detailsin existing New Zealand precast concrete buildingsrdquo Bulletin ofthe New Zealand Society for Earthquake Engineering vol 49 no2 pp 190ndash199 2016

[8] K A Soudki S H Rizkalla and R W Daikiw ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformations part 2 prestressed connectionsrdquo PCI journal vol40 no 5 pp 82ndash96 1995

[9] K A Soudki S H Rizkalla and B LeBlanc ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformationsmdashpart 1 mild steel connectionsrdquo PCI journal vol40 no 4 pp 78ndash96 1995

[10] K A Soudki J S West S H Rizkalla and B BlackettldquoHorizontal connections for precast concrete shear wall panelsunder cyclic shear loadingrdquo PCI journal vol 41 no 3 pp 64ndash80 1996

[11] R Vaghei F Hejazi H Taheri M S Jaafar and A A Ali ldquoEval-uate performance of precast concrete wall to wall connectionrdquoAPCBEE Procedia vol 9 pp 285ndash290 2014

[12] H Jiang Z Chen J Zhang B Wu Y Tian andW Liu ldquoQuasi-static test of precast reinforced concrete shear wall structurerdquoJianzhu Jiegou XuebaoJournal of Building Structures vol 32 no6 pp 34ndash40 2011

[13] Z ChenH Jiang J Zhang BWu Y Tian andW Liu ldquoPseudo-dynamic substructure test on precast reinforced concrete shearwall structurerdquo Jianzhu Jiegou XuebaoJournal of Building Struc-tures vol 32 no 6 pp 41ndash50 2011

[14] J R QianW L Han Z Z Zhao and H Qin ldquoPseudo-dynamicSubstructure Test on a 3-story Full-scaleModel of PrefabricatedConcrete Shear Wall Structure with Rebars Splicing by GroutSleevesrdquo Journal of Building Structures vol 38 no 3 pp 26ndash382017

[15] P Negro D A Bournas and F J Molina ldquoPseudodynamictests on a full-scale 3-storey precast concrete building Globalresponserdquo Engineering Structures vol 57 pp 594ndash608 2013

[16] M Oliva P Gavrilovic and R W Clough ldquoSeismic testing oflarge panel precast walls Comparison of pseudostatic and shak-ing table testsrdquo Earthquake Engineering amp Structural Dynamicsvol 19 no 6 pp 859ndash875 1990

[17] L H Lee W H Yi and S Y Seo ldquoShaking Table Test of PrecastConcrete Wall Structurerdquo in Proceedings of the Eleventh WorldConference on Earthquake Engineering

[18] W Wang A Q Li et al ldquoShaking table test of precast concreteshear wall isolation structurerdquo Journal of Building Structuresvol 36 no 6 pp 35ndash43 2015

[19] EERI Preliminary Reconnaissance Report Earthquake Engi-neering Research Institute 1994

[20] EERI Armenia Earthquake Reconnaissance Report EarthquakeEngineering Research Institute 1989

[21] GB 50011-2010 Code for Seismic Design of Buildings ChinaArchitecture and Building Press Beijing China 2010

[22] JGJ 1-2014 Technical Specification for Precast Concrete Struc-tures China Architecture amp Building Press Beijing China2014

[23] 15J939-1 Example of Architectural Design for PrefabricatedConcrete Structures China Planning Press Beijing China 2015

[24] E Buckingham ldquoOn physically similar systems Illustrations ofthe use of dimensional equationsrdquo Physical Review A AtomicMolecular and Optical Physics vol 4 no 4 pp 345ndash376 1914

[25] Y Zhou andX L LuTheexperimentalmethod and technology ofshaking table test about building structure Science Press BeijingChina 2012

[26] JGJT 101-2015 Specification for Seismic Test of Buildings ChinaArchitecture amp Building Press Beijing China 2014

[27] C Y Shen F L Zhou et al ldquoExperimental research on micro-concrete used in dynamic test modelrdquo Journal of GuangzhouUniversity (Natural Science Edition) vol 4 no 3 pp 250ndash2532005

[28] J Xiao T L Pham and T Ding ldquoShake table test on seismicresponse of a precast frame with recycled aggregate concreterdquoAdvances in Structural Engineering vol 18 no 9 pp 1517ndash15342015

[29] S C Chakrabarti G C Nayak and D K Paul ldquoShear charac-teristics of cast-in-place vertical joints in story-high precast wallassemblyrdquo ACI Structural Journal vol 85 no 1 pp 30ndash45 1988

[30] CWang and J Xiao ldquoStudy of the seismic response of a recycledaggregate concrete frame structurerdquo Earthquake Engineeringand Engineering Vibration vol 12 no 4 pp 669ndash680 2013

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 8: Comparative Study of Seismic Behavior between Monolithic

8 Shock and Vibration

D11DDDD11D11D11D11DDDD 11D

A1 D1

A2 V1

A3 D2

A4 V2

A5 D3

A6 V3

A7 D4

A8 V4

A9 D5

A10 V5

A11 D6

A12 D7

A13 D8 V6

A14

A15

A16

A17

A18

A19

A20

A21

A22

A23

A24

A25

A26

V7

V9

D9

V11

V8

V10

D10

V12

D12

Y

X

Figure 4 Arrangement of accelerometers displacement transduc-ers and velocity vibration sensors

CIPS MPCS

Figure 5 General view of CIPS and MPCS

the adjunction grouted sleeves and the additional stirrupsin the connections deduced the phenomenon As the firstearthquake excitation was inputted the natural frequency ofMPCS decreased about 20 which could be caused by theshrinkage and microcracks in connections as initial damage[29] With the increasing energy the influence of the initialdamage was not the main factor And then the two modelswith approximately equal frequency presented in failure stateThe curves of stiffness degradation of CIPS reduced graduallywith the increasing PGA In contrast withCIPS that ofMPCS

CIPS MPCS

Figure 6 Failure patterns of CIPS and MPCS

00

01

02

03

04

05

06

07

08

09

10

11

Stiff

ness

ratio

MPCS-XMPCS-Y

CIPS-XCIPS-Y

022 g014 g007 g 040 g0035 g 062 gPrimePGA (g)

Figure 7 Stiffness ratio versus frequency in119883 and 119884 directions

decreased evidently in the first phase and then diminishedslowly from PGA of 0035 g to 014 g Finally they hadsimilar residual stiffnesses The diverse tendency might beunderstood by the mentioned failure pattern

Damping ratio reflects the dissipation capacity of struc-ture As is shown in Figure 8 the damping ratio increasedgradually following the PGA In the first phase it was 42And then the damping ratio increased slowly before the PGAof 040 g which arranged from 42 to 50 The averagedamping ratio of each phase changed from 42 to 82in the loading process which was in the scope of cast-in-place concrete structure Nevertheless the energy dissipationmechanismofCIPS andMPCSwas different for the order anddistribution of cracks

The first and second mode shapes of the models aredescribed in Figure 9 In general the first-order mode shape

Shock and Vibration 9

000

001

002

003

004

005

006

007

008

009

010

011

Dam

ping

ratio

MPCS-XMPCS-Y

CIPS-XCIPS-Y

014 g 022 g007 g0035 g 040 g 062 gPrimePGA (g)

Figure 8 Damping ratios of CIPS and MPCS

exhibited bending deformation characteristics and the lateralstiffness was uniform distribution along the height of themodel The shapes of them were bending to 119909-axis littleby little The phenomenon would be caused by high-ordermodes And the trend of that of CIPSwasmore obvious in thePGAof 040 g and 062 g since the severe damage occurred inthe shear wall The second-order mode shapes of them weresimilar And the maximummode coefficient of second modeshape was in positon of 4th floor

42 Acceleration Response The ratio of measured accelera-tion to the corresponding input peak ground accelerationis called the acceleration amplifying factor It reflects thedynamic response of structure under earthquake excitationsThe acceleration amplifying factors along the height of themodels are described in Figure 10(a) under the B-WSMDZCand ELW ground motions for different seismic levels Obvi-ously that of the two models was gradually increased alongthe height of themodel in each test phaseWith the increasingPGA arranged from 007 g to 062 g the whole developmenttrend of that decreased gradually implying the progressivedegradation of structural stiffness In each test phase themodels presented different dynamic responses under thedifferent earthquake excitations In elastic stage with thePGA of 007 g and 014 g the CIPS had greater response toB-WSM and ELW than to DZC The MPCS had uniformresponse to the three excitations The earthquake waves withdifferent frequency-spectrum characteristics would lead tothis phenomenon The greatest response of CIPS occurredunder the ELW with PGA of 022 g and that of MPCSoccurred in the B-WSM with PGA of 040 g The diversitiesof twomodels about the acceleration amplifying factors couldbe caused by the connections between the precast elementsThe factors of CIPS and MPCS reduced from 546 to 319and from 523 to 308 in the final test phase respectivelyAs shown in Figure 10(b) the distribution feature was more

regular under the test phases with PGA from 0035 g to 014 gWhen the two models were subjected to severe damage theinfluence of high-order vibration modes increased graduallyand the acceleration amplifying factors at somemeasurementpoints no longer conformed to the distribution [30]

43 Earthquake Action The seismic force distribution char-acteristic of structure is very important reference for theaseismic design and application of the MPCS and CIPS Themaximum seismic force of the 119894th floor is derived as follows

119865119894max = 1198981198941003816100381610038161003816 119894 (119905) + 0 (119905) max

1003816100381610038161003816 (1)

where 119865119894max is the maximum seismic force 119898119894 is the lumpedmass of the 119894th floor 119894(119905) is the acceleration response of the119894th floor relative to the ground at the time of 119905 and 0(119905) is theground acceleration at the time of 119905

According to (1) the maximum seismic forces of CIPSand MPCS are presented and compared in Figure 11 Asillustrated in Figure 11(a) the results of them were presentedin linear behavior following the increasing earthquake waveswith PGA from 0035 g to 014 g The curves were increasedprogressively along the height of models They could reflectthe actual seismic force distribution in some extent of struc-ture in the elastic stage and the high-order modes could beignored in this time And then the cracks expanded graduallyin different ways of the twomodels and the influence of high-order modes worked The models entered into plastic stagewith obvious nonlinearity The maximum seismic forces ofthe CIPS emerged in the middle of model such as 5th 6thand 8th Those of MPCS were in 4th 5th and 8th In theplastic stage with PGA from 022 g to 062 g the distributionof seismic forces changed for the high-order modes

The interstory shear forceswere calculated by (1) and theyare illustrated along the height of models in Figure 11(b) Theinterstory shear force was increased gradually with the PGAand decreased along the height of model structure In elasticstage the distribution of the interstory shear of CIPS andMPCS showed similar rules like inverted triangle In plasticstage the vibration trend was effected by the high-ordermodes and interstory shear force was not strictlymeeting thedistribution pattern

44 Displacement Response The maximum story displace-ments of the CIPS and MPCS along the model heightobtained from the B-WSM DZC and ELW with thePGA from 0035 g to 062 g are compared and depicted inFigure 12(a) Since the shear wall structure is the effectivelateral resisting system [1] themaximum story displacementsof the models were small in 12th story arranged from 113mmto 756mm of CIPS and from 106mm to 699mm of MPCSin the elastic stageThe stage contained the PGA from 0035 gto 014 g With increasing the intensity of earthquake wavesthe two models demonstrated different damage phenomenadescribed in the above paragraph The concentrated damagein shear wall of 4th story of CIPS caused bigger displacementthan that of MPCS The bigger displacement formed thebigger torsion angle of floor and then brought about themuchbigger displacement in upside structure of CIPS And the

