Available online at www.sciencedirect.

com

ScienceDirect
Available
Available online
online at at www.sciencedirect.com
www.sciencedirect.com
Structural Integrity Procedia 00 (2017) 000–000
www.elsevier.com/locate/procedia
ScienceDirect
ScienceDirect
Procedia Structural
Structural IntegrityIntegrity
Procedia500
(2017) 179–186
(2016) 000–000
www.elsevier.com/locate/procedia

2nd International Conference on Structural Integrity, ICSI 2017, 4-7 September 2017, Funchal,
Madeira, Portugal

Simplified
XV PortugueseAssessment of the PCF
Conference on Fracture, Effects
2016, of Columns
10-12 Shortening
February 2016, on Portugal
Paço de Arcos, the
Response of Tall Concrete Buildings
Thermo-mechanical modeling of a high pressure turbine blade of an
airplane
Ruben Correiagas
a
turbine
, Paulo engine
Silva Lobo b,
*
aQueen’s University of Belfast, University Road, Belfast, BT7 1NN, Northern Ireland
a
a b c
P. Brandão , V. Infante , A.M. Deus *
Creagh Concrete Products ltd, 38 Blackpark Road, Toomebridge, BT41 3SL, Northern Ireland
bDepartamento de Engenharia Civil e Geologia, Universidade da Madeira, Campus Universitário da Penteada, 9000-390, Funchal, Portugal
b
a
Department of Mechanical
CERIS, Engineering,
Instituto Superior Instituto
Técnico, SuperiordeTécnico,
Universidade Lisboa,Universidade de Lisboa,
Av. Rovisco Pais Av. Rovisco
1, 1049-001, Pais,
Lisboa, 1, 1049-001 Lisboa,
Portugal
Portugal
b
IDMEC, Department of Mechanical Engineering, Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais, 1, 1049-001 Lisboa,
Portugal
c
CeFEMA, Department of Mechanical Engineering, Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais, 1, 1049-001 Lisboa,
Abstract Portugal

The constructive process as well as the time-dependent effects must be considered in the assessment of the response of complex
concrete
Abstractstructures. For tall buildings, the adequate prediction of vertical elements shortening is required to determine its effects
on other structural and nonstructural elements, usually overestimated by linear elastic analysis. Thus, simple numerical methods
which make
During it possible
their to consider
operation, modern the mostengine
aircraft relevantcomponents
aspects of theare structural
subjectedbehaviour may bedemanding
to increasingly useful in the early stages
operating of a
conditions,
project. In the
especially theresearch presented
high pressure herein
turbine (HPT) a simplified
blades. Suchmethod, which
conditions considers
cause the viscoelasticity
these parts of concrete
to undergo different types ofastime-dependent
well as the
construction
degradation, sequence, was used.
one of which Its adequacy
is creep. was assessed
A model using the finitebyelement
comparison
methodof (FEM)
the results
was for a tall concrete
developed, building
in order withtothose
to be able predict
the creep
obtained withbehaviour of HPT
a commercial blades.
software Flight
which data records
incorporates (FDR) staged
a nonlinear for a specific aircraft,
construction provided
analysis package.by The
a commercial aviation
good correlation
company,
between were used
the obtained to obtain
results thermal
indicates andsimplified
that the mechanical data for
method usedthree
may different
be appliedflight cycles.
to help makeInappropriate
order to create
designthe 3D model
choices.
© needed
2017 The forAuthors.
the FEM analysis,
Published byaElsevier
HPT bladeB.V. scrap was scanned, and its chemical composition and material properties were
© 2017 The Authors.
obtained. The Published
data that wasbygathered
Elsevier B.V.
was fed into the FEMofmodel and different simulations were run, first with a simplified 3D
Peer-review under responsibility of the Scientific Committee ICSI 2017.
Peer-review
rectangularunder responsibility
block shape, in of the Scientific
order to better Committee of ICSI
establish the 2017
model, and then with the real 3D mesh obtained from the blade scrap. The
overall expected behaviour in terms of displacement was observed, in particular at the trailing edge of the blade. Therefore such a
Keywords: Tall buildings; Concrete; Time-dependent effects; Columns shortening; Constructive process.
model can be useful in the goal of predicting turbine blade life, given a set of FDR data.

© 2016 The Authors. Published by Elsevier B.V.

