DESIGN OF MULTI-STOREY BUILDINGS TO …...to serviceability (temperature, creep, shrinkage)...

http://www.iaeme.com/IJCIET/index.asp 881 [email protected]

International Journal of Civil Engineering and Technology (IJCIET)

Volume 9, Issue 9, September 2018, pp. 881–893, Article ID: IJCIET_09_09_084

Available online at http://www.iaeme.com/ijciet/issues.asp?JType=IJCIET&VType=9&IType=9

ISSN Print: 0976-6308 and ISSN Online: 0976-6316

©IAEME Publication Scopus Indexed

DESIGN OF MULTI-STOREY BUILDINGS TO

MITIGATE LOW TEMPERATURES AND

EARTHQUAKE HAZARD

Olga Markogiannaki

Aristotle University of Thessaloniki

Department of Civil Engineering

Konstantinos Psarras



Ioannis Tegos



Emeritus Professor

ABSTRACT

Expansion joints are often used in buildings to divide large floor plans to statically

independent structures. This practice is used to lower the effect of contraction due to

temperature and creep and shrinkage phenomena. When seismic hazard is considered

the required joint widths in buildings become even larger which results in high

construction and maintenance costs. In the last years, the perspective on the

phenomena that induce restraints and the quantitative tools to calculate them have

changed and there have been several research efforts on the development of jointless

structures. It has been observed that stiffness of vertical structural members plays a

key role in the response of buildings under temperature constraints. In contrast to the

beneficial effect on the seismic response, large stiffness of vertical structural members

increases the intensity of temperature effects. Therefore, it is important to develop

effective novel methods to control the stiffness of vertical members apart from

changing cross-section dimensions or ground floor height. The present paper focuses

on a novel method for “controlling” the stiffness of columns of multi-story structures.

A 15-storey building is used as a case study. The aim is to compromise the two

contradictory actions (seismic and thermal). 3D models of the building with and

without the proposed mechanism are developed and the analysis results are

compared. Based on the inter-story drift effects on jointless high–rise structures, it is

illustrated that even large floor plans can be treated with no joints. The results show

the improved response of the jointless structure under extreme temperatures and

effective performance when subjected to seismic hazard.

Design of Multi-Storey Buildings to Mitigate Low Temperatures and Earthquake Hazard


Key words: temperature; seismic response; reinforced concrete buildings; jointless

structure; stiffness control

Cite this Article: Olga Markogiannaki, Konstantinos Psarras, Ioannis Tegos, Design

of Multi-Storey Buildings to Mitigate Low Temperatures and Earthquake Hazard.

International Journal of Civil Engineering and Technology, 9(9), 2018, pp. 881-893.

http://www.iaeme.com/IJCIET/issues.asp?JType=IJCIET&VType=9&IType=9

1. INTRODUCTION

Current building design practice and regulations [1] suggest expansion joints to structures

with large floor plans, as means to avoid increased stresses due to imposed deformations in

the serviceability limit state including temperature and creep and shrinkage phenomena. The

vulnerable members of such buildings are the columns of the ground floor, as the differential

deformations between ground and first floor slab induce additional stresses on these columns.

Moreover, in seismic areas, requirements are further increased due to seismic hazard and the

finally required joint widths are significantly large (i.e. EC8 specifications). A reason for the

wide use of joints has been the idea that is more convenient to let the structures expand rather

than resist the increased demand produced by the imposed deformations (i.e. temperature,

shrinkage) in integral structures. The wide use of joints has also been based partly on the

difficulties of traditional design in analyzing large undetermined structures, as nonlinearities

have to be considered. Furthermore, the presence of joints allows the use of various

construction techniques, i.e. the efficient use of precast members. However, experience of

construction practice has shown that joints in structures arise even more issues than those they

accommodate. The presence of joints undermines durability of structures with problems such

as moisture and corrosion or additional deformations induced by faulty joints. There are also

aesthetic-design issues, since they need double supports at their ends. In addition, their cost

(construction and maintenance) is not negligible to that total lifetime cost of structures [2].

Therefore, joints can be uneconomical design solutions, especially, in seismic prone areas,

where seismic demand increases significantly joint widths due to large seismic relative

displacements.