10 Shock and Vibration

0123456789

101112

Floo

r

01 02 03 04 05 06 07 08 09 10 1100Mode coefficient of first mode shape

0123456789

101112

Floo

r

01 02 03 04 05 06 07 08 09 10 1100Mode coefficient of first mode shape

minus08 minus06 minus04 minus02minus10 02 04 06 08 1000Mode coefficient of second mode shape

minus06 minus04 minus02minus08minus10 02 04 06 08 1000Mode coefficient of second mode shape

0123456789

101112

Floo

r

0123456789

101112

Floo

r

MPCS-initialMPCS-0035 gMPCS-007 gMPCS-014 g

MPCS-022 gMPCS-040 gMPCS-062 g

MPCS-initialMPCS-0035 gMPCS-007 gMPCS-014 g

MPCS-022 gMPCS-040 gMPCS-062 g

CIPS-initialCIPS-0035 gCIPS-007 gCIPS-014 g

CIPS-022 gCIPS-040 gCIPS-062 g

CIPS-initialCIPS-0035 gCIPS-007 gCIPS-014 g

CIPS-022 gCIPS-040 gCIPS-062 g

Figure 9 Mode shapes of CIPS and MPCS

maximumdestructive story driftwas in 4th floorMeanwhilethe two models showed obviously nonlinear behavior withthe PGA of 040 g and 062 g The shape of maximum storydisplacements of them was similar to the first mode shapeThe maximum story displacements were 4867mm in CIPSand 4198mm in MPCS

The maximum interstory drift ratios were calculated andpresented in Figure 12(b) The maximum value of story driftof CIPS was 11005 in 5th story under the DZC earthquakewave with the PGA of 070 g and that of MPCS was 11020in 5th story under the same wave They met the provisionabout the maximum story drift in elastic stage in area of 8-degree seismic fortification intensity in code of GB 50011-2010 [21] In the stage from PGA of 0035 g to 014 g thestory drift increased progressively The seismic design codeasks for the limiting value of plastic interstory drift ratio toforbid collapse We checked the story drift under the PGA of040 g The values of CIPS and MPCS were 1121 and 1127

which met the limiting value in rare earthquake action of 8-degree seismic regions Under the excitation of PGAof 062 gthe values exceeded the limiting value in rare earthquakeaction of 9-degree seismic regions And then the moreintensive earthquakewaves with PGAof 080 gwere inputtedthe models presented better seismic behavior withoutcollapse

5 Conclusion

A comparative seismic study between CIPS and MPCS wasconducted by shaking table test which contained a 15 scaled12-story CIPS model and MPCS model The experimentalresults of dynamic characteristic failure pattern and mecha-nism and seismic response of the models were discussed andcompared with each other to better understand their seismicbehavior Based on the intensive analysis of test results thefollowing conclusions are derived

Shock and Vibration 11

PGA of 062 gPGA of 040 gPGA of 022 gPGA of 014 gPGA of 007 g PGA of 062 gPGA of 040 gPGA of 022 gPGA of 014 gPGA of 007 g

EE E EB B B BDD DD

E(LW

)

D(Z

C)

B(-W

SM)

Peak ground acceleration of CIPS (g)

E EEEB B BB DDD D

E(LW

)

D(Z

C)

B(-W

SM)

Peak ground acceleration of MPCS (g)

12th floor11th floor10th floor9th floor8th floor

7th floor6th floor5th floor4th floor3rd floor

2nd floor1st floorFloor

12th floor11th floor10th floor9th floor8th floor

7th floor6th floor5th floor4th floor3rd floor

2nd floor1st floorFloor

081216202428323640444852566064

Acce

lera

tion

ampl

ifyin

g co

effici

ent

081216202428323640444852566064

Acce

lera

tion

ampl

ifyin

g co

effici

ent

(a) Acceleration amplifying factors changed along the test phases of CIPS and MPCS

0123456789

101112

Floo

r

15 20 25 30 35 40 45 50 5510Acceleration amplifying coefficient of B-WSM

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 g

MPCS-040 gMPCS-062 gCIPS-0035 gCIPS-007 g

CIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Acceleration amplifying factors of B-WSMwave along the height oftwo models

Figure 10 Acceleration amplifying factors of CIPS and MPCS

(1) The typical failure pattern of the CIPS was damageconcentration in coupling beams in first and thenthe cracks emerged in the shear wall in plasticstage However besides the concentrated damage incoupling beam the connections between the precastelement and the CIP floor were the weakness and

the vertical crack came after the horizontal crack inplastic stage The precast element with high qualitiesdid not present cracks in the test

(2) The initial natural frequency as the initial stiffnessof MPCS degraded more obviously than CIPS forthe fine cracks in the connections The damping

12 Shock and Vibration

1 2 3 4 5 6 7 8 9 10 11 120Seismic force of DZC (kN)

1 2 3 4 5 6 7 8 9 10 11 120Seismic force of B-WSM (kN)

1 2 3 4 5 6 7 8 9 10 11 12 130Seismic force of ELW (kN)

123456789

101112

Floo

r

123456789

101112

Floo

r

123456789

101112

Floo

r

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(a) Distribution of seismic forces

10 20 30 40 50 60 70 80 90 1000Story shear of ELW (kN)

0123456789

101112

Floo

r10 20 30 40 50 60 70 80 900

Story shear of B-WSM (kN)

0123456789

101112

Floo

r

0123456789

101112

Floo

r

10 20 30 40 50 60 70 800Story shear of DZC (kN)

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Distribution of story shear

Figure 11 Earthquake actions of CIPS and MPCS

ratio of them had similar parameter but the way ofenergy dissipation of them was different And theyhad similar mode shapes

(3) The acceleration amplifying factors of them increasedalong the height of models and decreased progres-sively with the increasing PGA Nevertheless theyhad different responses under the different earth-quake waves The maximum acceleration amplifyingfactor of CIPS was 546 under the ELW with PGA of022 g and that of MPCS was 523 under the B-WSMwith PGA of 040 g

(4) Distribution of seismic forces of them increasedlinearly along the height ofmodels and then presentednonlinearity for the influence of high-order modesThe distribution of the interstory shear of CIPS and

MPCS showed similar rules like inverted triangle inelastic stage

(5) The maximum story displacements of them werealmost equal in elastic stage However the maximumstory displacement of CIPS was greater than thatof MPCS in plastic stage which was caused bythe concentrated failure in 4th story of CIPS Themaximum interstory ratio under the seismic waveswith PGA of 007 g and 040 g was in accordancewith the provision of Chinese code Under the moreintensive earthquake wave of PGA of 080 g the twomodels had enough capacity to resist collapse

Conflicts of Interest

The authors declare that they have no conflicts of interest

Shock and Vibration 13

5 10 15 20 25 30 35 40 450Displacement of B-WSM (mm)

0123456789

101112

Floo

r

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of DZC (mm)

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of ELW (mm)

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(a) Maximum story displacements of CIPS and MPCS

0123456789

101112

Floo

r

000

1

001

0

000

8

000

4

001

2

000

0

000

7

000

5

000

9

000

3

001

1

000

2

000

6

Story drift of DZC

000

1

000

8

000

4

001

0

000

0

000

60

007

000

3

000

9

000

2

001

1

000

5

Story drift of B-WSM

0123456789

101112

Floo

r

0123456789

101112

Floo

r

000

1

001

2

001

0

000

4

000

8

001

3

000

0

000

6

000

9

000

5

001

1

000

30

002

000

7

Story drift of ELW

Elastic limit Plastic limitElastic limit Plastic limitElastic limit Plastic limit

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Story drift ratio of CIPS and MPCS

Figure 12 Displacement responses of CIPS and MPCS

Acknowledgments

This research has been supported by both Innovation TeamofXirsquoan University of Architecture and Technology and ShaanxiProvince Science and Technology Coordination InnovationProject (nos 2015KTZDSF03-05-01 2015KTZDSF03-04 and2014SZS04-Z01)The support of the National Natural ScienceFoundation of China (Grants nos 51408456 51578444 and51478381) and Ministry of Education Plan for Yangtze RiverScholar and Innovation TeamDevelopment (no IRT13089) isalso acknowledged

References

[1] R Park and T Paulay Reinforced Concrete Structures ohnWileyamp Sons 1975

[2] M Fintel ldquoPerformance of buildings with shear walls in earth-quakes of the last thirty yearsrdquo PCI journal vol 40 no 3 pp62ndash80 1995

[3] A A Yee ldquoSplice sleeve for reinforcing barsrdquo US Pat vol 35401968

[4] J H Ling A B Abd Rahman I S Ibrahim and Z AbdulHamid ldquoBehaviour of grouted pipe splice under incrementaltensile loadrdquo Construction and Building Materials vol 33 no8 pp 90ndash98 2012

[5] M T El-Sheikh R Sause S Pessiki and L-W Lu ldquoSeismicbehavior and design of unbonded post-tensioned precast con-crete framesrdquo PCI journal vol 44 no 3 pp 54ndash71 1999

[6] R S Henry S Sritharan and J M Ingham ldquoUnbonded tendonstresses in post-tensioned concrete walls at nominal flexuralstrengthrdquo ACI Structural Journal vol 109 no 2 pp 171ndash1822012

14 Shock and Vibration

[7] P Seifi R S Henry and JM Ingham ldquoPanel connection detailsin existing New Zealand precast concrete buildingsrdquo Bulletin ofthe New Zealand Society for Earthquake Engineering vol 49 no2 pp 190ndash199 2016

[8] K A Soudki S H Rizkalla and R W Daikiw ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformations part 2 prestressed connectionsrdquo PCI journal vol40 no 5 pp 82ndash96 1995

[9] K A Soudki S H Rizkalla and B LeBlanc ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformationsmdashpart 1 mild steel connectionsrdquo PCI journal vol40 no 4 pp 78ndash96 1995

[10] K A Soudki J S West S H Rizkalla and B BlackettldquoHorizontal connections for precast concrete shear wall panelsunder cyclic shear loadingrdquo PCI journal vol 41 no 3 pp 64ndash80 1996

[11] R Vaghei F Hejazi H Taheri M S Jaafar and A A Ali ldquoEval-uate performance of precast concrete wall to wall connectionrdquoAPCBEE Procedia vol 9 pp 285ndash290 2014

[12] H Jiang Z Chen J Zhang B Wu Y Tian andW Liu ldquoQuasi-static test of precast reinforced concrete shear wall structurerdquoJianzhu Jiegou XuebaoJournal of Building Structures vol 32 no6 pp 34ndash40 2011

[13] Z ChenH Jiang J Zhang BWu Y Tian andW Liu ldquoPseudo-dynamic substructure test on precast reinforced concrete shearwall structurerdquo Jianzhu Jiegou XuebaoJournal of Building Struc-tures vol 32 no 6 pp 41ndash50 2011

[14] J R QianW L Han Z Z Zhao and H Qin ldquoPseudo-dynamicSubstructure Test on a 3-story Full-scaleModel of PrefabricatedConcrete Shear Wall Structure with Rebars Splicing by GroutSleevesrdquo Journal of Building Structures vol 38 no 3 pp 26ndash382017