1. Introduction
Peer-review under responsibility of the Scientific Committee of PCF 2016.

Over theHigh
Keywords: years, buildings
Pressure heightCreep;
Turbine Blade; has Finite
increased toMethod;
Element economize on Simulation.
3D Model; land area, and the construction of reinforced
concrete (RC) high-rise buildings became popular. As a result, problems due to differential axial shortening of vertical

* Corresponding author.
E-mail address: [email protected]

2452-3216 © 2017 The Authors. Published by Elsevier B.V.

Peer-review underauthor.
* Corresponding responsibility
Tel.: +351of218419991.
the Scientific Committee of ICSI 2017.
E-mail address: [email protected]

2452-3216 © 2016 The Authors. Published by Elsevier B.V.

Peer-review under responsibility of the Scientific Committee of PCF 2016.
180 Paulo Silva Lobo et al. / Procedia Structural Integrity 5 (2017) 179–186
2 Correia and Silva Lobo / Structural Integrity Procedia 00 (2017) 000–000

elements have been observed and reported (Moragaspitiya et al., 2009a; Kurc and Lulec, 2011). The total column
shortenings are rarely of practical interest. However, the effects of differential axial shortenings between vertical
elements can lead to excessive deflection and unacceptable crack widths on horizontal elements. The former may
additionally result in damage on nonstructural elements, such as façades, partitions, claddings and mechanical
installations (Fintel et al., 1987). Furthermore, when vertical elements are rigidly connected through slabs or beams,
relative vertical deformations can generate substantial internal forces (Pan et al., 1993; Kim et al., 2010), and
significant redistribution of forces may be required at the ultimate limit states. Thus, the axial deformations of columns,
both elastic and inelastic, require special consideration in the design and construction of tall building structures (Fintel
et al., 1987).
It should be noted that linear elastic analysis often overestimates these effects, thus more complex analyses must
be performed to prevent the exclusion of adequate structural solutions based on inadequate analysis. Several methods
were developed to quantify the magnitude of the mentioned effects, making it possible to define competent design
provisions and procedures (Moragaspitiya et al., 2010). The finite element method, combined with step-by-step
integration methods, for example, is a reliable solution for the prediction of the time-dependent deformations.
However, these procedures are often complicated and time-consuming, or do not capture the complexity of the
problem, because they are limited to a few analysis parameters (Au et al., 2007; Huang et al., 2007; Moragaspitiya et
al., 2009b; Kurc and Lulec, 2011). For early design stages, it is useful to use simple numerical methods, adequate for
an engineering practice environment, which consider both the constructive process and the time-dependent response
of concrete.
A simplified method to predict the internal forces due to axial shortening of columns was used in the research
presented herein, making it possible to consider the more relevant parameters in the response assessment of concrete
structures. The study reported now is intended to assess the adequacy of the simplified method used through the
comparison of the results for a RC tall building with those obtained with SAP2000, using its nonlinear staged
construction analysis package.

2. Axial shortening of concrete columns

The construction and loading sequence may be relevant for the analysis of complex structures, namely regarding
the axial deformation of columns (CTBUH, 1980; Smith and Coull, 1991). During construction, dead loads are applied
step-by-step. As indicated in Fig. 1, at the time of construction of a given level N there are N-1 previous load stages
due to slabs concrete pouring, to which adds all subsequent load stages, owing to other installations, such as cladding
and partitions, that occurred up to the time of construction of level N. Because each column segment undergoes elastic
and inelastic deformations due to the different loads, axial shortening of columns is directly related to the construction
sequence and its pace. At the design stage, an assumption of the loading history, as realistically as possible, is required
to assess the differential shortening consequences.

Fig. 1. Schematic of a multi-story building.