These concerns have motivated a turn to the construction of jointless, also called integral,

structures to accommodate the problems that arise with the use of joints. Integral structures

have long lifetime without additional maintenance costs. The elimination of joints in

combination with efficient response against thermal and seismic actions can be considered as

a design solution. In the last years, the advancements in technology and in numerical analysis

for undetermined structures has led to successful construction of a large number of jointless

structures, i.e. Kaltern swimming center in Germany, that have reliable behavior against

design loads. Kumar and Vidhya [3] performed a numerical study on the effect of thermal

and seismic actions on a multistory jointless building that showed the significance of seasonal

thermal and seismic actions on buildings. There have also been other publications for jointless

or with limited joints structures [4,5] but they are mainly based on empirical experience.

A critical aspect considered in the design of jointless structures is the accommodation of

the restraints induced by temperature and creep and shrinkage. There have been some

research efforts regarding the response of structures under induced restraints. Hock [6] studied

the origins and size of the imposed deformations on structures showing the importance of

column capacity and deformability under large horizontal slab displacements, while

Wohlfahrt et al [7] tested columns under restrained deformations. Noakowski et al [8] studied

the relation of column stiffness and crack growth to shear forces developed in columns under

imposed restraints. Knowledge on crack growth mechanism [9] and nonlinear response of

reinforced concrete members [10] are necessary to understand the nature of these phenomena.

Olga Markogiannaki, Konstantinos Psarras, Ioannis Tegos


Other researchers, like Caldentey et al [11] and Groli et al [12], have focused on experimental

and analytical investigations on the improvement of the design methods used for columns of

long jointless structures, which have been identified as critical structural members in terms of

serviceability requirements, concluding that current Codes are quite conservative considering

the limit lengths for jointless structures. However, these research studies lack any suggestions

to improve the response of columns subjected to temperature and other phenomena that

induce restraints and to eliminate the intermediate expansion joints.

In the light of the above, a method that contributes to the development of jointless

structures is presented herein. The authors have conducted a preliminary research on this

method in Markogiannaki et al. [13]. The target is to control column stiffness and increase

flexibility which is the dominant factor that affects stresses and deformations on columns by

the imposed temperature and seismic loading. Stiffness is controlled with a novel type of

cross-section for the outer corner columns which has the advantage of flexibility and consists

of a bundle of columns, symmetrically arranged, statically independent, circular cross-

sections in contact with each other. The cross-section has the characteristic of multiple

smaller cross sections replacing a larger solid one. A similar approach has been proposed by

Tegos and Pilitsis [14] for bridge piers to increase bridge flexibility without the use of any

isolation devices. The proposed method is applied in a 15-story building with 80mx20m

layout dimensions and the structural performance against extreme temperature and seismic

hazard is evaluated. The as-built building has small local joints and non-integral connections

between the outer columns and slabs at the ground floor level. Despite the fact that extensive

expansion joints through the building layout were avoided in this design, the solution of a

complete elimination of joints was not explored. The analysis results indicate the improved

structural safety of the fully integral building.

2. DESCRIPTION OF GROUND FLOOR COLUMN GROUP

The alternative column cross-section that is explored in the present paper aims to control

column stiffness and increase column flexibility. The ultimate goal is to eliminate the need of

joints and address both temperature and seismic demand. In typical building structural

systems, the ground floor columns and beams are the most vulnerable structural components

to serviceability (temperature, creep, shrinkage) deformations due to the varying response of

the top and bottom of columns at the ground floor level which results in high drift values at

this floor level. Furthermore, the demand on columns is higher as the column distance from

the center increases. Therefore, the proposed column design approach discussed herein is

applied at the corner columns of the ground floor. The proposed column approach includes

the following design aspects: Smaller circular cross-sections, equal in diameter, replace the

large solid cross section of the perimeter building columns. The schematic representation of

the proposed configuration is illustrated in Fig. 1. In this way the massive solid columns are

fragmentized and have increased flexibility to address deformations imposed by temperature,

creep and shrinkage. Although the system becomes more flexible, if key elements for seismic

response are used, like shear walls, the system behaves effectively against seismic hazard, as

well. The columns are replaced only in the ground floor, while in the upper floors the initial

column cross section is used. The continuity between the lower and upper floor columns is

achieved through the height of the concrete slab between them. Such buildings often have

slabs with large heights which provide adequate space for the appropriate anchorage of the

vertical column reinforcement.