[15] P Negro D A Bournas and F J Molina ldquoPseudodynamictests on a full-scale 3-storey precast concrete building Globalresponserdquo Engineering Structures vol 57 pp 594ndash608 2013

[16] M Oliva P Gavrilovic and R W Clough ldquoSeismic testing oflarge panel precast walls Comparison of pseudostatic and shak-ing table testsrdquo Earthquake Engineering amp Structural Dynamicsvol 19 no 6 pp 859ndash875 1990

[17] L H Lee W H Yi and S Y Seo ldquoShaking Table Test of PrecastConcrete Wall Structurerdquo in Proceedings of the Eleventh WorldConference on Earthquake Engineering

[18] W Wang A Q Li et al ldquoShaking table test of precast concreteshear wall isolation structurerdquo Journal of Building Structuresvol 36 no 6 pp 35ndash43 2015

[19] EERI Preliminary Reconnaissance Report Earthquake Engi-neering Research Institute 1994

[20] EERI Armenia Earthquake Reconnaissance Report EarthquakeEngineering Research Institute 1989

[21] GB 50011-2010 Code for Seismic Design of Buildings ChinaArchitecture and Building Press Beijing China 2010

[22] JGJ 1-2014 Technical Specification for Precast Concrete Struc-tures China Architecture amp Building Press Beijing China2014

[23] 15J939-1 Example of Architectural Design for PrefabricatedConcrete Structures China Planning Press Beijing China 2015

[24] E Buckingham ldquoOn physically similar systems Illustrations ofthe use of dimensional equationsrdquo Physical Review A AtomicMolecular and Optical Physics vol 4 no 4 pp 345ndash376 1914

[25] Y Zhou andX L LuTheexperimentalmethod and technology ofshaking table test about building structure Science Press BeijingChina 2012

[26] JGJT 101-2015 Specification for Seismic Test of Buildings ChinaArchitecture amp Building Press Beijing China 2014

[27] C Y Shen F L Zhou et al ldquoExperimental research on micro-concrete used in dynamic test modelrdquo Journal of GuangzhouUniversity (Natural Science Edition) vol 4 no 3 pp 250ndash2532005

[28] J Xiao T L Pham and T Ding ldquoShake table test on seismicresponse of a precast frame with recycled aggregate concreterdquoAdvances in Structural Engineering vol 18 no 9 pp 1517ndash15342015

[29] S C Chakrabarti G C Nayak and D K Paul ldquoShear charac-teristics of cast-in-place vertical joints in story-high precast wallassemblyrdquo ACI Structural Journal vol 85 no 1 pp 30ndash45 1988

[30] CWang and J Xiao ldquoStudy of the seismic response of a recycledaggregate concrete frame structurerdquo Earthquake Engineeringand Engineering Vibration vol 12 no 4 pp 669ndash680 2013

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 9: Comparative Study of Seismic Behavior between Monolithic

Shock and Vibration 9

000

001

002

003

004

005

006

007

008

009

010

011

Dam

ping

ratio

MPCS-XMPCS-Y

CIPS-XCIPS-Y

014 g 022 g007 g0035 g 040 g 062 gPrimePGA (g)

Figure 8 Damping ratios of CIPS and MPCS

exhibited bending deformation characteristics and the lateralstiffness was uniform distribution along the height of themodel The shapes of them were bending to 119909-axis littleby little The phenomenon would be caused by high-ordermodes And the trend of that of CIPSwasmore obvious in thePGAof 040 g and 062 g since the severe damage occurred inthe shear wall The second-order mode shapes of them weresimilar And the maximummode coefficient of second modeshape was in positon of 4th floor

42 Acceleration Response The ratio of measured accelera-tion to the corresponding input peak ground accelerationis called the acceleration amplifying factor It reflects thedynamic response of structure under earthquake excitationsThe acceleration amplifying factors along the height of themodels are described in Figure 10(a) under the B-WSMDZCand ELW ground motions for different seismic levels Obvi-ously that of the two models was gradually increased alongthe height of themodel in each test phaseWith the increasingPGA arranged from 007 g to 062 g the whole developmenttrend of that decreased gradually implying the progressivedegradation of structural stiffness In each test phase themodels presented different dynamic responses under thedifferent earthquake excitations In elastic stage with thePGA of 007 g and 014 g the CIPS had greater response toB-WSM and ELW than to DZC The MPCS had uniformresponse to the three excitations The earthquake waves withdifferent frequency-spectrum characteristics would lead tothis phenomenon The greatest response of CIPS occurredunder the ELW with PGA of 022 g and that of MPCSoccurred in the B-WSM with PGA of 040 g The diversitiesof twomodels about the acceleration amplifying factors couldbe caused by the connections between the precast elementsThe factors of CIPS and MPCS reduced from 546 to 319and from 523 to 308 in the final test phase respectivelyAs shown in Figure 10(b) the distribution feature was more

regular under the test phases with PGA from 0035 g to 014 gWhen the two models were subjected to severe damage theinfluence of high-order vibration modes increased graduallyand the acceleration amplifying factors at somemeasurementpoints no longer conformed to the distribution [30]

43 Earthquake Action The seismic force distribution char-acteristic of structure is very important reference for theaseismic design and application of the MPCS and CIPS Themaximum seismic force of the 119894th floor is derived as follows

119865119894max = 1198981198941003816100381610038161003816 119894 (119905) + 0 (119905) max

1003816100381610038161003816 (1)

where 119865119894max is the maximum seismic force 119898119894 is the lumpedmass of the 119894th floor 119894(119905) is the acceleration response of the119894th floor relative to the ground at the time of 119905 and 0(119905) is theground acceleration at the time of 119905

According to (1) the maximum seismic forces of CIPSand MPCS are presented and compared in Figure 11 Asillustrated in Figure 11(a) the results of them were presentedin linear behavior following the increasing earthquake waveswith PGA from 0035 g to 014 g The curves were increasedprogressively along the height of models They could reflectthe actual seismic force distribution in some extent of struc-ture in the elastic stage and the high-order modes could beignored in this time And then the cracks expanded graduallyin different ways of the twomodels and the influence of high-order modes worked The models entered into plastic stagewith obvious nonlinearity The maximum seismic forces ofthe CIPS emerged in the middle of model such as 5th 6thand 8th Those of MPCS were in 4th 5th and 8th In theplastic stage with PGA from 022 g to 062 g the distributionof seismic forces changed for the high-order modes

The interstory shear forceswere calculated by (1) and theyare illustrated along the height of models in Figure 11(b) Theinterstory shear force was increased gradually with the PGAand decreased along the height of model structure In elasticstage the distribution of the interstory shear of CIPS andMPCS showed similar rules like inverted triangle In plasticstage the vibration trend was effected by the high-ordermodes and interstory shear force was not strictlymeeting thedistribution pattern

44 Displacement Response The maximum story displace-ments of the CIPS and MPCS along the model heightobtained from the B-WSM DZC and ELW with thePGA from 0035 g to 062 g are compared and depicted inFigure 12(a) Since the shear wall structure is the effectivelateral resisting system [1] themaximum story displacementsof the models were small in 12th story arranged from 113mmto 756mm of CIPS and from 106mm to 699mm of MPCSin the elastic stageThe stage contained the PGA from 0035 gto 014 g With increasing the intensity of earthquake wavesthe two models demonstrated different damage phenomenadescribed in the above paragraph The concentrated damagein shear wall of 4th story of CIPS caused bigger displacementthan that of MPCS The bigger displacement formed thebigger torsion angle of floor and then brought about themuchbigger displacement in upside structure of CIPS And the

10 Shock and Vibration

0123456789

101112

Floo

r

01 02 03 04 05 06 07 08 09 10 1100Mode coefficient of first mode shape

0123456789

101112

Floo

r

01 02 03 04 05 06 07 08 09 10 1100Mode coefficient of first mode shape

minus08 minus06 minus04 minus02minus10 02 04 06 08 1000Mode coefficient of second mode shape

minus06 minus04 minus02minus08minus10 02 04 06 08 1000Mode coefficient of second mode shape

0123456789

101112

Floo

r

0123456789

101112

Floo

r

MPCS-initialMPCS-0035 gMPCS-007 gMPCS-014 g

MPCS-022 gMPCS-040 gMPCS-062 g

MPCS-initialMPCS-0035 gMPCS-007 gMPCS-014 g

MPCS-022 gMPCS-040 gMPCS-062 g

CIPS-initialCIPS-0035 gCIPS-007 gCIPS-014 g

CIPS-022 gCIPS-040 gCIPS-062 g

CIPS-initialCIPS-0035 gCIPS-007 gCIPS-014 g

CIPS-022 gCIPS-040 gCIPS-062 g

Figure 9 Mode shapes of CIPS and MPCS

maximumdestructive story driftwas in 4th floorMeanwhilethe two models showed obviously nonlinear behavior withthe PGA of 040 g and 062 g The shape of maximum storydisplacements of them was similar to the first mode shapeThe maximum story displacements were 4867mm in CIPSand 4198mm in MPCS

The maximum interstory drift ratios were calculated andpresented in Figure 12(b) The maximum value of story driftof CIPS was 11005 in 5th story under the DZC earthquakewave with the PGA of 070 g and that of MPCS was 11020in 5th story under the same wave They met the provisionabout the maximum story drift in elastic stage in area of 8-degree seismic fortification intensity in code of GB 50011-2010 [21] In the stage from PGA of 0035 g to 014 g thestory drift increased progressively The seismic design codeasks for the limiting value of plastic interstory drift ratio toforbid collapse We checked the story drift under the PGA of040 g The values of CIPS and MPCS were 1121 and 1127

which met the limiting value in rare earthquake action of 8-degree seismic regions Under the excitation of PGAof 062 gthe values exceeded the limiting value in rare earthquakeaction of 9-degree seismic regions And then the moreintensive earthquakewaves with PGAof 080 gwere inputtedthe models presented better seismic behavior withoutcollapse

5 Conclusion

A comparative seismic study between CIPS and MPCS wasconducted by shaking table test which contained a 15 scaled12-story CIPS model and MPCS model The experimentalresults of dynamic characteristic failure pattern and mecha-nism and seismic response of the models were discussed andcompared with each other to better understand their seismicbehavior Based on the intensive analysis of test results thefollowing conclusions are derived

Shock and Vibration 11

PGA of 062 gPGA of 040 gPGA of 022 gPGA of 014 gPGA of 007 g PGA of 062 gPGA of 040 gPGA of 022 gPGA of 014 gPGA of 007 g

EE E EB B B BDD DD

E(LW

)

D(Z

C)

B(-W

SM)

Peak ground acceleration of CIPS (g)

E EEEB B BB DDD D

E(LW

)

D(Z

C)

B(-W

SM)

Peak ground acceleration of MPCS (g)

12th floor11th floor10th floor9th floor8th floor

7th floor6th floor5th floor4th floor3rd floor

2nd floor1st floorFloor

12th floor11th floor10th floor9th floor8th floor

7th floor6th floor5th floor4th floor3rd floor

2nd floor1st floorFloor

081216202428323640444852566064

Acce

lera

tion

ampl

ifyin

g co

effici

ent

081216202428323640444852566064

Acce

lera

tion

ampl

ifyin

g co

effici

ent

(a) Acceleration amplifying factors changed along the test phases of CIPS and MPCS

0123456789

101112

Floo

r

15 20 25 30 35 40 45 50 5510Acceleration amplifying coefficient of B-WSM

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 g

MPCS-040 gMPCS-062 gCIPS-0035 gCIPS-007 g

CIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Acceleration amplifying factors of B-WSMwave along the height oftwo models