 Paulo Silva Lobo et al. / Procedia Structural Integrity 5 (2017) 179–186 181
Correia and Silva Lobo / Structural Integrity Procedia 00 (2017) 000–000 3

In the case of cast-in-situ RC structures, the deformations that occur prior to installing a slab are corrected through
the levelling of the formworks of the horizontal components of the floor (Fintel et al., 1987). If accurate estimates of
the vertical displacements are available, it is even possible to prescribe further corrective measures, eliminating long-
term deformations. Nonetheless, it is not possible to avoid the structural effects of the resulting differential
deformations.
The variation of the elasticity modulus with time may be of primary importance when time-dependent behaviour
of concrete is considered. The ageing of the material is considered through the use of the concrete strength at the time
of loading rather than the usual 28 days. The time-dependent Young’s modulus used for determining axial shortenings
may be obtained by the equation presented in NP EN 1992-1-1:2010, given by

0.3
 f (t ) 
Ecm (t )   cm   Ecm (1)
 f cm 

where Ecm and fcm are the Young’s Modulus and the compressive strength, respectively, at the age of 28 days. The
tangent elasticity modulus Ec(t) is obtained by multiplying Ecm(t) by 1.05.
The loads are applied on the vertical elements at the loading stages, and the columns shortening results from both
elastic and time-dependent deformations. Assuming that the axial stress of a column segment varies in small
increments due to the loading cycles, the total axial strain at any instant t, due to a load applied at ti, may be obtained
by the sum of all strains which occur during that time interval, given by (Ghali et al., 2002)

 c (t )   ce (t i )   cc (t , t i )   cs (t , t i ) (2)

in which εce(ti) is the elastic strain due to the load applied at time ti. εcc(t,ti) and εcs(t,ti) represent the creep and shrinkage
deformations, respectively. If the stress is constant over the time period from ti to t, the total deformation εc(t) is caused
by two components: the shrinkage that takes place during the considered time period; the stress applied at ti, σc(ti), in
which case its contribution is a function of the Young’s modulus and of the creep coefficient, φ(t,ti), which may be
determined according to NP EN 1992-1-1:2010. Thus, the Equation in (2) can be rewritten as (Bažant, 1982; Ghali et
al., 2002)

 1  (t , ti ) 
 c (t )   c (ti )       cs (t , ti )
 (3)
 Ec (ti ) Ec (28) 

In this study, the columns shortening which occur before and after the casting of the slab of each level were
calculated separately. This procedure made it possible to determine the deformations that are eliminated during the
construction process and, therefore, do not contribute to the final displacements. Considering a segment j of a vertical
element, the corresponding total shortening up to the construction of a level N has N-j loading cycles applied at instants
ti, with tj equal to the time instant of the initial loading of the segment being analysed - the construction of the slab at
level j. The mentioned shortening may be determined by the superposition of the deformations of each loading cycle.
Thus, using the Equation in (3), the corresponding strain may be given, in a simplified manner, as

N 1  1  (t N  t j , ti )  
 j (t N )    (ti )        cs (t N  t j , t j ) (4)
 Ec (28)  
i j   Ec (ti )

Thus, for a given column, the total shortening which occurs before the construction of level N, eliminated by the
construction process, may be obtained by
182 Paulo Silva Lobo et al. / Procedia Structural Integrity 5 (2017) 179–186
4 Correia and Silva Lobo / Structural Integrity Procedia 00 (2017) 000–000

N 1
 c , N    j (t N )  h j (5)
j 1

where hj is the height of the segment j of the vertical element.

The long-term shortenings of each segment of a column were determined in a similar manner to what was
considered for the construction shortenings. In this case, it is necessary to account for all the deformations that occur
up to an instant T after the construction of the structure, with n loading stages. Thus, the long-term strains are given
by

n   1  (T  t j , t i )  
 j (T )    (t i )        cs (T  t j , t j ) (6)
 E (t ) E ( 28)
i 1   c i c 

The total long-term shortening of a vertical element is then computed in a similar manner to what is presented in the
Equation in (5), resulting

N
 T , N    j (T )  h j (7)
j 1

The final values of the columns shortening of a level N at time instant T, δf,N, are determined by the difference
between the total vertical displacement of the column at the analysed level and the displacement eliminated by the
constructive process, and is written as

 f , N   T , N   c, N (8)

The simplified method adopted uses the columns displacements obtained as described above to predict the shear
and bending moment diagrams of the horizontal elements of each floor. These internal forces are then computed
assuming that a differential shortening has the same effect on a horizontal element as a differential settlement of
supports. Usually, on a concrete building, the differential shortenings are more substantial between columns and shear
walls, given the more pronounced difference of mean axial strains. Erro! A origem da referência não foi encontrada.
shows the simple beam model adopted in this study as well as the corresponding internal forces due to the differential
settlement of the supports. The spring represents the flexural stiffness of the column at the beam-column joint.

a b
Bending moment

Shear

Fig. 2. (a) deformations due to the axial shortening of columns; (b) corresponding bending moment and shear diagrams.