(a) (b)

Figure 1 Schematic plan of the group of columns for (a) rectangular solid cross section, (b) circular

column cross section

The proposed column type is easy to construct, as it is possible to use cardboard

disposable formworks that are widely available and of low cost in current construction

practice and aesthetics are also considered for the final column configuration.

Regarding the structural response of the smaller circular columns buckling issues are

considered. Therefore, a criterion is determined regarding the minimum column dimensions.

The criterion relates the Euler buckling load to the upper bound of the normalized axial force

due to ductility demand. In common building structures, the normalized axial load vd shall be

smaller or equal in absolute values to 0.65 [15]. Eq. (1) and (2) show the derivation of the

formula of the minimum critical diameter of each column cross section with respect to the

upper bound of the axial force.

2

2

2

2 2

2

2

H

0.65 0.65H ( / 4)

4

0.65 H

c ccr

cr c cd

c ck cr ck

c ccr

ck

Pk

Pv

A f k d f

dk f

(1)

(2)

Where, Ec is modulus of elasticity of column, IC column moment of inertia, H is the

column height and k factor depending on fixity conditions , Ac is the column cross section, fck

is the concrete strength, dcr is the column diameter. P-Delta effects are also considered in the

selection of the appropriate column cross sections and column reinforcement is calculated

from critical seismic and non-seismic load combinations.

3. MULTI-STOREY BENCHMARK BUILDING

The building used as the benchmark structure in the case study is a 15 – storey concrete

building with one basement and a plan layout of 79m*19.5m. It is a typical example of

buildings with long dimensions. It consists of a basement, a ground floor that is 4.5m high

and 14 typical floors that are 3.2m high. In Fig. 2 the structural layout of the typical floors is

illustrated. The building is plan symmetric in the Y-direction, while the staircase and elevators

break the symmetry in the X-direction.

According to common practice expansion joints are required for lengths beyond 30-35m.

In particular, in this building due to the presence of the concrete core in the middle a single



joint is not feasible and two expansion joints are required. However, the building was studied

with a limited local joint solution at the column-slab connections to avoid the need of large

joint widths due to seismic demand and the need of double elements at joint locations.

There are circular columns with a diameter equal to 90cm around the perimeter, 4 shear

walls at the perimeter and a central core of shear walls. A characteristic of the system due to

operational needs is the absence of inner columns that result in span lengths equal to 18m and

the absence of beams. The concrete slab (Fig. 2(b)) has a total height of 60cm, is sandwich-

voided filled with polystyrene for lowering the self-weight of the slab and is prestressed due

to the increased demand for controlling deflection, tension stresses and to avoid cracking. The

tendons have parabolic geometry and consist of 5 strands. The foundation is a concrete slab of

1m height and prestressed as well. Further information on dimensions of the structural

elements and their reinforcement detailing can be found in Fig. 3(b-d). The concrete strength

is fck=35MPa and steel strength fyk=500MPa. The tendons are St1670/1860. The building is

located in a moderate seismic hazard region, seismic zone II – 0.24g, and soil class is B [15].

To address the temperature, creep and shrinkage deformations two strategies were

employed: a) the first strategy included the division of the cross – section of the outer

concrete shear walls at the ground floor level with polystyrene to 2 equal smaller cross

sections, as shown in Fig. 3 (a). The initial width of the shear wall is 50cm and is divided to

two 25cm wide cross sections. This separation results to smaller shear wall stiffness in the

critical (longitudinal) direction which lowers the resistance to deformations, as well, without

reducing the stiffness in the transverse direction, Fig. 4. b) the second strategy involved the

introduction of joints, 2cm wide, at the connections of the 4 corner columns of the ground

floor with the 1st floor slab and the extension of the column to slab support with a height of

40cm, as shown in Fig. 3(b).. As the serviceability demand is smaller in the middle of the

structure, the width of local joints is reduced from the corner to the middle accordingly. These

column to slab joints are located at all column positions that exceed 15m distance from the

middle at both sides of the building. The local joints are introduced to avoid the development

of cracking in the slabs. Regarding the seismic response of the building in the benchmark

solution, it should be taken into account that since the introduced joints for temperature and

creep and shrinkage have small width they will close very quickly under seismic excitation.