Figure 10 Acceleration amplifying factors of CIPS and MPCS

(1) The typical failure pattern of the CIPS was damageconcentration in coupling beams in first and thenthe cracks emerged in the shear wall in plasticstage However besides the concentrated damage incoupling beam the connections between the precastelement and the CIP floor were the weakness and

the vertical crack came after the horizontal crack inplastic stage The precast element with high qualitiesdid not present cracks in the test

(2) The initial natural frequency as the initial stiffnessof MPCS degraded more obviously than CIPS forthe fine cracks in the connections The damping

12 Shock and Vibration

1 2 3 4 5 6 7 8 9 10 11 120Seismic force of DZC (kN)

1 2 3 4 5 6 7 8 9 10 11 120Seismic force of B-WSM (kN)

1 2 3 4 5 6 7 8 9 10 11 12 130Seismic force of ELW (kN)

123456789

101112

Floo

r

123456789

101112

Floo

r

123456789

101112

Floo

r

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(a) Distribution of seismic forces

10 20 30 40 50 60 70 80 90 1000Story shear of ELW (kN)

0123456789

101112

Floo

r10 20 30 40 50 60 70 80 900

Story shear of B-WSM (kN)

0123456789

101112

Floo

r

0123456789

101112

Floo

r

10 20 30 40 50 60 70 800Story shear of DZC (kN)

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Distribution of story shear

Figure 11 Earthquake actions of CIPS and MPCS

ratio of them had similar parameter but the way ofenergy dissipation of them was different And theyhad similar mode shapes

(3) The acceleration amplifying factors of them increasedalong the height of models and decreased progres-sively with the increasing PGA Nevertheless theyhad different responses under the different earth-quake waves The maximum acceleration amplifyingfactor of CIPS was 546 under the ELW with PGA of022 g and that of MPCS was 523 under the B-WSMwith PGA of 040 g

(4) Distribution of seismic forces of them increasedlinearly along the height ofmodels and then presentednonlinearity for the influence of high-order modesThe distribution of the interstory shear of CIPS and

MPCS showed similar rules like inverted triangle inelastic stage

(5) The maximum story displacements of them werealmost equal in elastic stage However the maximumstory displacement of CIPS was greater than thatof MPCS in plastic stage which was caused bythe concentrated failure in 4th story of CIPS Themaximum interstory ratio under the seismic waveswith PGA of 007 g and 040 g was in accordancewith the provision of Chinese code Under the moreintensive earthquake wave of PGA of 080 g the twomodels had enough capacity to resist collapse

Conflicts of Interest

The authors declare that they have no conflicts of interest

Shock and Vibration 13

5 10 15 20 25 30 35 40 450Displacement of B-WSM (mm)

0123456789

101112

Floo

r

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of DZC (mm)

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of ELW (mm)

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(a) Maximum story displacements of CIPS and MPCS

0123456789

101112

Floo

r

000

1

001

0

000

8

000

4

001

2

000

0

000

7

000

5

000

9

000

3

001

1

000

2

000

6

Story drift of DZC

000

1

000

8

000

4

001

0

000

0

000

60

007

000

3

000

9

000

2

001

1

000

5

Story drift of B-WSM

0123456789

101112

Floo

r

0123456789

101112

Floo

r

000

1

001

2

001

0

000

4

000

8

001

3

000

0

000

6

000

9

000

5

001

1

000

30

002

000

7

Story drift of ELW

Elastic limit Plastic limitElastic limit Plastic limitElastic limit Plastic limit

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Story drift ratio of CIPS and MPCS

Figure 12 Displacement responses of CIPS and MPCS

Acknowledgments

This research has been supported by both Innovation TeamofXirsquoan University of Architecture and Technology and ShaanxiProvince Science and Technology Coordination InnovationProject (nos 2015KTZDSF03-05-01 2015KTZDSF03-04 and2014SZS04-Z01)The support of the National Natural ScienceFoundation of China (Grants nos 51408456 51578444 and51478381) and Ministry of Education Plan for Yangtze RiverScholar and Innovation TeamDevelopment (no IRT13089) isalso acknowledged

References

[1] R Park and T Paulay Reinforced Concrete Structures ohnWileyamp Sons 1975

[2] M Fintel ldquoPerformance of buildings with shear walls in earth-quakes of the last thirty yearsrdquo PCI journal vol 40 no 3 pp62ndash80 1995

[3] A A Yee ldquoSplice sleeve for reinforcing barsrdquo US Pat vol 35401968

[4] J H Ling A B Abd Rahman I S Ibrahim and Z AbdulHamid ldquoBehaviour of grouted pipe splice under incrementaltensile loadrdquo Construction and Building Materials vol 33 no8 pp 90ndash98 2012

[5] M T El-Sheikh R Sause S Pessiki and L-W Lu ldquoSeismicbehavior and design of unbonded post-tensioned precast con-crete framesrdquo PCI journal vol 44 no 3 pp 54ndash71 1999

[6] R S Henry S Sritharan and J M Ingham ldquoUnbonded tendonstresses in post-tensioned concrete walls at nominal flexuralstrengthrdquo ACI Structural Journal vol 109 no 2 pp 171ndash1822012

14 Shock and Vibration

[7] P Seifi R S Henry and JM Ingham ldquoPanel connection detailsin existing New Zealand precast concrete buildingsrdquo Bulletin ofthe New Zealand Society for Earthquake Engineering vol 49 no2 pp 190ndash199 2016

[8] K A Soudki S H Rizkalla and R W Daikiw ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformations part 2 prestressed connectionsrdquo PCI journal vol40 no 5 pp 82ndash96 1995

[9] K A Soudki S H Rizkalla and B LeBlanc ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformationsmdashpart 1 mild steel connectionsrdquo PCI journal vol40 no 4 pp 78ndash96 1995

[10] K A Soudki J S West S H Rizkalla and B BlackettldquoHorizontal connections for precast concrete shear wall panelsunder cyclic shear loadingrdquo PCI journal vol 41 no 3 pp 64ndash80 1996

[11] R Vaghei F Hejazi H Taheri M S Jaafar and A A Ali ldquoEval-uate performance of precast concrete wall to wall connectionrdquoAPCBEE Procedia vol 9 pp 285ndash290 2014

[12] H Jiang Z Chen J Zhang B Wu Y Tian andW Liu ldquoQuasi-static test of precast reinforced concrete shear wall structurerdquoJianzhu Jiegou XuebaoJournal of Building Structures vol 32 no6 pp 34ndash40 2011

[13] Z ChenH Jiang J Zhang BWu Y Tian andW Liu ldquoPseudo-dynamic substructure test on precast reinforced concrete shearwall structurerdquo Jianzhu Jiegou XuebaoJournal of Building Struc-tures vol 32 no 6 pp 41ndash50 2011

[14] J R QianW L Han Z Z Zhao and H Qin ldquoPseudo-dynamicSubstructure Test on a 3-story Full-scaleModel of PrefabricatedConcrete Shear Wall Structure with Rebars Splicing by GroutSleevesrdquo Journal of Building Structures vol 38 no 3 pp 26ndash382017

[15] P Negro D A Bournas and F J Molina ldquoPseudodynamictests on a full-scale 3-storey precast concrete building Globalresponserdquo Engineering Structures vol 57 pp 594ndash608 2013

[16] M Oliva P Gavrilovic and R W Clough ldquoSeismic testing oflarge panel precast walls Comparison of pseudostatic and shak-ing table testsrdquo Earthquake Engineering amp Structural Dynamicsvol 19 no 6 pp 859ndash875 1990

[17] L H Lee W H Yi and S Y Seo ldquoShaking Table Test of PrecastConcrete Wall Structurerdquo in Proceedings of the Eleventh WorldConference on Earthquake Engineering

[18] W Wang A Q Li et al ldquoShaking table test of precast concreteshear wall isolation structurerdquo Journal of Building Structuresvol 36 no 6 pp 35ndash43 2015

[19] EERI Preliminary Reconnaissance Report Earthquake Engi-neering Research Institute 1994

[20] EERI Armenia Earthquake Reconnaissance Report EarthquakeEngineering Research Institute 1989

[21] GB 50011-2010 Code for Seismic Design of Buildings ChinaArchitecture and Building Press Beijing China 2010

[22] JGJ 1-2014 Technical Specification for Precast Concrete Struc-tures China Architecture amp Building Press Beijing China2014

[23] 15J939-1 Example of Architectural Design for PrefabricatedConcrete Structures China Planning Press Beijing China 2015

[24] E Buckingham ldquoOn physically similar systems Illustrations ofthe use of dimensional equationsrdquo Physical Review A AtomicMolecular and Optical Physics vol 4 no 4 pp 345ndash376 1914

[25] Y Zhou andX L LuTheexperimentalmethod and technology ofshaking table test about building structure Science Press BeijingChina 2012

[26] JGJT 101-2015 Specification for Seismic Test of Buildings ChinaArchitecture amp Building Press Beijing China 2014

[27] C Y Shen F L Zhou et al ldquoExperimental research on micro-concrete used in dynamic test modelrdquo Journal of GuangzhouUniversity (Natural Science Edition) vol 4 no 3 pp 250ndash2532005

[28] J Xiao T L Pham and T Ding ldquoShake table test on seismicresponse of a precast frame with recycled aggregate concreterdquoAdvances in Structural Engineering vol 18 no 9 pp 1517ndash15342015

[29] S C Chakrabarti G C Nayak and D K Paul ldquoShear charac-teristics of cast-in-place vertical joints in story-high precast wallassemblyrdquo ACI Structural Journal vol 85 no 1 pp 30ndash45 1988

[30] CWang and J Xiao ldquoStudy of the seismic response of a recycledaggregate concrete frame structurerdquo Earthquake Engineeringand Engineering Vibration vol 12 no 4 pp 669ndash680 2013

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 10: Comparative Study of Seismic Behavior between Monolithic

10 Shock and Vibration

0123456789

101112

Floo

r

01 02 03 04 05 06 07 08 09 10 1100Mode coefficient of first mode shape

0123456789

101112

Floo

r

01 02 03 04 05 06 07 08 09 10 1100Mode coefficient of first mode shape

minus08 minus06 minus04 minus02minus10 02 04 06 08 1000Mode coefficient of second mode shape

minus06 minus04 minus02minus08minus10 02 04 06 08 1000Mode coefficient of second mode shape

0123456789

101112

Floo

r

0123456789

101112

Floo

r

MPCS-initialMPCS-0035 gMPCS-007 gMPCS-014 g

MPCS-022 gMPCS-040 gMPCS-062 g

MPCS-initialMPCS-0035 gMPCS-007 gMPCS-014 g

MPCS-022 gMPCS-040 gMPCS-062 g

CIPS-initialCIPS-0035 gCIPS-007 gCIPS-014 g

CIPS-022 gCIPS-040 gCIPS-062 g

CIPS-initialCIPS-0035 gCIPS-007 gCIPS-014 g

CIPS-022 gCIPS-040 gCIPS-062 g

Figure 9 Mode shapes of CIPS and MPCS

maximumdestructive story driftwas in 4th floorMeanwhilethe two models showed obviously nonlinear behavior withthe PGA of 040 g and 062 g The shape of maximum storydisplacements of them was similar to the first mode shapeThe maximum story displacements were 4867mm in CIPSand 4198mm in MPCS