On the determination of the internal forces of the beams, both age-adjusted and effective elasticity moduli were
used. The former was used to estimate the stiffness values when assessing the effects of differential shortenings
resulting from the loading cycles up to the construction of the level of the beam under analysis. The latter elasticity
modulus was used in the assessment regarding the effects of the loads applied after the construction of the level under
 Paulo Silva Lobo et al. / Procedia Structural Integrity 5 (2017) 179–186 183
Correia and Silva Lobo / Structural Integrity Procedia 00 (2017) 000–000 5

analysis. For a load introduced at ti, the effective and age-adjusted Young’s modulus at an age t may be estimated per
the equations presented in CEB-FIP Model Code 90, given by

E c (t i )
E c,eff (t , t i )  (9)
E c (t i )
1   (t , t i )
E c (28)

and

E c (t i )
E c,adj (t , t i )  (10)
E c (t i )
1  .   (t , t i )
E c (28)

In this study, the ageing coefficient, χ, was considered equal to 0.8.

3. Case study

A multi-storey building with 45 floors and total height of 135 m was used to assess the adequacy of the methodology
described above. The building has a typical high-rise structural beam-column system, with a central core and a simple
symmetrical plan, as shown in Fig. 3. The chosen building made it possible to perform a straightforward analysis of
the results and to focus on the effects of differential shortening of vertical elements.

6m 6m 6m
6m
6m 6m

Fig. 3. 3D perspective, plan and elevation of the building analyzed.

The horizontal and vertical members are constructed with C30/37 and C60/70 concrete, respectively. The interior
beams, which connect the core shear walls to the peripheral frame, have a 0.30×0.50 m2 cross-section, while the
exterior beams have 0.40×0.70 m2 cross-section. Regarding the vertical elements, a continuous cross-section area of
1.00×0.40 m2 was adopted for the columns, whilst the central core has 0.40 m thick concrete walls.
The stress values due to the axial deformations of the vertical supports obtained with the simplified method adopted
were compared with the results of a 3D analysis using SAP2000 Ultimate v.15.0. This software makes it possible to
consider the construction process and the time-dependent behaviour of the material, using a specific package called
“Staged Construction”, based on the input of the material properties which affect ageing, creep, and shrinkage
parameters. These properties were computed per the constitutive laws recommended in CEB-FIP Model Code 90.
184 Paulo Silva Lobo et al. / Procedia Structural Integrity 5 (2017) 179–186
6 Correia and Silva Lobo / Structural Integrity Procedia 00 (2017) 000–000

On both nonlinear analyses, floor construction and loading cycles of seven days were considered. The total
permanent loads were considered to be imposed 14 days after the floor self-weight was applied. Moreover, the
long-term results were determined for 10 and 30 years after the construction of the building.

4. Results

The validity of the simplified method used was assessed by comparison of its results with the values obtained with
SAP2000. The study focused the differential axial shortening between the central core and the exterior column P1 (see
Fig. 3) as well as the internal forces of the interior beams resulting from those displacements. If the structure presents
adequate ductility, the problem here evaluated, regarding both internal forces and displacements, is a matter of
serviceability limit states verification, thus service-level loads combination was considered in the assessment presented
below.
The total columns shortening values at every five levels are shown in Fig. 4.
a b
7,0 7,0

6,0 6,0

Total shortening [cm]

Total shortening [cm]

5,0 5,0

4,0 4,0

3,0 3,0

2,0 2,0

1,0 1,0

0,0 0,0
1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9
Level Level

Column P1 (10 years) Column P1 (30 years) Wall (10 years) Wall (30 years)

Fig. 4. (a) simplified method; (b) “Staged Construction”.