After the closure of the gap, the columns will be activated and will contribute to the seismic

resistance of the building along with the concrete core and the shear walls. In the present

study, the change of the column stiffness with a group of columns with smaller cross-section

is an attempt to develop a fully integral structure with the elimination of all types (local or

extended) of joints that hasn’t been achieved for the building under study.

(a)



(b) (c)

(d)

(e)

Figure 2 (a) typical floor plan, (b) concrete slab detail, (c) typical column detail, (d) concrete core

(upper part) detail, (e) foundation slab

(a)



(b)

Figure 3 (a) Ground floor corner concrete shear walls (b) Column-Slab joint detail

Figure 4 Location of local joints

4. 3D BUILDING MODELS

For both solutions 3D non-linear models were developed in SAP2000 [16]. The finite element

model of the benchmark building is shown in Fig. 5(a). The columns are modeled as frame

elements with concentrated plasticity. The moment-curvature curves were derived from cross-

section analysis in the free software Any Section [17] based on the dimensions and

reinforcement of the circular solid columns. The concrete shear walls and cores are modeled

with frame elements. Rigid elements are introduced at the top and bottom of the shear walls at

each floor. The floor slab was modeled with equivalent I-shaped beams, beff=1,2m, h=0.6m,

and shell elements in the perimetric full-concrete zones. The tendons were modeled with

tendon elements with the respective dimensions, geometry and values for losses according to

Eurocode 2[1]. The foundation slab was modeled accordingly, with equivalent I-shaped

beams, beff=6m, and shell elements in the perimetric full-concrete zones and the basement

concrete shear walls. The soil was modeled with equivalent linear springs with

ks=20000kN/m. P-Delta effects are taken into account in the analysis. The mass of each floor

is calculated based on the values of the permanent and part of the live loads. The total values

for the characteristic floors are presented in Table 1.



Figure 5 (a) finite element model, (b) frame elements, (c) shell elements

The finite element model of the group of columns (modified) solution has the same

properties with the finite element model of the benchmark building. It only has different

column configuration at the four corners of the building at the ground floor, as shown in

Fig.6, which results in a more flexible system with smaller column stiffness. After a crack

development analysis according to [18] and from Eq. 2, the critical (minimum) column

diameter is derived, dcr=0.30m. It is found that such a diameter can accommodate up to 5cm

of column displacement. Longitudinal reinforcement ratio for the small 30cm diameter

columns is assumed the minimum, ρl=1%, which is verified for all critical load combinations

according to Eurocode 8 [19]. In Fig. 6, the column characteristics and the respective

modeling are presented.

Figure 6 Modified building 3D Model

Table 1 Mass distribution

Floor m(tn)

Roof 2312.33

Typical Floor 2276.40

Ground Floor 2355.91

The input ground motions used for seismic analysis are seven ground acceleration records

matching EC8 spectrum derived from Rexel [20], Fig. 7.



Figure 7 Accelerograms matching code spectrum for seismic zone II

5. ANALYSES AND DISCUSSION OF RESULTS

Comparative analysis was conducted for the two structures, the benchmark building and the

modified- fully integral with the group of columns with smaller cross section. The analyses

included static analysis for temperature deformations and dynamic analysis for the seismic

demand. The target of the analysis was to verify that the imposed deformations are acceptable

in the modified case where joints are avoided and that seismic forces are efficiently received.

Firstly, the response of the benchmark building was investigated against temperature

loading. Regarding seasonal temperature effects, current regulations (i.e. CEN [Comité

Européen de Normalisation] 2003) determine extreme low and high values that are

considered in the design. Low values cause contraction, while higher cause expansion of the

floor slabs. Due to the presence of creep and shrinkage thermal actions that cause expansion

are counterbalanced and are not critical in the analysis. On the contrary, lower temperatures

cause further contraction and their effect is crucial for the investigation of the fully integral

building response. The extreme temperature value considered in the analysis is -25 o

C. An

additional -25oC temperature (Rhodes and Carreira 1997) is used as a simplification to

account for creep and shrinkage. Therefore, equivalent temperature of -50oC was applied to

account for thermal contraction due to extremely low temperatures, creep and shrinkage. The

deformed shape of the benchmark building is shown in a three dimensional view in Fig. 8a.