The maximum interstory drift ratios were calculated andpresented in Figure 12(b) The maximum value of story driftof CIPS was 11005 in 5th story under the DZC earthquakewave with the PGA of 070 g and that of MPCS was 11020in 5th story under the same wave They met the provisionabout the maximum story drift in elastic stage in area of 8-degree seismic fortification intensity in code of GB 50011-2010 [21] In the stage from PGA of 0035 g to 014 g thestory drift increased progressively The seismic design codeasks for the limiting value of plastic interstory drift ratio toforbid collapse We checked the story drift under the PGA of040 g The values of CIPS and MPCS were 1121 and 1127

which met the limiting value in rare earthquake action of 8-degree seismic regions Under the excitation of PGAof 062 gthe values exceeded the limiting value in rare earthquakeaction of 9-degree seismic regions And then the moreintensive earthquakewaves with PGAof 080 gwere inputtedthe models presented better seismic behavior withoutcollapse

5 Conclusion

A comparative seismic study between CIPS and MPCS wasconducted by shaking table test which contained a 15 scaled12-story CIPS model and MPCS model The experimentalresults of dynamic characteristic failure pattern and mecha-nism and seismic response of the models were discussed andcompared with each other to better understand their seismicbehavior Based on the intensive analysis of test results thefollowing conclusions are derived

Shock and Vibration 11

PGA of 062 gPGA of 040 gPGA of 022 gPGA of 014 gPGA of 007 g PGA of 062 gPGA of 040 gPGA of 022 gPGA of 014 gPGA of 007 g

EE E EB B B BDD DD

E(LW

)

D(Z

C)

B(-W

SM)

Peak ground acceleration of CIPS (g)

E EEEB B BB DDD D

E(LW

)

D(Z

C)

B(-W

SM)

Peak ground acceleration of MPCS (g)

12th floor11th floor10th floor9th floor8th floor

7th floor6th floor5th floor4th floor3rd floor

2nd floor1st floorFloor

12th floor11th floor10th floor9th floor8th floor

7th floor6th floor5th floor4th floor3rd floor

2nd floor1st floorFloor

081216202428323640444852566064

Acce

lera

tion

ampl

ifyin

g co

effici

ent

081216202428323640444852566064

Acce

lera

tion

ampl

ifyin

g co

effici

ent

(a) Acceleration amplifying factors changed along the test phases of CIPS and MPCS

0123456789

101112

Floo

r

15 20 25 30 35 40 45 50 5510Acceleration amplifying coefficient of B-WSM

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 g

MPCS-040 gMPCS-062 gCIPS-0035 gCIPS-007 g

CIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Acceleration amplifying factors of B-WSMwave along the height oftwo models

Figure 10 Acceleration amplifying factors of CIPS and MPCS

(1) The typical failure pattern of the CIPS was damageconcentration in coupling beams in first and thenthe cracks emerged in the shear wall in plasticstage However besides the concentrated damage incoupling beam the connections between the precastelement and the CIP floor were the weakness and

the vertical crack came after the horizontal crack inplastic stage The precast element with high qualitiesdid not present cracks in the test

(2) The initial natural frequency as the initial stiffnessof MPCS degraded more obviously than CIPS forthe fine cracks in the connections The damping

12 Shock and Vibration

1 2 3 4 5 6 7 8 9 10 11 120Seismic force of DZC (kN)

1 2 3 4 5 6 7 8 9 10 11 120Seismic force of B-WSM (kN)

1 2 3 4 5 6 7 8 9 10 11 12 130Seismic force of ELW (kN)

123456789

101112

Floo

r

123456789

101112

Floo

r

123456789

101112

Floo

r

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(a) Distribution of seismic forces

10 20 30 40 50 60 70 80 90 1000Story shear of ELW (kN)

0123456789

101112

Floo

r10 20 30 40 50 60 70 80 900

Story shear of B-WSM (kN)

0123456789

101112

Floo

r

0123456789

101112

Floo

r

10 20 30 40 50 60 70 800Story shear of DZC (kN)

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Distribution of story shear

Figure 11 Earthquake actions of CIPS and MPCS

ratio of them had similar parameter but the way ofenergy dissipation of them was different And theyhad similar mode shapes

(3) The acceleration amplifying factors of them increasedalong the height of models and decreased progres-sively with the increasing PGA Nevertheless theyhad different responses under the different earth-quake waves The maximum acceleration amplifyingfactor of CIPS was 546 under the ELW with PGA of022 g and that of MPCS was 523 under the B-WSMwith PGA of 040 g

(4) Distribution of seismic forces of them increasedlinearly along the height ofmodels and then presentednonlinearity for the influence of high-order modesThe distribution of the interstory shear of CIPS and

MPCS showed similar rules like inverted triangle inelastic stage

(5) The maximum story displacements of them werealmost equal in elastic stage However the maximumstory displacement of CIPS was greater than thatof MPCS in plastic stage which was caused bythe concentrated failure in 4th story of CIPS Themaximum interstory ratio under the seismic waveswith PGA of 007 g and 040 g was in accordancewith the provision of Chinese code Under the moreintensive earthquake wave of PGA of 080 g the twomodels had enough capacity to resist collapse

Conflicts of Interest

The authors declare that they have no conflicts of interest

Shock and Vibration 13

5 10 15 20 25 30 35 40 450Displacement of B-WSM (mm)

0123456789

101112

Floo

r

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of DZC (mm)

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of ELW (mm)

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(a) Maximum story displacements of CIPS and MPCS

0123456789

101112

Floo

r

000

1

001

0

000

8

000

4

001

2

000

0

000

7

000

5

000

9

000

3

001

1

000

2

000

6

Story drift of DZC

000

1

000

8

000

4

001

0

000

0

000

60

007

000

3

000

9

000

2

001

1

000

5

Story drift of B-WSM

0123456789

101112

Floo

r

0123456789

101112

Floo

r

000

1

001

2

001

0

000

4

000

8

001

3

000

0

000

6

000

9

000

5

001

1

000

30

002

000

7

Story drift of ELW

Elastic limit Plastic limitElastic limit Plastic limitElastic limit Plastic limit

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Story drift ratio of CIPS and MPCS

Figure 12 Displacement responses of CIPS and MPCS

Acknowledgments

This research has been supported by both Innovation TeamofXirsquoan University of Architecture and Technology and ShaanxiProvince Science and Technology Coordination InnovationProject (nos 2015KTZDSF03-05-01 2015KTZDSF03-04 and2014SZS04-Z01)The support of the National Natural ScienceFoundation of China (Grants nos 51408456 51578444 and51478381) and Ministry of Education Plan for Yangtze RiverScholar and Innovation TeamDevelopment (no IRT13089) isalso acknowledged

References

[1] R Park and T Paulay Reinforced Concrete Structures ohnWileyamp Sons 1975

[2] M Fintel ldquoPerformance of buildings with shear walls in earth-quakes of the last thirty yearsrdquo PCI journal vol 40 no 3 pp62ndash80 1995

[3] A A Yee ldquoSplice sleeve for reinforcing barsrdquo US Pat vol 35401968

[4] J H Ling A B Abd Rahman I S Ibrahim and Z AbdulHamid ldquoBehaviour of grouted pipe splice under incrementaltensile loadrdquo Construction and Building Materials vol 33 no8 pp 90ndash98 2012

[5] M T El-Sheikh R Sause S Pessiki and L-W Lu ldquoSeismicbehavior and design of unbonded post-tensioned precast con-crete framesrdquo PCI journal vol 44 no 3 pp 54ndash71 1999

[6] R S Henry S Sritharan and J M Ingham ldquoUnbonded tendonstresses in post-tensioned concrete walls at nominal flexuralstrengthrdquo ACI Structural Journal vol 109 no 2 pp 171ndash1822012

14 Shock and Vibration

[7] P Seifi R S Henry and JM Ingham ldquoPanel connection detailsin existing New Zealand precast concrete buildingsrdquo Bulletin ofthe New Zealand Society for Earthquake Engineering vol 49 no2 pp 190ndash199 2016

[8] K A Soudki S H Rizkalla and R W Daikiw ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformations part 2 prestressed connectionsrdquo PCI journal vol40 no 5 pp 82ndash96 1995

[9] K A Soudki S H Rizkalla and B LeBlanc ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformationsmdashpart 1 mild steel connectionsrdquo PCI journal vol40 no 4 pp 78ndash96 1995

[10] K A Soudki J S West S H Rizkalla and B BlackettldquoHorizontal connections for precast concrete shear wall panelsunder cyclic shear loadingrdquo PCI journal vol 41 no 3 pp 64ndash80 1996

[11] R Vaghei F Hejazi H Taheri M S Jaafar and A A Ali ldquoEval-uate performance of precast concrete wall to wall connectionrdquoAPCBEE Procedia vol 9 pp 285ndash290 2014

[12] H Jiang Z Chen J Zhang B Wu Y Tian andW Liu ldquoQuasi-static test of precast reinforced concrete shear wall structurerdquoJianzhu Jiegou XuebaoJournal of Building Structures vol 32 no6 pp 34ndash40 2011

[13] Z ChenH Jiang J Zhang BWu Y Tian andW Liu ldquoPseudo-dynamic substructure test on precast reinforced concrete shearwall structurerdquo Jianzhu Jiegou XuebaoJournal of Building Struc-tures vol 32 no 6 pp 41ndash50 2011

[14] J R QianW L Han Z Z Zhao and H Qin ldquoPseudo-dynamicSubstructure Test on a 3-story Full-scaleModel of PrefabricatedConcrete Shear Wall Structure with Rebars Splicing by GroutSleevesrdquo Journal of Building Structures vol 38 no 3 pp 26ndash382017

[15] P Negro D A Bournas and F J Molina ldquoPseudodynamictests on a full-scale 3-storey precast concrete building Globalresponserdquo Engineering Structures vol 57 pp 594ndash608 2013

[16] M Oliva P Gavrilovic and R W Clough ldquoSeismic testing oflarge panel precast walls Comparison of pseudostatic and shak-ing table testsrdquo Earthquake Engineering amp Structural Dynamicsvol 19 no 6 pp 859ndash875 1990

[17] L H Lee W H Yi and S Y Seo ldquoShaking Table Test of PrecastConcrete Wall Structurerdquo in Proceedings of the Eleventh WorldConference on Earthquake Engineering

[18] W Wang A Q Li et al ldquoShaking table test of precast concreteshear wall isolation structurerdquo Journal of Building Structuresvol 36 no 6 pp 35ndash43 2015

[19] EERI Preliminary Reconnaissance Report Earthquake Engi-neering Research Institute 1994

[20] EERI Armenia Earthquake Reconnaissance Report EarthquakeEngineering Research Institute 1989

[21] GB 50011-2010 Code for Seismic Design of Buildings ChinaArchitecture and Building Press Beijing China 2010

[22] JGJ 1-2014 Technical Specification for Precast Concrete Struc-tures China Architecture amp Building Press Beijing China2014

[23] 15J939-1 Example of Architectural Design for PrefabricatedConcrete Structures China Planning Press Beijing China 2015

[24] E Buckingham ldquoOn physically similar systems Illustrations ofthe use of dimensional equationsrdquo Physical Review A AtomicMolecular and Optical Physics vol 4 no 4 pp 345ndash376 1914