The similarities in the results of the simplified method and of the “Staged Construction” analyses can be easily
observed. Also similar is the elements shortening tendency along the height of the building obtained with both
analyses. The axial displacements increase to about two-thirds of the height of the building, after which it decreases.
As could have been expected, much of the deformations take place in the first 10 years after construction. Maximum
total shortening values on both assessments occur at about levels 30 and 35. According to the “Staged Construction”
analysis, after 30 years, column P1 undertakes the highest shortening value at level 33, with 6.3 cm, while on the
concrete shear wall the maximum displacement of 4.5 cm is obtained at level 38. Using the simplified method, those
values are of 5.9 cm at level 32 and of 4.3 cm at level 36.
Furthermore, a linear elastic analysis was performed to compare its results with those of the more realistic methods,
which consider the time-dependent effects and the construction sequence. As noted above, this method often
overestimates the internal forces of structural elements as well as the displacements of complex structures such as
those of tall buildings. The model used is the same of the “Staged Construction” analysis, but without the consideration
of the construction sequence and with linear elastic constitutive laws for the materials. The long-term values of
displacements were obtained by multiplication of the elastic results by 2.5, which corresponds to a value of φ equal to
1.5, considered adequate for the concrete adopted.
A comparison of the differential displacements between the column P1 and the shear wall obtained with the three
applied analyses are presented in Fig. 5.
 Correia and Silva
Paulo Silva Lobo
Lobo et
/ Structural Integrity
al. / Procedia Procedia
Structural 00 (2017)
Integrity 000–000
5 (2017) 179–186 1857

3,0

2,5

Δ shortening [cm]
2,0

1,5

1,0

0,5

0,0
5 10 15 20 25 30 35 40 45
Level

Simplified method "Staged Construction" Linear elastic analysis

Fig. 5. Differential shortening values at 30 years after the construction of the building.

The maximum values of the differential shortening occur at approximately mid-height of the building, according
to both the “Staged Construction” and the simplified method, whilst via the linear elastic analysis the maximum value
is registered at the top of the building, at level 45. The results obtained with the latter analysis are continuously
increasing from the bottom to the top of the structure. Once more, both time-dependent methods show similar values,
but some inconsistency was obtained at the top levels, where it is noticeable a decrease at a higher rate when using the
simplified method. Nevertheless, the maximum values obtained with both formulations are very similar. In contrast,
the elastic analysis provided very unrealistic differential shortening values, assuming that the “Staged Construction”
method delivers the most precise results.
The shear and bending moment values of the interior beams due to the long-term differential deformations are
shown for every 10 levels in Table 1Erro! A origem da referência não foi encontrada.. The beams are identified
with the letter “B” followed by the number of the floor level they belong to.

Table 1. Internal forces of beams due to the differential shortening of vertical elements.
“Staged Construction” analysis Simplified method Elastic analysis
Shear [kN] Moment [kN·m] Shear [kN] Moment [kN·m] Shear [kN] Moment [kN·m]
B10 23.1 -67.5 23.0 -70.0 28.2 -80.0
64.0 67.5 74.6
B20 35.5 -104.5 34.4 -105.3 47.5 -136.0
98.8 101.1 126.0
B30 38.4 -113.0 35.1 -106.4 59.2 -168.8
106.3 102.0 156.2
B40 31.5 -92.1 23.6 -72.4 64.7 -184.5
86.5 69.6 17.6

The values obtained with the “Staged Construction” method and with the simplified method are very similar.
However, following the differences in the differential shortenings, the weaker correlation is found in the top floors of
the building. Nonetheless, the final internal forces obtained with both methods correlate well, especially for the
maximum values at approximately mid-height of the building. This indicates that the simplified method could be used
to help structural designers make adequate choices at the early stages of a project.
The shear and bending moment values obtained using the linear elastic analysis are significantly different from the
results of the other two methods, especially for the higher floors, following, as expected, the differences observed on
186 Paulo Silva Lobo et al. / Procedia Structural Integrity 5 (2017) 179–186
8 Correia and Silva Lobo / Structural Integrity Procedia 00 (2017) 000–000

the differential shortening values. Noteworthy are the high shear and bending moment values of B40 when using the
linear elastic analysis, which are approximately twice the values obtained with the other two analyses.