As shown in Fig. 8b and 8c, the deformations are larger at the corner columns when compared

to the middle of the structure. Furthermore, it is evident that large drift values are developed

only in the ground floor columns due to the varying response of the foundation and the 1st

floor. In the upper floors drift is very small, since both at their top and bottom the same

restraints are induced. It is observed that the temperature analysis results are in accordance

with the application of the proposed system at the ground floor of the building.

The temperature effects are also investigated for the modified-fully integral structural

system with the smaller column cross sections and smaller column stiffness. In Fig. 9, the

comparative drift results for both structures (benchmark and modified) are illustrated. For the

benchmark building, maximum drift value at the corner column is smaller than the width of

the joints at the column to slab connections (1.7cm) which shows that the limited joint

selection is efficient for thermal actions. In the modified structural system, contraction due to

temperature induces cracking due to the integral (monolithic) column to slab connection.

Based on the diagrams provided in Tegos, Giannakas, and Chrysanidis (2011) crack growth is



calculated and crack widths are found lower than 0.2mm which are acceptable according to

the allowable cracking level provided by EC2 [1].

(a)

(b)

(c)

Figure 8 Deformed Shape benchmark structure a) 3D view, b) plan view, c) elevation view



Figure 9 Drift in longitudinal x direction due to temperature effects

The two buildings were also investigated for their response when subjected to seismic

hazard. It is noted that the aim of the analysis under is to verify that the proposed method has

efficient seismic response. The investigation is driven by the concept that temperature

restraints and seismicity are contradictory and often interventions that are beneficial for the

one have negative effects on the other.

In Fig. 10, the seismic drift for each floor for the benchmark and the modified building are

illustrated and compared to drift limits provided by Eurocode 8 - Part1 [15]. The modified

building has smaller column stiffness, therefore it is more flexible. Although there are slightly

larger drifts in the modified building, the resulting values are far lower than the failure

threshold. The modified system affects slightly the response since the core and the shear walls

contribute the most to the building seismic resistance. This is evident in Fig. 11 where the

seismic base shear for the longitudinal and transverse earthquake direction is presented. Base

shear is divided into two categories; shear wall-core, column and it is observed that the

contribution of the shear walls and concrete core in receiving seismic forces is much higher

than that of the columns.

(a) (b)

Figure 10 Seismic Drift for seismic zone II (a) x direction (b) y direction



(a) (b)

Figure 11 Base seismic shear ratio for vertical structural components

6. CONCLUSIONS

The present paper focuses on an alternative column configuration for buildings with long

layouts to avoid the use of expansion joints. Smaller circular cross-sections, equal in diameter,

replace the large solid cross section of the perimeter building columns at the ground floor

level. The presented group of columns has smaller stiffness than the initial solid one resulting

in higher flexibility. Analyses were conducted in a 15 storey reinforced concrete building to

evaluate the structural response against temperature and earthquake hazard. The following

conclusions are derived from the analysis results for the benchmark and the modified system

without expansion joints:

The fully integral structure avoids the introduction of joints, even the limited joints provided

in the studied building case. This elimination of joints addresses issues, like high construction

and maintenance costs.

The study for the imposed restraints due to temperature and creep and shrinkage has shown

that they are accommodated in the modified building case. The top ground floor displacements

are lower than the limit values for cracking development

Although the modified system is more flexible and could affect the seismic response, it is

observed in the analyses results that the modified system’s seismic response is very close to

the benchmark building’s response which is attributed to the fact that the concrete shear walls

and not the columns have the major contribution to the seismic resistance. For both

benchmark and modified systems the seismic drift values are below EC8 limits.

Regarding the aesthetic aspect of the presented solution in the modified system, according to

the authors’ opinion, aesthetics is not harmed and, the column configuration can be used as an

architectural feature of the building.