[25] Y Zhou andX L LuTheexperimentalmethod and technology ofshaking table test about building structure Science Press BeijingChina 2012

[26] JGJT 101-2015 Specification for Seismic Test of Buildings ChinaArchitecture amp Building Press Beijing China 2014

[27] C Y Shen F L Zhou et al ldquoExperimental research on micro-concrete used in dynamic test modelrdquo Journal of GuangzhouUniversity (Natural Science Edition) vol 4 no 3 pp 250ndash2532005

[28] J Xiao T L Pham and T Ding ldquoShake table test on seismicresponse of a precast frame with recycled aggregate concreterdquoAdvances in Structural Engineering vol 18 no 9 pp 1517ndash15342015

[29] S C Chakrabarti G C Nayak and D K Paul ldquoShear charac-teristics of cast-in-place vertical joints in story-high precast wallassemblyrdquo ACI Structural Journal vol 85 no 1 pp 30ndash45 1988

[30] CWang and J Xiao ldquoStudy of the seismic response of a recycledaggregate concrete frame structurerdquo Earthquake Engineeringand Engineering Vibration vol 12 no 4 pp 669ndash680 2013

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 11: Comparative Study of Seismic Behavior between Monolithic

Shock and Vibration 11

PGA of 062 gPGA of 040 gPGA of 022 gPGA of 014 gPGA of 007 g PGA of 062 gPGA of 040 gPGA of 022 gPGA of 014 gPGA of 007 g

EE E EB B B BDD DD

E(LW

)

D(Z

C)

B(-W

SM)

Peak ground acceleration of CIPS (g)

E EEEB B BB DDD D

E(LW

)

D(Z

C)

B(-W

SM)

Peak ground acceleration of MPCS (g)

12th floor11th floor10th floor9th floor8th floor

7th floor6th floor5th floor4th floor3rd floor

2nd floor1st floorFloor

12th floor11th floor10th floor9th floor8th floor

7th floor6th floor5th floor4th floor3rd floor

2nd floor1st floorFloor

081216202428323640444852566064

Acce

lera

tion

ampl

ifyin

g co

effici

ent

081216202428323640444852566064

Acce

lera

tion

ampl

ifyin

g co

effici

ent

(a) Acceleration amplifying factors changed along the test phases of CIPS and MPCS

0123456789

101112

Floo

r

15 20 25 30 35 40 45 50 5510Acceleration amplifying coefficient of B-WSM

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 g

MPCS-040 gMPCS-062 gCIPS-0035 gCIPS-007 g

CIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Acceleration amplifying factors of B-WSMwave along the height oftwo models

Figure 10 Acceleration amplifying factors of CIPS and MPCS

(1) The typical failure pattern of the CIPS was damageconcentration in coupling beams in first and thenthe cracks emerged in the shear wall in plasticstage However besides the concentrated damage incoupling beam the connections between the precastelement and the CIP floor were the weakness and

the vertical crack came after the horizontal crack inplastic stage The precast element with high qualitiesdid not present cracks in the test

(2) The initial natural frequency as the initial stiffnessof MPCS degraded more obviously than CIPS forthe fine cracks in the connections The damping

12 Shock and Vibration

1 2 3 4 5 6 7 8 9 10 11 120Seismic force of DZC (kN)

1 2 3 4 5 6 7 8 9 10 11 120Seismic force of B-WSM (kN)

1 2 3 4 5 6 7 8 9 10 11 12 130Seismic force of ELW (kN)

123456789

101112

Floo

r

123456789

101112

Floo

r

123456789

101112

Floo

r

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(a) Distribution of seismic forces

10 20 30 40 50 60 70 80 90 1000Story shear of ELW (kN)

0123456789

101112

Floo

r10 20 30 40 50 60 70 80 900

Story shear of B-WSM (kN)

0123456789

101112

Floo

r

0123456789

101112

Floo

r

10 20 30 40 50 60 70 800Story shear of DZC (kN)

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Distribution of story shear

Figure 11 Earthquake actions of CIPS and MPCS

ratio of them had similar parameter but the way ofenergy dissipation of them was different And theyhad similar mode shapes

(3) The acceleration amplifying factors of them increasedalong the height of models and decreased progres-sively with the increasing PGA Nevertheless theyhad different responses under the different earth-quake waves The maximum acceleration amplifyingfactor of CIPS was 546 under the ELW with PGA of022 g and that of MPCS was 523 under the B-WSMwith PGA of 040 g

(4) Distribution of seismic forces of them increasedlinearly along the height ofmodels and then presentednonlinearity for the influence of high-order modesThe distribution of the interstory shear of CIPS and

MPCS showed similar rules like inverted triangle inelastic stage

(5) The maximum story displacements of them werealmost equal in elastic stage However the maximumstory displacement of CIPS was greater than thatof MPCS in plastic stage which was caused bythe concentrated failure in 4th story of CIPS Themaximum interstory ratio under the seismic waveswith PGA of 007 g and 040 g was in accordancewith the provision of Chinese code Under the moreintensive earthquake wave of PGA of 080 g the twomodels had enough capacity to resist collapse

Conflicts of Interest

The authors declare that they have no conflicts of interest

Shock and Vibration 13

5 10 15 20 25 30 35 40 450Displacement of B-WSM (mm)

0123456789

101112

Floo

r

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of DZC (mm)

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of ELW (mm)

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(a) Maximum story displacements of CIPS and MPCS

0123456789

101112

Floo

r

000

1

001

0

000

8

000

4

001

2

000

0

000

7

000

5

000

9

000

3

001

1

000

2

000

6

Story drift of DZC

000

1

000

8

000

4

001

0

000

0

000

60

007

000

3

000

9

000

2

001

1

000

5

Story drift of B-WSM

0123456789

101112

Floo

r

0123456789

101112

Floo

r

000

1

001

2

001

0

000

4

000

8

001

3

000

0

000

6

000

9

000

5

001

1

000

30

002

000

7

Story drift of ELW

Elastic limit Plastic limitElastic limit Plastic limitElastic limit Plastic limit

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Story drift ratio of CIPS and MPCS

Figure 12 Displacement responses of CIPS and MPCS

Acknowledgments

This research has been supported by both Innovation TeamofXirsquoan University of Architecture and Technology and ShaanxiProvince Science and Technology Coordination InnovationProject (nos 2015KTZDSF03-05-01 2015KTZDSF03-04 and2014SZS04-Z01)The support of the National Natural ScienceFoundation of China (Grants nos 51408456 51578444 and51478381) and Ministry of Education Plan for Yangtze RiverScholar and Innovation TeamDevelopment (no IRT13089) isalso acknowledged

References

[1] R Park and T Paulay Reinforced Concrete Structures ohnWileyamp Sons 1975

[2] M Fintel ldquoPerformance of buildings with shear walls in earth-quakes of the last thirty yearsrdquo PCI journal vol 40 no 3 pp62ndash80 1995

[3] A A Yee ldquoSplice sleeve for reinforcing barsrdquo US Pat vol 35401968

[4] J H Ling A B Abd Rahman I S Ibrahim and Z AbdulHamid ldquoBehaviour of grouted pipe splice under incrementaltensile loadrdquo Construction and Building Materials vol 33 no8 pp 90ndash98 2012

[5] M T El-Sheikh R Sause S Pessiki and L-W Lu ldquoSeismicbehavior and design of unbonded post-tensioned precast con-crete framesrdquo PCI journal vol 44 no 3 pp 54ndash71 1999

[6] R S Henry S Sritharan and J M Ingham ldquoUnbonded tendonstresses in post-tensioned concrete walls at nominal flexuralstrengthrdquo ACI Structural Journal vol 109 no 2 pp 171ndash1822012

14 Shock and Vibration

[7] P Seifi R S Henry and JM Ingham ldquoPanel connection detailsin existing New Zealand precast concrete buildingsrdquo Bulletin ofthe New Zealand Society for Earthquake Engineering vol 49 no2 pp 190ndash199 2016

[8] K A Soudki S H Rizkalla and R W Daikiw ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformations part 2 prestressed connectionsrdquo PCI journal vol40 no 5 pp 82ndash96 1995

[9] K A Soudki S H Rizkalla and B LeBlanc ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformationsmdashpart 1 mild steel connectionsrdquo PCI journal vol40 no 4 pp 78ndash96 1995

[10] K A Soudki J S West S H Rizkalla and B BlackettldquoHorizontal connections for precast concrete shear wall panelsunder cyclic shear loadingrdquo PCI journal vol 41 no 3 pp 64ndash80 1996

[11] R Vaghei F Hejazi H Taheri M S Jaafar and A A Ali ldquoEval-uate performance of precast concrete wall to wall connectionrdquoAPCBEE Procedia vol 9 pp 285ndash290 2014

[12] H Jiang Z Chen J Zhang B Wu Y Tian andW Liu ldquoQuasi-static test of precast reinforced concrete shear wall structurerdquoJianzhu Jiegou XuebaoJournal of Building Structures vol 32 no6 pp 34ndash40 2011

[13] Z ChenH Jiang J Zhang BWu Y Tian andW Liu ldquoPseudo-dynamic substructure test on precast reinforced concrete shearwall structurerdquo Jianzhu Jiegou XuebaoJournal of Building Struc-tures vol 32 no 6 pp 41ndash50 2011

[14] J R QianW L Han Z Z Zhao and H Qin ldquoPseudo-dynamicSubstructure Test on a 3-story Full-scaleModel of PrefabricatedConcrete Shear Wall Structure with Rebars Splicing by GroutSleevesrdquo Journal of Building Structures vol 38 no 3 pp 26ndash382017

[15] P Negro D A Bournas and F J Molina ldquoPseudodynamictests on a full-scale 3-storey precast concrete building Globalresponserdquo Engineering Structures vol 57 pp 594ndash608 2013

[16] M Oliva P Gavrilovic and R W Clough ldquoSeismic testing oflarge panel precast walls Comparison of pseudostatic and shak-ing table testsrdquo Earthquake Engineering amp Structural Dynamicsvol 19 no 6 pp 859ndash875 1990

[17] L H Lee W H Yi and S Y Seo ldquoShaking Table Test of PrecastConcrete Wall Structurerdquo in Proceedings of the Eleventh WorldConference on Earthquake Engineering

[18] W Wang A Q Li et al ldquoShaking table test of precast concreteshear wall isolation structurerdquo Journal of Building Structuresvol 36 no 6 pp 35ndash43 2015

[19] EERI Preliminary Reconnaissance Report Earthquake Engi-neering Research Institute 1994

[20] EERI Armenia Earthquake Reconnaissance Report EarthquakeEngineering Research Institute 1989

[21] GB 50011-2010 Code for Seismic Design of Buildings ChinaArchitecture and Building Press Beijing China 2010

[22] JGJ 1-2014 Technical Specification for Precast Concrete Struc-tures China Architecture amp Building Press Beijing China2014

[23] 15J939-1 Example of Architectural Design for PrefabricatedConcrete Structures China Planning Press Beijing China 2015

[24] E Buckingham ldquoOn physically similar systems Illustrations ofthe use of dimensional equationsrdquo Physical Review A AtomicMolecular and Optical Physics vol 4 no 4 pp 345ndash376 1914

[25] Y Zhou andX L LuTheexperimentalmethod and technology ofshaking table test about building structure Science Press BeijingChina 2012

[26] JGJT 101-2015 Specification for Seismic Test of Buildings ChinaArchitecture amp Building Press Beijing China 2014