5. Conclusions

Tall buildings are complex structures that need careful design in all aspects, one of which is the consideration of
the constructive process. Also, the time-dependent response of concrete must be considered in the structural analysis.
A RC tall building with 45 floors and a structural system composed of a central core connected to a peripheral
frame was considered. A simplified method for the assessment of the effects mentioned above on the response of tall
concrete buildings was used. The results obtained were compared with the ones of a commercial software of structural
analysis which incorporates a nonlinear “Staged Construction” analysis package. The total and differential shortening
values obtained with both analysis are similar, with the largest difference occurring in the differential column/wall
shortening values at top floors of the building. The shear and bending moment values due to the long-term differential
shortenings computed with the nonlinear analyses are similar. Although smaller internal forces were determined for
the interior beams of the top floors with the simplified method, the maximum values are identical and occur in the
floors above the mid-height of the building, around level 30. These results were compared with those of a linear elastic
analysis, which significantly overestimated the vertical displacements of the building as well as the resulting internal
forces in the horizontal elements.
The comparison of the long-term results of the simplified method with those of the “Staged Construction” analysis
indicates that the former may be useful in the early stages of the project of a tall building. It constitutes a simple method
to estimate the long-term differential displacements of vertical elements as well as the resulting effects on the response
of structures.

References

Au, F. T. K., Liu, C. H., Lee, P. K. K., 2007. Creep and Shrinkage Analysis of Reinforced Concrete Frames by History-Adjusted and
Shrinkage-Adjusted Elasticity Moduli. The Structural Design of Tall and Special Buildings 18(1), 13–35.
Bažant, Z. P., 1982. Mathematical Models for Creep and Shrinkage of Concrete, in “Creep and Shrinkage in Concrete Structures”. In: Bažant, Z.
P., Wittmann, F. H. (Ed.). John Wiley & Sons, Chichester, New York, Brisbane, Toronto and Singapore, pp. 93.
Comité Euro-Internacional du Béton - Fédération Internationale de la Précontrainte, 1993. CEB-FIP Model Code 90 (MC-90). Thomas Telford,
London, pp. 460.
CTBUH, 1980. Tall Buildings Criteria and Loading, in "Monograph on Planning and Design of Tall Buildings". In: Robertson, L., Naka, T. (Ed.).
American Society of Civil Engineers, New York.
Fintel, M., Ghosh, S. K., Iyengar, H., 1987. “Column Shortening in Tall Structures: Prediction and Compensation”. Portland Cement Association,
pp. 35.
Ghali, A., Favre, R., Elbadry, M., 2002. “Concrete Structures: Stresses and Deformations”. Spon Press, London and New York, pp. 584.
Huang, T.-T, Stewart, R., Doh, J.-H, Song, D., 2007. Risk distribution profile for differential column shortening using a possibility theory approach,
Fourth International Conference on Construction in the 21 st Century (CITC-IV): “Accelerating Innovation in Engineering, Management and
Technology”. Australia, pp. 709–716.
Kim, H.-S., Jeong, S.-H., Shin, S.-H., Park, J.-P., 2010. Simplified Column Shortening Analysis of a Multi-Storey Reinforced Concrete Frame. The
Structural Design of Tall and Special Buildings 21(6), 405–415.
Kurc, O., Lulec, A., 2011. A Comparative Study on Different Analysis Approaches for Estimating the Axial Loads on Columns and Structural
Walls at Tall Buildings. The Structural Design of Tall and Special Buildings 22(6), 485–499.
Moragaspitiya, P., Thambiratnam, D., Perera, N., Chan, T., 2009a. Differential axial shortening of concrete structures, 2nd Infrastructure Theme
Postgraduate Conference. Queensland University of Technology, Brisbane, 1–11.
Moragaspitiya, P., Thambiratnam, D., Perera, N. Chan, T., 2009b. Axial shortening in reinforced concretes members using vibration characteristics
- Part 1 - Theory, 3rd Smart Systems Postgraduate Student Conference. Queensland University of Technology, Brisbane, pp. 126–131.
Moragaspitiya, P., Thambiratnam, D., Perera, N., Chan, T., 2010. A Numerical Method to Quantify Differential Axial Shortening in Concrete
Buildings. Engineering Structures 32(8), 2310–2317.
NP EN 1992-1-1:2010. Eurocode 2 – Design of Concrete Structures Part 1-1: General Rules and Rules for Buildings. CEN, Brussels.
Pan, L. B., Liu, P. C., Bakoss, S. L., 1993. Long‐Term Shortening of Concrete Columns in Tall Buildings. Journal of Structural Engineering 119(7),
pp. 2258-2262.
Smith, B. S., Coull, A., 1991. “Tall Building Structures: Analysis and Design”. John Willey & Sons, Inc., New York, pp. 552.