ACKNOWLEDGMENTS

The authors wish to express their gratitude to METE SYSM S.A. for providing the original

study of the building for the purposes of the present paper.

REFERENCES

[1] CEN [Comité Européen de Normalisation]. Eurocode 2: Design of concrete structures -

Part 1: General rules and rules for buildings. 2002,1(November).

[2] Wanninger, R.; Brinsa C. Kostensenkung durch Modifizierung vorhandener Regelwerke.

Kurzberichte aus der Bauforschung. 2006,47(3), pp.15–20.



[3] Kumar, B. Vidhya K. Numerical Study on Seismic and Temperature Effects in a RCC

Building. International Journal of Engineering Research & Technology. 2014,3(5),

pp.827–31.

[4] Glimm, M.; Quast U. SR. Gemesse-ne Beanspruchungen langer fugenloser ICranbahn-

balken. Bautechnik 80. 2003,7, pp.434–41.

[5] Rapolder H. Das Parkhaus für die Allianz - Arena, München. 2004.

[6] Hock B. Über die Verformungen und Bean-spruchungen von Stahlbetonskelettbauten

infolge von Temperatur- und Feuchtigkeitsänderungen. 1983.

[7] Wohlfahrt, R.; Koch R. Versuche an Stützen mit Normalkraft Zwangsverschiebungen.

DAfStb, Heft 376,. Berlin: DAfStb, Heft 376,Ernst & Sohn; 1986.

[8] Noakowski, P.; Schäfer H 0. Steifigkeits -orientierte Statik im Stahlbetonbau. Berlin:

Ernst & Sohn; 2003.

[9] Carpinteri A. Mechanical damage and crack growth in concrete: Plastic collapse to brittle

fracture. Springer Science & Business Media; 2012.

[10] Chen W-F. Plasticity in Reinforced Concrete [Internet]. J. Ross Publishing; 2007. 474 p.

[11] Caldentey AP, Peiretti HC, Petschke TP, Parrota JE, Soto AG. Serviceability design of

columns of long jointless structures. Engineering Structures. 2012,44, pp.359–71.

[12] Groli G, Pérez Caldentey A, Giraldo Soto A, Marchetto F, Ezeberry Parrotta J. Simplified

serviceability design of jointless structures. Experimental verification and application to

typical bridge and building structures. Engineering Structures. 2014,59, pp.469–83.

[13] Markogiannaki O, Psarras K, Tegos I. Jointless Construction – Optimization of Aseismic

Multi- Storey Buildings. In: 16ECEE. 2018.

[14] Tegos I, Pilitsis V. The Use of Piers of High Flexibility for the Seismic Inertial Forces on

Integral Bridges. In: IBSBI 2014. Athens, Greece; 2014.

[15] CEN[Comité Européen de Normalisation]. Eurocode 8 : Design of structures for

earthquake resistance — Part 1: General rules, seismic actions and rules for buildings.

2003,.

[16] Computers and Structures Inc. SAP2000 Integrated Finite Element Analysis and Design of

Structures. 1998.

[17] Papanikolaou VK. AnySection : Software for the analysis of arbitrary composite sections

in biaxial bending and axial load. Aristotle University of Thessaloniki, Greece; 2015.

[18] Tegos I, Giannakas N, Chrysanidis T. Serviceability cracking check of circular section

piers. Bridge Structures. 2011,7(1), pp.43–52.

[19] CEN [Comité Européen de Normalisation]. Eurocode 8 — Design of Structures for

Earthquake Resistance — Part 2: Bridges. Vol. 3. 2005.

[20] Iervolino I, Galasso C, Cosenza E. REXEL: Computer Aided Record Selection for Code-

Based Seismic Structural Analysis. Bulletin of Earthquake Engineering. 2009,8(2),

pp.339–62.

[21] CEN [Comité Européen de Normalisation]. Eurocode 1: Actions on structures - Part 1-5:

General actions -Thermal actions. 2003.

DESIGN OF MULTI-STOREY BUILDINGS TO …...to serviceability (temperature, creep, shrinkage)...

Documents

Transcript of DESIGN OF MULTI-STOREY BUILDINGS TO …...to serviceability (temperature, creep, shrinkage)...