[27] C Y Shen F L Zhou et al ldquoExperimental research on micro-concrete used in dynamic test modelrdquo Journal of GuangzhouUniversity (Natural Science Edition) vol 4 no 3 pp 250ndash2532005

[28] J Xiao T L Pham and T Ding ldquoShake table test on seismicresponse of a precast frame with recycled aggregate concreterdquoAdvances in Structural Engineering vol 18 no 9 pp 1517ndash15342015

[29] S C Chakrabarti G C Nayak and D K Paul ldquoShear charac-teristics of cast-in-place vertical joints in story-high precast wallassemblyrdquo ACI Structural Journal vol 85 no 1 pp 30ndash45 1988

[30] CWang and J Xiao ldquoStudy of the seismic response of a recycledaggregate concrete frame structurerdquo Earthquake Engineeringand Engineering Vibration vol 12 no 4 pp 669ndash680 2013

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 12: Comparative Study of Seismic Behavior between Monolithic

12 Shock and Vibration

1 2 3 4 5 6 7 8 9 10 11 120Seismic force of DZC (kN)

1 2 3 4 5 6 7 8 9 10 11 120Seismic force of B-WSM (kN)

1 2 3 4 5 6 7 8 9 10 11 12 130Seismic force of ELW (kN)

123456789

101112

Floo

r

123456789

101112

Floo

r

123456789

101112

Floo

r

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(a) Distribution of seismic forces

10 20 30 40 50 60 70 80 90 1000Story shear of ELW (kN)

0123456789

101112

Floo

r10 20 30 40 50 60 70 80 900

Story shear of B-WSM (kN)

0123456789

101112

Floo

r

0123456789

101112

Floo

r

10 20 30 40 50 60 70 800Story shear of DZC (kN)

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Distribution of story shear

Figure 11 Earthquake actions of CIPS and MPCS

ratio of them had similar parameter but the way ofenergy dissipation of them was different And theyhad similar mode shapes

(3) The acceleration amplifying factors of them increasedalong the height of models and decreased progres-sively with the increasing PGA Nevertheless theyhad different responses under the different earth-quake waves The maximum acceleration amplifyingfactor of CIPS was 546 under the ELW with PGA of022 g and that of MPCS was 523 under the B-WSMwith PGA of 040 g

(4) Distribution of seismic forces of them increasedlinearly along the height ofmodels and then presentednonlinearity for the influence of high-order modesThe distribution of the interstory shear of CIPS and

MPCS showed similar rules like inverted triangle inelastic stage

(5) The maximum story displacements of them werealmost equal in elastic stage However the maximumstory displacement of CIPS was greater than thatof MPCS in plastic stage which was caused bythe concentrated failure in 4th story of CIPS Themaximum interstory ratio under the seismic waveswith PGA of 007 g and 040 g was in accordancewith the provision of Chinese code Under the moreintensive earthquake wave of PGA of 080 g the twomodels had enough capacity to resist collapse

Conflicts of Interest

The authors declare that they have no conflicts of interest

Shock and Vibration 13

5 10 15 20 25 30 35 40 450Displacement of B-WSM (mm)

0123456789

101112

Floo

r

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of DZC (mm)

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of ELW (mm)

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(a) Maximum story displacements of CIPS and MPCS

0123456789

101112

Floo

r

000

1

001

0

000

8

000

4

001

2

000

0

000

7

000

5

000

9

000

3

001

1

000

2

000

6

Story drift of DZC

000

1

000

8

000

4

001

0

000

0

000

60

007

000

3

000

9

000

2

001

1

000

5

Story drift of B-WSM

0123456789

101112

Floo

r

0123456789

101112

Floo

r

000

1

001

2

001

0

000

4

000

8

001

3

000

0

000

6

000

9

000

5

001

1

000

30

002

000

7

Story drift of ELW

Elastic limit Plastic limitElastic limit Plastic limitElastic limit Plastic limit

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Story drift ratio of CIPS and MPCS

Figure 12 Displacement responses of CIPS and MPCS

Acknowledgments

This research has been supported by both Innovation TeamofXirsquoan University of Architecture and Technology and ShaanxiProvince Science and Technology Coordination InnovationProject (nos 2015KTZDSF03-05-01 2015KTZDSF03-04 and2014SZS04-Z01)The support of the National Natural ScienceFoundation of China (Grants nos 51408456 51578444 and51478381) and Ministry of Education Plan for Yangtze RiverScholar and Innovation TeamDevelopment (no IRT13089) isalso acknowledged

References

[1] R Park and T Paulay Reinforced Concrete Structures ohnWileyamp Sons 1975

[2] M Fintel ldquoPerformance of buildings with shear walls in earth-quakes of the last thirty yearsrdquo PCI journal vol 40 no 3 pp62ndash80 1995

[3] A A Yee ldquoSplice sleeve for reinforcing barsrdquo US Pat vol 35401968

[4] J H Ling A B Abd Rahman I S Ibrahim and Z AbdulHamid ldquoBehaviour of grouted pipe splice under incrementaltensile loadrdquo Construction and Building Materials vol 33 no8 pp 90ndash98 2012

[5] M T El-Sheikh R Sause S Pessiki and L-W Lu ldquoSeismicbehavior and design of unbonded post-tensioned precast con-crete framesrdquo PCI journal vol 44 no 3 pp 54ndash71 1999

[6] R S Henry S Sritharan and J M Ingham ldquoUnbonded tendonstresses in post-tensioned concrete walls at nominal flexuralstrengthrdquo ACI Structural Journal vol 109 no 2 pp 171ndash1822012

14 Shock and Vibration

[7] P Seifi R S Henry and JM Ingham ldquoPanel connection detailsin existing New Zealand precast concrete buildingsrdquo Bulletin ofthe New Zealand Society for Earthquake Engineering vol 49 no2 pp 190ndash199 2016

[8] K A Soudki S H Rizkalla and R W Daikiw ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformations part 2 prestressed connectionsrdquo PCI journal vol40 no 5 pp 82ndash96 1995

[9] K A Soudki S H Rizkalla and B LeBlanc ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformationsmdashpart 1 mild steel connectionsrdquo PCI journal vol40 no 4 pp 78ndash96 1995

[10] K A Soudki J S West S H Rizkalla and B BlackettldquoHorizontal connections for precast concrete shear wall panelsunder cyclic shear loadingrdquo PCI journal vol 41 no 3 pp 64ndash80 1996

[11] R Vaghei F Hejazi H Taheri M S Jaafar and A A Ali ldquoEval-uate performance of precast concrete wall to wall connectionrdquoAPCBEE Procedia vol 9 pp 285ndash290 2014

[12] H Jiang Z Chen J Zhang B Wu Y Tian andW Liu ldquoQuasi-static test of precast reinforced concrete shear wall structurerdquoJianzhu Jiegou XuebaoJournal of Building Structures vol 32 no6 pp 34ndash40 2011

[13] Z ChenH Jiang J Zhang BWu Y Tian andW Liu ldquoPseudo-dynamic substructure test on precast reinforced concrete shearwall structurerdquo Jianzhu Jiegou XuebaoJournal of Building Struc-tures vol 32 no 6 pp 41ndash50 2011

[14] J R QianW L Han Z Z Zhao and H Qin ldquoPseudo-dynamicSubstructure Test on a 3-story Full-scaleModel of PrefabricatedConcrete Shear Wall Structure with Rebars Splicing by GroutSleevesrdquo Journal of Building Structures vol 38 no 3 pp 26ndash382017

[15] P Negro D A Bournas and F J Molina ldquoPseudodynamictests on a full-scale 3-storey precast concrete building Globalresponserdquo Engineering Structures vol 57 pp 594ndash608 2013

[16] M Oliva P Gavrilovic and R W Clough ldquoSeismic testing oflarge panel precast walls Comparison of pseudostatic and shak-ing table testsrdquo Earthquake Engineering amp Structural Dynamicsvol 19 no 6 pp 859ndash875 1990

[17] L H Lee W H Yi and S Y Seo ldquoShaking Table Test of PrecastConcrete Wall Structurerdquo in Proceedings of the Eleventh WorldConference on Earthquake Engineering

[18] W Wang A Q Li et al ldquoShaking table test of precast concreteshear wall isolation structurerdquo Journal of Building Structuresvol 36 no 6 pp 35ndash43 2015

[19] EERI Preliminary Reconnaissance Report Earthquake Engi-neering Research Institute 1994

[20] EERI Armenia Earthquake Reconnaissance Report EarthquakeEngineering Research Institute 1989

[21] GB 50011-2010 Code for Seismic Design of Buildings ChinaArchitecture and Building Press Beijing China 2010

[22] JGJ 1-2014 Technical Specification for Precast Concrete Struc-tures China Architecture amp Building Press Beijing China2014

[23] 15J939-1 Example of Architectural Design for PrefabricatedConcrete Structures China Planning Press Beijing China 2015

[24] E Buckingham ldquoOn physically similar systems Illustrations ofthe use of dimensional equationsrdquo Physical Review A AtomicMolecular and Optical Physics vol 4 no 4 pp 345ndash376 1914

[25] Y Zhou andX L LuTheexperimentalmethod and technology ofshaking table test about building structure Science Press BeijingChina 2012

[26] JGJT 101-2015 Specification for Seismic Test of Buildings ChinaArchitecture amp Building Press Beijing China 2014

[27] C Y Shen F L Zhou et al ldquoExperimental research on micro-concrete used in dynamic test modelrdquo Journal of GuangzhouUniversity (Natural Science Edition) vol 4 no 3 pp 250ndash2532005

[28] J Xiao T L Pham and T Ding ldquoShake table test on seismicresponse of a precast frame with recycled aggregate concreterdquoAdvances in Structural Engineering vol 18 no 9 pp 1517ndash15342015

[29] S C Chakrabarti G C Nayak and D K Paul ldquoShear charac-teristics of cast-in-place vertical joints in story-high precast wallassemblyrdquo ACI Structural Journal vol 85 no 1 pp 30ndash45 1988

[30] CWang and J Xiao ldquoStudy of the seismic response of a recycledaggregate concrete frame structurerdquo Earthquake Engineeringand Engineering Vibration vol 12 no 4 pp 669ndash680 2013

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 13: Comparative Study of Seismic Behavior between Monolithic

Shock and Vibration 13

5 10 15 20 25 30 35 40 450Displacement of B-WSM (mm)

0123456789

101112

Floo

r

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of DZC (mm)

0123456789

101112

Floo

r

5 10 15 20 25 30 35 40 45 500Displacement of ELW (mm)

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(a) Maximum story displacements of CIPS and MPCS

0123456789

101112

Floo

r

000

1

001

0

000

8

000

4

001

2

000

0

000

7

000

5

000

9

000

3

001

1

000

2

000

6

Story drift of DZC

000

1

000

8

000

4

001

0

000

0

000

60

007

000

3

000

9

000

2

001

1

000

5

Story drift of B-WSM

0123456789

101112

Floo

r

0123456789

101112

Floo

r

000

1

001

2

001

0

000

4

000

8

001

3

000

0

000

6

000

9

000

5

001

1

000

30

002

000

7

Story drift of ELW

Elastic limit Plastic limitElastic limit Plastic limitElastic limit Plastic limit

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

MPCS-0035 gMPCS-007 gMPCS-014 gMPCS-022 gMPCS-040 gMPCS-062 g

CIPS-0035 gCIPS-007 gCIPS-014 gCIPS-022 gCIPS-040 gCIPS-062 g

(b) Story drift ratio of CIPS and MPCS

Figure 12 Displacement responses of CIPS and MPCS

Acknowledgments

This research has been supported by both Innovation TeamofXirsquoan University of Architecture and Technology and ShaanxiProvince Science and Technology Coordination InnovationProject (nos 2015KTZDSF03-05-01 2015KTZDSF03-04 and2014SZS04-Z01)The support of the National Natural ScienceFoundation of China (Grants nos 51408456 51578444 and51478381) and Ministry of Education Plan for Yangtze RiverScholar and Innovation TeamDevelopment (no IRT13089) isalso acknowledged

References

[1] R Park and T Paulay Reinforced Concrete Structures ohnWileyamp Sons 1975

[2] M Fintel ldquoPerformance of buildings with shear walls in earth-quakes of the last thirty yearsrdquo PCI journal vol 40 no 3 pp62ndash80 1995

[3] A A Yee ldquoSplice sleeve for reinforcing barsrdquo US Pat vol 35401968

[4] J H Ling A B Abd Rahman I S Ibrahim and Z AbdulHamid ldquoBehaviour of grouted pipe splice under incrementaltensile loadrdquo Construction and Building Materials vol 33 no8 pp 90ndash98 2012

[5] M T El-Sheikh R Sause S Pessiki and L-W Lu ldquoSeismicbehavior and design of unbonded post-tensioned precast con-crete framesrdquo PCI journal vol 44 no 3 pp 54ndash71 1999

[6] R S Henry S Sritharan and J M Ingham ldquoUnbonded tendonstresses in post-tensioned concrete walls at nominal flexuralstrengthrdquo ACI Structural Journal vol 109 no 2 pp 171ndash1822012

14 Shock and Vibration

[7] P Seifi R S Henry and JM Ingham ldquoPanel connection detailsin existing New Zealand precast concrete buildingsrdquo Bulletin ofthe New Zealand Society for Earthquake Engineering vol 49 no2 pp 190ndash199 2016

[8] K A Soudki S H Rizkalla and R W Daikiw ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformations part 2 prestressed connectionsrdquo PCI journal vol40 no 5 pp 82ndash96 1995

[9] K A Soudki S H Rizkalla and B LeBlanc ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformationsmdashpart 1 mild steel connectionsrdquo PCI journal vol40 no 4 pp 78ndash96 1995

[10] K A Soudki J S West S H Rizkalla and B BlackettldquoHorizontal connections for precast concrete shear wall panelsunder cyclic shear loadingrdquo PCI journal vol 41 no 3 pp 64ndash80 1996

[11] R Vaghei F Hejazi H Taheri M S Jaafar and A A Ali ldquoEval-uate performance of precast concrete wall to wall connectionrdquoAPCBEE Procedia vol 9 pp 285ndash290 2014

[12] H Jiang Z Chen J Zhang B Wu Y Tian andW Liu ldquoQuasi-static test of precast reinforced concrete shear wall structurerdquoJianzhu Jiegou XuebaoJournal of Building Structures vol 32 no6 pp 34ndash40 2011

[13] Z ChenH Jiang J Zhang BWu Y Tian andW Liu ldquoPseudo-dynamic substructure test on precast reinforced concrete shearwall structurerdquo Jianzhu Jiegou XuebaoJournal of Building Struc-tures vol 32 no 6 pp 41ndash50 2011

[14] J R QianW L Han Z Z Zhao and H Qin ldquoPseudo-dynamicSubstructure Test on a 3-story Full-scaleModel of PrefabricatedConcrete Shear Wall Structure with Rebars Splicing by GroutSleevesrdquo Journal of Building Structures vol 38 no 3 pp 26ndash382017

[15] P Negro D A Bournas and F J Molina ldquoPseudodynamictests on a full-scale 3-storey precast concrete building Globalresponserdquo Engineering Structures vol 57 pp 594ndash608 2013

[16] M Oliva P Gavrilovic and R W Clough ldquoSeismic testing oflarge panel precast walls Comparison of pseudostatic and shak-ing table testsrdquo Earthquake Engineering amp Structural Dynamicsvol 19 no 6 pp 859ndash875 1990

[17] L H Lee W H Yi and S Y Seo ldquoShaking Table Test of PrecastConcrete Wall Structurerdquo in Proceedings of the Eleventh WorldConference on Earthquake Engineering

[18] W Wang A Q Li et al ldquoShaking table test of precast concreteshear wall isolation structurerdquo Journal of Building Structuresvol 36 no 6 pp 35ndash43 2015

[19] EERI Preliminary Reconnaissance Report Earthquake Engi-neering Research Institute 1994

[20] EERI Armenia Earthquake Reconnaissance Report EarthquakeEngineering Research Institute 1989

[21] GB 50011-2010 Code for Seismic Design of Buildings ChinaArchitecture and Building Press Beijing China 2010

[22] JGJ 1-2014 Technical Specification for Precast Concrete Struc-tures China Architecture amp Building Press Beijing China2014

[23] 15J939-1 Example of Architectural Design for PrefabricatedConcrete Structures China Planning Press Beijing China 2015

[24] E Buckingham ldquoOn physically similar systems Illustrations ofthe use of dimensional equationsrdquo Physical Review A AtomicMolecular and Optical Physics vol 4 no 4 pp 345ndash376 1914

[25] Y Zhou andX L LuTheexperimentalmethod and technology ofshaking table test about building structure Science Press BeijingChina 2012

[26] JGJT 101-2015 Specification for Seismic Test of Buildings ChinaArchitecture amp Building Press Beijing China 2014

[27] C Y Shen F L Zhou et al ldquoExperimental research on micro-concrete used in dynamic test modelrdquo Journal of GuangzhouUniversity (Natural Science Edition) vol 4 no 3 pp 250ndash2532005

[28] J Xiao T L Pham and T Ding ldquoShake table test on seismicresponse of a precast frame with recycled aggregate concreterdquoAdvances in Structural Engineering vol 18 no 9 pp 1517ndash15342015

[29] S C Chakrabarti G C Nayak and D K Paul ldquoShear charac-teristics of cast-in-place vertical joints in story-high precast wallassemblyrdquo ACI Structural Journal vol 85 no 1 pp 30ndash45 1988

[30] CWang and J Xiao ldquoStudy of the seismic response of a recycledaggregate concrete frame structurerdquo Earthquake Engineeringand Engineering Vibration vol 12 no 4 pp 669ndash680 2013

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 14: Comparative Study of Seismic Behavior between Monolithic

14 Shock and Vibration

[7] P Seifi R S Henry and JM Ingham ldquoPanel connection detailsin existing New Zealand precast concrete buildingsrdquo Bulletin ofthe New Zealand Society for Earthquake Engineering vol 49 no2 pp 190ndash199 2016

[8] K A Soudki S H Rizkalla and R W Daikiw ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformations part 2 prestressed connectionsrdquo PCI journal vol40 no 5 pp 82ndash96 1995

[9] K A Soudki S H Rizkalla and B LeBlanc ldquoHorizontalconnections for precast concrete shear walls subjected to cyclicdeformationsmdashpart 1 mild steel connectionsrdquo PCI journal vol40 no 4 pp 78ndash96 1995

[10] K A Soudki J S West S H Rizkalla and B BlackettldquoHorizontal connections for precast concrete shear wall panelsunder cyclic shear loadingrdquo PCI journal vol 41 no 3 pp 64ndash80 1996

[11] R Vaghei F Hejazi H Taheri M S Jaafar and A A Ali ldquoEval-uate performance of precast concrete wall to wall connectionrdquoAPCBEE Procedia vol 9 pp 285ndash290 2014

[12] H Jiang Z Chen J Zhang B Wu Y Tian andW Liu ldquoQuasi-static test of precast reinforced concrete shear wall structurerdquoJianzhu Jiegou XuebaoJournal of Building Structures vol 32 no6 pp 34ndash40 2011

[13] Z ChenH Jiang J Zhang BWu Y Tian andW Liu ldquoPseudo-dynamic substructure test on precast reinforced concrete shearwall structurerdquo Jianzhu Jiegou XuebaoJournal of Building Struc-tures vol 32 no 6 pp 41ndash50 2011

[14] J R QianW L Han Z Z Zhao and H Qin ldquoPseudo-dynamicSubstructure Test on a 3-story Full-scaleModel of PrefabricatedConcrete Shear Wall Structure with Rebars Splicing by GroutSleevesrdquo Journal of Building Structures vol 38 no 3 pp 26ndash382017

[15] P Negro D A Bournas and F J Molina ldquoPseudodynamictests on a full-scale 3-storey precast concrete building Globalresponserdquo Engineering Structures vol 57 pp 594ndash608 2013

[16] M Oliva P Gavrilovic and R W Clough ldquoSeismic testing oflarge panel precast walls Comparison of pseudostatic and shak-ing table testsrdquo Earthquake Engineering amp Structural Dynamicsvol 19 no 6 pp 859ndash875 1990

[17] L H Lee W H Yi and S Y Seo ldquoShaking Table Test of PrecastConcrete Wall Structurerdquo in Proceedings of the Eleventh WorldConference on Earthquake Engineering

[18] W Wang A Q Li et al ldquoShaking table test of precast concreteshear wall isolation structurerdquo Journal of Building Structuresvol 36 no 6 pp 35ndash43 2015

[19] EERI Preliminary Reconnaissance Report Earthquake Engi-neering Research Institute 1994

[20] EERI Armenia Earthquake Reconnaissance Report EarthquakeEngineering Research Institute 1989

[21] GB 50011-2010 Code for Seismic Design of Buildings ChinaArchitecture and Building Press Beijing China 2010

[22] JGJ 1-2014 Technical Specification for Precast Concrete Struc-tures China Architecture amp Building Press Beijing China2014

[23] 15J939-1 Example of Architectural Design for PrefabricatedConcrete Structures China Planning Press Beijing China 2015

[24] E Buckingham ldquoOn physically similar systems Illustrations ofthe use of dimensional equationsrdquo Physical Review A AtomicMolecular and Optical Physics vol 4 no 4 pp 345ndash376 1914

[25] Y Zhou andX L LuTheexperimentalmethod and technology ofshaking table test about building structure Science Press BeijingChina 2012

[26] JGJT 101-2015 Specification for Seismic Test of Buildings ChinaArchitecture amp Building Press Beijing China 2014

[27] C Y Shen F L Zhou et al ldquoExperimental research on micro-concrete used in dynamic test modelrdquo Journal of GuangzhouUniversity (Natural Science Edition) vol 4 no 3 pp 250ndash2532005

[28] J Xiao T L Pham and T Ding ldquoShake table test on seismicresponse of a precast frame with recycled aggregate concreterdquoAdvances in Structural Engineering vol 18 no 9 pp 1517ndash15342015

[29] S C Chakrabarti G C Nayak and D K Paul ldquoShear charac-teristics of cast-in-place vertical joints in story-high precast wallassemblyrdquo ACI Structural Journal vol 85 no 1 pp 30ndash45 1988

[30] CWang and J Xiao ldquoStudy of the seismic response of a recycledaggregate concrete frame structurerdquo Earthquake Engineeringand Engineering Vibration vol 12 no 4 pp 669ndash680 2013

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 15: Comparative Study of Seismic Behavior between Monolithic

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom