Steel and Composite Structures, Vol. 20, No.

2 (2016) 000-000
DOI: http://dx.doi.org/10.12989/scs.2016.20.1.000

Structural behavior of non-symmetrical

steel cable-stayed bridges
Juan J. Jorquera-Lucerga 1, Jose A. Lozano-Galant 2 and Jose Turmo 3
1
Department of Civil Engineering, Polytechnic University of Cartagena,
Paseo Alfonso XIII 52, 30203, Cartagena, Spain
2
Department of Civil Engineering, University of Castilla-La Mancha,
Av. Camilo Jose Cela SN 13071, Ciudad Real, Spain
3
Department of Construction Engineering, Universitat Politècnica de Catalunya,
Barcelonatech, C/Jordi Girona 1-3, Edif B1, 08034 Barcelona, Spain)

(Received April 22, 2015, Revised July 14, 2015, Accepted October 20, 2015)

Abstract. Despite of the growing number of built examples, the analysis of non-symmetrical cable stayed bridges
has not received considerable attention from the researchers. In fact, the effects of the main design parameters in the
structural behavior of these bridges are not addressed in detail in the literature. To fill this gap, this paper studies the
structural response of a number of non-symmetrical cable-stayed bridges. With this aim, a parametric analysis is
performed to evaluate the effect of each of the main design parameters (the ratio between the main and the back span
length, the pylon, the deck and backstay stiffnesses, the pylon inclination, and the stay configuration) of this kind of
bridges. Furthermore, the role of the geometrical nonlinearity and the steel consumption in stays are evaluated.
Keywords: non-symmetrical cable-stayed bridges; parametric study; design parameters; steel bridge; design
guidelines

1. Introduction

The rapid bloom of cable-stayed bridges has been propitiated by the development of erection
techniques and improvements of construction materials (Svensson 2012). According to many
authors (see Agrawal 1997), this bridge type is the most economical one for bridge spans ranging
between 150 and 360 m. A major concern in the structural efficiency of a cable-stayed bridge and,
therefore, in its cost, lies in the symmetry of its longitudinal configuration, i.e., when the side span
is roughly the 40% of the main span for a three span bridge. For this reason, structurally speaking,
designers prefer symmetric designs rather than non-symmetric ones, which are limited to those
cases where the symmetric solution is discouraged for site conditions. Nevertheless, economic
boom of the last decade has deflected attention from economic cost to iconic aesthetical
appearance increasing significantly designer‟s freedom (Jorquera-Lucerga 2013). This structural
freedom has enabled the proliferation of a number of non-symmetric cable-stayed bridges that
might even go without basic structural elements such as backstays (as the Alamillo Bridge in Spain,
Casas and Aparicio 1998).

Corresponding author, Ph.D., E-mail: juanjo.jorquera@upct.es

Copyright © 2016 Techno-Press, Ltd.

http://www.techno-press.org/?journal=scs&subpage=6 ISSN: 1229-9367 (Print), 1598-6233 (Online)
 Juan J. Jorquera-Lucerga, Jose A. Lozano-Galant and Jose Turmo

In cable-stayed bridges, all the resistant mechanisms are related, to a greater or lesser extent, to
the tensile forces of the stay cables in service. Since current cable-stayed bridges include a highly
indeterminate structural system, no unique solution for pre-tensioning stay cable forces exists
(Marchetti and Lecinq 1999). In fact, these represent a design parameter that can be tailored to
achieve an effective design for the bridge (such as full use of material strength, achievement of
target geometry or minimization of structural cost).
The stay cable forces are usually defined to achieve a certain designer‟s criterion in service, in a
stage known as the Objective Service Stage or OSS (see Lozano-Galant et al. 2012a, b). The
criteria proposed in the literature to estimate the stay forces in the OSS are reviewed by many
authors (see Guan 2000, Chen et al. 2000, Hassan et al. 2012, Lozano-Galant et al. 2013).
According to them, some of the most important criteria are as follows: (1) The Pendulus Rule (see
Virlogeux 1994, SETRA 2001). This is a simplified method to carry out a first approximation of
the stay cable forces in nearly horizontal decks. This criterion assumes that the tension of each stay
cable is close to the tension obtained when the deck is hinged at each stay anchorage. In this way,
the cable forces can be estimated as the projection in the stay direction of the averaged load
introduced into the two adjacent deck segments. (2) Minimization of the Creep Effect Criterion
(see Scotti 2003). This criterion defines the stay cable forces in the OSS to minimize the creep
effect in concrete structures with vertical stay cables. In order to analyze the creep effects, this
method applies the age-adjusted effective modulus method (Bazant 1972). The calculation of the
stay forces is performed by solving a system of equations that relates the deflections of the deck
when unitary forces are applied into the stay cables and when the permanent load is applied into
the structure. (3) Minimal Bending Energy Criterion (Du 1989). This criterion is based on the
minimization of the bending energy of the structure. Additionally, this criterion can be used to
minimize the effects of the time-dependent phenomena Lozano-Galant and Turmo (2014) and the
staggered erection of the superstructure Lozano-Galant et al. (2014, 2015). (4) Rigidly Supported
Continuous Beam Criterion (Manterola et al. 2006, Chen et al. 2000, Gimsing 1997). This
criterion assumes that the long-term behavior of a cable-stayed bridge corresponds with that of a
fictitious rigidly continuous beam. This beam is defined by removing the stay cables and adding
fictitious bearings at the bridge deck anchorages. In this way, the stay cable forces can be obtained
by projecting the vertical reactions of the corresponding fictitious supports into the stay cable
direction. (5) Zero Displacement Criterion (Lazar et al. 1972, Wang et al. 1993). This criterion
defines the stay cable forces to achieve zero deflections at certain control points of the structure. A
common criterion consists of defining zero vertical deflection at the deck-stay connection and zero
horizontal deflection at the top of the pylon in structures with backstays. To achieve this goal
(Wang et al. 1993) proposed the use of an iterative process in which the prestressing stay cable
forces are successively updated to minimize the deflections at the control points. Some other
methods such as the B-spline curve criterion (Hassam et al. 2012, Hassam 2013) are also based, to
a great extent, on the analysis of the bridge deflections. The Minimization of the Sum of Squares
Method criterion defines the stay cable forces by an optimization analysis in which the sum of the
squares of the vertical deflections along the bridge deck is minimized. On the other hand, the B-
spline curves criterion proposes the optimization, by mean of genetic algorithms, of the post-
tensioning functions to achieve minimum deck deflections. These post-tensioning functions are
defined by B-spline curves. (6) The Unit Load Method (Janjic et al. 2002, 2003). This criterion is
based on a linear system of equations that includes a degree of freedom for each stay cable force.
This system relates the bending moments at some control points for two types of load cases: unit
prestressing loads at each stay cable and the target load of the superstructure. This system of
 Structural behavior of non-symmetrical steel cable-stayed bridges

equations can be directly solved by mean of an influence matrix after defining the desired moment
distribution to be achieved. A similar approach was used by Guan in terms of deflections. (7)
Optimization Criterion (Baldomir et al. 2010, Negrao and Simoes 1997, Simoes and Negrao 2000).
In this criterion, the prestressing stay cable forces are defined by the minimization of a scalar
objective function. Different tendencies are used to define these objective functions. Some
designers base their objective functions on the structural efficiency while some others base it on
the economy of the structure. To reduce the computation time of the optimization process, Lute et
al. (2009) proposed the use of genetic algorithms.
All criteria presented above might be applied to symmetric cable-stayed structures.
Nevertheless, not all of them are adequate to non-symmetric ones. This is the case of the Pendulus
rule and the Rigidly Continuous Beam Criterion as they provide no information of the backstay
forces. For this reason, sometimes the designers have to look for alternative methods. This was the
case of the design of the Alamillo Bridge (Casas and Aparicio 1998). In this structure the stay
cables forces in the main span were defined to correspond with those of an equivalent continuous
beam. Nevertheless, the weight of the inclined pylon was defined to assure that the resultant of the
stay forces remained inside the pylon. The lack of studies referring to non-symmetrical cable-
stayed bridges is not limited to the backstay forces as the effects of the main design parameters
(such as the ratio between the main and the back span length, the pylon and the deck stiffnesses,
the pylon inclination or the stay configuration) are not conveniently addressed neither.
To fill the detected gaps, this paper studies the structural behavior of highly non-symmetrical
two span cable-stayed bridges made of steel. With this aim, a number of parametric analyses
focused on each of the main design parameters (ratio between the main and the back span length,
the pylon and the deck stiffnesses, the pylon inclination of the stay configuration) are presented.
Furthermore, the effect of the geometrical nonlinearity in these structures is not significant.
This paper is organized as follows: In Section 2, the layout and the load cases in the case of
study are presented. In Section 3, the calculation and the assumptions of the study are presented. In
Section 4, the results of the parametric analyses are discussed. Finally, a set of conclusions are
drawn in Section 5.

2. Cases of study

The cases of study consist of a cable-stayed bridge whose geometry is modified to analyze the
effects of different non-symmetrical configurations. In this section, the configurations and load
cases of the analyzed non-symmetric bridges are presented.

2.1 Configurations

The main parameters that define the geometry of a two span non-symmetric cable-stayed
bridge may be defined as follows: (1) Back span, L1, (2) Main span, L2, (3) Pylon height over the
deck, H1, (4) Pylon height below the deck, H2, (5) Pylon inclination, α, (6) Backstay inclination, α1,
(7) Lower main span stay inclination, α2, (8) Cable arrangement, in which the stay cables are
distributed throughout the pylon. The studied arrangements are Harp (parallel cables) and Semi-
Fan (cables distributed throughout a certain length h, near the top of the pylon), (9) Anchorage
separation in the pylon back span, h1, (10) Anchorage separation in the pylon main span, h2, (11)
Number of stay cables in the back span, n1, (12) Number of stay cables in the main span, n2, (13)
 Juan J. Jorquera-Lucerga, Jose A. Lozano-Galant and Jose Turmo

Fig. 1 Main parameters of a non-symmetrical cable-stayed bridge with different stay configurations

Separation of the backstay, s0, (14) Anchorage separation in the deck back span, s1, (15)
Anchorage separation in the deck main span, s2. (16) Separation of the deck last stay, s3. All these
parameters are summarized in Fig. 1.
The assumptions considered in this study might be summarized as follows: (1) Pylon-deck
connection: As considered in many of actual bridges, a configuration pioneered by Severins
Bridge (1960) over the River Rhine in Cologne: see e.g., Zárate-Brazo Largo cable-stayed bridge
(Svensson 2012) no pylon-deck connection is considered. (2) Valid range of stay inclinations: To
assure an adequate efficiency of the backstays, α1 and α2 must remain into a certain range of
validity. In this study, this range has been assumed between 30° and 60°, although smaller values
outside this range could also be considered (e.g., Menn (1990) considered minimum stay
inclinations of 22°). (3) Fixed parameters in non-symmetrical configurations: To homogenize the
characteristics of the analyzed bridges, all the examples share the following fixed parameters L2 =
100 m, s0 = s1 = s2 = 6 m, s3 = 7 m, H2 = 15 m and n2 = 15. (4) Parameters defining the non-
symmetrical configuration: The non-symmetrical configurations are defined in terms of the fixed
parameters and fixing α, α1 and h. Firstly, the pylon top can be obtained by intersecting the last
main span stay and the pylon. In this way, H2 can be calculated in terms of α1 and α as presented in
Eq. (1). This height can be used to calculate L1 as presented in Eq. (2). In order to satisfy s0 = 6 m
and S1 = 6 m with an integer number of stay cables, the admissible L1 are restricted to 10 m, 21 m,
33 m, 51 m, 63 m, 75 m, 81 m and 99 m. For each of these values of L1, n1 can be estimated from
Eq. (3). Finally, h1 can be obtained in terms of h and n1 as presented in Eq. (4). The calculation of
h2 can be carried out by substituting n1 to n2 in Eq. (4). (5) The effects of the staggered erection of
the superstructure are neglected. (6) Steel relaxation is neglected. The simulation of this
phenomena can be found in Gunaydin et al. (2014) and Atmaca and Ates (2012).
tan 𝛼1 ·𝐿2
𝐻2 = (1)
1 − tan 𝛼 ·tan 𝛼1
1
𝐿1 = 𝐻2 tan 𝛼 + < 𝐿2 (2)
tan 𝛼1
 Structural behavior of non-symmetrical steel cable-stayed bridges

𝐿 − 𝑆3
𝑁2 = + 1 − 𝑁1 (3)
𝑆1
ℎ
ℎ1 = (4)
𝑁1
The valid cases that satisfy stay cables efficiency condition for L1=100m are located into the
shadowed region of Fig. 2(A) in terms of L1, α and α1. To reduce the number of bridge geometries
available for the parametric analysis only some α (-20°, -10°, 0°, 10° and 20°) are studied, where a
negative angle represents a pylon leaning backwards and a positive angle represents a pylon
leaning forward the main span. The agreement between the studied α and the admissible L1 results
in the 23 analyzed cases summarized in Fig. 2(A). Each of these geometries is analyzed with the
two alternative pylon and deck sections presented in Fig. 2(B). Differences between both sections
(Stiff and Flexible) refer to their stiffness. The sections stiff and flexible sections in the steel pylon
are 4000×2000 and 2000×2000 mm with 30mm thickness, respectively. In the case of the deck, the
Stiff and Flexible sections are 6000×2000 and 10000×500 mm with 30 mm thickness. It is
remarked that all analyzed pylons and decks are assumed to be made of steel. Therefore, time-
dependent phenomena that occur in concrete bridges are not considered. The mechanical

Fig. 2 Cases analyzed in the parametric study. (a) Valid and analyzed cases in the parametric study; and
(b) Analyzed pylon and deck cross-sections
 Juan J. Jorquera-Lucerga, Jose A. Lozano-Galant and Jose Turmo

Table 1 Mechanical properties of the structural elements

Area (m2) Inertia (m4) E (GPa)
Flexible pylon 0,236 0,153 210
Stiff pylon 0,356 0,779 210
Flexible deck 0,476 0,034 210
Stiff deck 0,626 0,386 210
Stay * 0,000 195
* The stay areas are defined for each model

properties of both the deck and the pylon are summarized in Table 1. The steel Young‟s modulus
is 210000 MPa for the deck and the pylon and 195000 MPa for the stay cables. The ultimate stay
strength, fGUTS, is 1860 MPa. These values are obtained from (Eurocode 2005). The areas of the
stay cables are not included in this table as they are calculated for each cable in each model.
The configurations of the studied examples vary to analyze the influence of the following
variables: (1) Pylon inclination α (ranging from -20° to 20°), (2) Deck cross section (Stiff or
Flexible in Fig. 2(B)), (3) Pylon cross section (Stiff or Flexible in Fig. 2(B)), and (4) Stay cable
configuration (Semi-Fan or Harp in Fig. 1(A)). In addition to these analyses, a study of the effects
of the geometrical nonlinearity is presented. All these variations results in a number of 368
analyzed structures.

2.2 Analyzed load cases

The analyzed load cases include: (1) Permanent loads. These loads include the self-weight and
a superimposed dead load of 3 kN/m2, that is to say 30 kN/m for a 10 m wide deck. (2) Live loads.
The load LM4 (5 kN/m2) of the European Code-1, that is 50 kN/m, is considered.

3. Calculation and assumptions

In this section after summarizing the calculation procedure, the main simulation assumptions
are presented.

3.1 Calculations

In the parametric analysis, the service behavior of a number of non-symmetric cable-stayed

bridges is analyzed with SAP2000. These structures are simulated by Finite Element Models
(FEMs). These FEMs simulate each deck and pylon segment between stay anchorages by mean of
a different beam element with six degrees of freedom. The stay cables are simulated by beam
elements pinned at both edges.
Some authors recommend the calculation of the stress envelopes in cable-stayed bridges by
analyzing a limited number of load cases. This is the case of Walther et al. (1999), who only
considers two load cases to study the live load effects. These load cases correspond with placing
the live load either in the main span or in the back span. Using this limited number of load cases
provides accurate prediction of the maximum stresses in deck and pylon. Nevertheless, the
obtained stresses at the inner part of the deck remain on the non-conservative side. To illustrate
this, Fig. 3 is analyzed. This figure compares the Walther‟s bending moment envelope, in blue
 Structural behavior of non-symmetrical steel cable-stayed bridges

Fig. 3 Envelope of bending moments in service. The envelope proposed in Walther et al. (1999) is
presented by a dotted line and the one used in this paper by a continuous one

solid dash, with the bending moment envelope obtained when every live load location is
considered, in red solid dash. This figure shows that both envelopes present the same stresses in
the pylon and in the proximities of the deck abutments (where the maximum deck bending
moments are found). Nevertheless, this is not the case at the inner part of the deck where the
accuracy of the Walther‟s envelope is lost (as higher bending moments are obtained by the red
envelope). In view of these results, and to assure an accurate simulation of the whole bridge, the
stress and deflection envelopes calculated in this paper have been obtained considering all possible
live load locations. To do so, the analyzed FEMs include a number of load cases equal to the
number of deck beam elements.

3.2 Assumptions

The analyses carried out are based on the following assumptions:

(1) Geometry of the FEMs and load cases. The geometry of the FEMs and the load cases
analyzed are presented in Section 2.1 and 2.2, respectively.
(2) Definition of all stay cable cross sections but the backstay one. The stay cable cross
sections must fulfill two criteria simultaneously. The first one is the limitation of the
maximum stresses. To prevent fatigue failure, the stay cables are designed to avoid
exceeding the 45% of the ultimate stay strength, fGUTS under the characteristic combination
(see SETRA 2001). The second one is the limitation of increment of stresses. To prevent
fatigue failure, the maximum increment of stress in stay cables due to live loads, 𝛥𝜎max ,
must be controlled. In this paper, this value has been fixed to 200 MPa. The amount of live
loads that must be considered for fatigue checking are specified in codes. In this paper, the
fatigue load considered is equal to the 100% of the characteristic live load. The area of
every stay cable, Ac, can be expressed as

𝑃𝐿 + 𝐿𝐿 ·𝑠 𝐿𝐿 ·𝑠
𝐴𝑐 = max⁡ , (5)
sin 𝛼 ·0.45 ·𝑓𝐺𝑈𝑇𝑆 sin 𝛼 ·𝛥𝜎𝑚𝑎𝑥
 Juan J. Jorquera-Lucerga, Jose A. Lozano-Galant and Jose Turmo

where PL and LL are the Permanent and Live loads, s is the cable spacing and α is the
cable inclination defined in Fig. 1. This procedure is based on the method of the tributary
area proposed by Svensson (2012).
(3) Definition of stay cable forces in all stays but the backstays due to permanent loads.
According to Svensson (2012), the Rigidly Supported Continuous Beam Criterion has
been used to define the axial load of the stay cables due to permanent loads.
(4) Definition of backstay area and force due to permanent loads. The definition of the area
and force of the backstay is not as straightforward as in the case of the rest of the stay
cables. The difficulty in defining these elements comes from the fact that backstay force
and stiffness are strongly linked with bridge forces and deflections. This problem has to be
solved by an iterative process, in which each iteration represents a different FEM of the
whole bridge with a certain backstay area. In this model, the axial force of the backstay is
defined to set the bending moment at the bottom of the pylon to zero when the permanent
loads and the axial forces in the rest of stays calculated according to the preceding
paragraph are applied. Once defined the axial force of the backstay for permanent loads,
the variation of stresses, 𝛥𝜎𝑚𝑎𝑥 , and the maximum tensile stresses in the backstay are
calculated using a FEM model. If any of these values exceeds the design thresholds, a new
area of the backstay is defined. This new area is introduced in the FEM that is used in the
following iteration.

4. Parametric studies, results and discussion

This section studies the effects of the main design parameters of the non-symmetrical cable-
stayed bridges in the structural behavior of their structural elements (stays, pylon and deck). The
analyzed design parameters are: the back span length, the pylon inclination, the pylon stiffness, the
deck stiffness, the backstay stiffness, the stay arrangement and the geometrical nonlinearity, After
presenting the results of these analyses, they are discussed in detail.

4.1 Parametric studies

In this section, the results of the analyzed design parameters are presented. It is remarked that
the geometry and the stay areas of all the proposed models are defined according to the procedures
described in Sections 1 and 3, respectively. The Stiff and Flexible sections for the pylon and deck
are described in Fig. 2.

4.1.1 Study of the Backspan Length, L1

This section studies the effect of the back span length, L1, on non-symmetrical cable-stayed
bridges with five different pylon inclinations (α = -20°, -10°, 0°, 10° and 20°). The analyzed back
span lengths are: (1) α = -20°: 5 models (L1 = 10 m, 21 m, 33 m, 51 m and 63 m). (2) α = -10°: 5
models (L1 = 21 m, 33 m, 51 m, 63 m and 75 m). (3) α = 0°: 6 models (L1 = 33 m, 51 m, 63 m, 75
m, 81 m and 99 m). (4) α = +10°: 5 models (L1 = 51 m, 63 m, 75 m, 81 m and 99 m). (5) α = +20°:
2 models (L1 = 75 m and 81 m). The area of the cables, Ac, assuming stiff sections in both the deck
and the pylon are presented in Fig. 4(A). The stress variations in cables due to live loads, Δσc, are
presented in Fig. 4(B).
The structural responses in the deck and the pylon of each model are detailed in Fig. 5 on the
same scale. This figure includes the envelopes of bending moments in the deck, MD (Fig. 5(A)),
Structural behavior of non-symmetrical steel cable-stayed bridges

Fig. 4 Definition of the models

 Juan J. Jorquera-Lucerga, Jose A. Lozano-Galant and Jose Turmo

Fig. 5 Study of the back span L1

the envelopes of vertical deflections in the deck, vD, (Fig. 5(B)), the envelopes of bending
moments in the pylon, MP (Fig. 5(C)) and the envelopes of horizontal deflection in the pylon, uP,
(Fig. 5(D)).
 Structural behavior of non-symmetrical steel cable-stayed bridges

4.1.2 Study of the pylon inclination α

This section analyzes the effect of the pylon inclination, α, in non-symmetrical cable-stayed
bridges with the same back span length (L1 = 51 m). The areas of the cable, Ac, obtained for the
five analyzed pylon inclinations (α = -20°, -10°, 0° and 10°) are summarized in Fig. 6(A).
Fig. 6 also includes the stress variations in cables due to live loads, Δσc, (Fig. 6(B)), the
envelopes of bending moments in the deck, MD, (Fig. 6(C)), the envelopes of vertical deflections
in the deck, vD, (Fig. 6(D)), the envelopes of bending moments in the pylon, MP, (Fig. 6(E)), and
the envelopes of horizontal deflection in the pylon, uP, (Fig. 6(F)).

4.1.3 Study of the pylon stiffness

This section studies how the pylon stiffness influences the structural behavior of non -

Fig. 6 Study of the pylon inclination α with L1 = 51 m

Fig. 7 Study of the pylon stiffness

 Juan J. Jorquera-Lucerga, Jose A. Lozano-Galant and Jose Turmo

Fig. 8 Study of the deck stiffness

symmetrical cable-stayed bridges with a vertical pylon (α = 0°) and six different back span lengths
(L1 = 33 m, 51 m, 63 m, 75 m, 81 m and 99 m). With this aim, the effects of two different pylon
stiffnesses (Stiff and Flexible) are analyzed. The stiffness of the deck is assumed as Flexible.
Fig. 7 summarizes the area of the cables, Ac, (Fig. 7(A)), the stress variations in cables due to
live loads, Δσc, (Fig. 7(B)), the envelope of bending moments in the deck, MD (Fig. 7(C)), the
envelope of vertical deflections in the deck, vD, (Fig. 7(D)), the envelope of bending moments in
the pylon, MP, (Fig. 7(E)), and the envelope of horizontal deflections in the pylon, uP, (Fig. 7(F)).
In this figure continuous lines indicate the results for the Stiff pylon and the dotted ones the results
of the Flexible pylon.

4.1.4 Study of the deck stiffness

This section analyzes the effects of the deck stiffness in non-symmetrical cable stayed bridges
with a vertical pylon (α = 0°) and six different back span lengths (L1 = 33 m, 51 m, 63 m, 75 m, 81
m and 99 m). In this analysis, two different deck stiffnesses (Stiff and Flexible) are considered.
The stiffness of the pylon is assumed as Stiff.
The results of the parametric analysis are summarized in Fig. 8. This figure includes: the area
of the cables, Ac, (Fig. 8(A)), the stress variations in cables due to live loads, Δσc, (Fig. 8(B)), the
envelope of bending moments in the deck, MD, (Fig. 8(C)), the envelope of vertical deflections in
the deck, vD, (Fig. 8(D)), the envelope of bending moments in the pylon, MP, (Fig. 8(E)), and the
envelope of horizontal deflections in the pylon, uP, (Fig. 8(F)). It is remarked that continuous and
dotted lines are used to identify the results of the Stiff and the Flexible deck, respectively.

4.1.5 Study of the backstay stiffness

This section studies the effects of the backstay stiffness in non-symmetrical cable stayed
bridges with an inclined pylon (α = 10°) and a back span length, L1, of 51 m. The areas of the stay
cables strictly required by the procedure described above, ABS, are presented in Fig. 4(A).
With the aim of analyzing the effects of the backstay stiffness, the minimum backstay area, ABs,
has been factored by γBS. Values of γBS = 2, 5 and 10 have been considered and the obtained results
 Structural behavior of non-symmetrical steel cable-stayed bridges

Fig. 9 Study of the backstay stiffness

are compared with the reference values (γBS = 1) in Fig. 9. This figure summarizes the envelope of
bending moments in the deck, MD, (Fig. 9(A)), the envelope of vertical deflections in the deck, vD,
(Fig. 9.B), the envelope of bending moments in the pylon, MP, (Fig. 9(C)), and the envelope of
horizontal deflections in the pylon, uP, (Fig. 9(D)).

4.1.6 Study of the stay configuration

This section analyzes how the stay configuration (Semi-Fan or Fan in Fig. 1) influences the
response of non-symmetrical cable-stayed bridges. The analyzed structures include a vertical
pylon (α = 0°) and six different back span lengths (L1 = 33 m, 51 m, 63 m, 75 m, 81 m and 99 m).
The results of the parametric analysis are summarized in Fig. 10. This figure includes: the cable
areas, Ac, (Fig. 10(A)), the stress variations in cables due to live loads, Δσc, (Fig. 10(B)), the
envelope of bending moments in the deck, MD, (Fig. 10(C)), the envelope of vertical deflections in
the deck, vD, (Fig. 10(D)), the envelope of bending moments in the pylon, MP, (Fig. 10(E)), and the
envelope of horizontal deflections in the pylon, uP, (Fig. 10(F)). It is to highlight that continuous
and dotted lines are used to identify the results of the Semi-Fan and Harp system, respectively.

4.1.7 Study of the geometrical nonlinearity

This section studies the effects of the geometrical nonlinearity (P-δ) in the non-symmetrical

Fig. 10 Study of the stay configuration

 Juan J. Jorquera-Lucerga, Jose A. Lozano-Galant and Jose Turmo

Fig. 11 Study of the geometrical nonlinearity

cable-stayed bridges. The analyzed structures include a vertical pylon (α = 0°) and six different
back span lengths (L1 = 33 m, 51 m, 63 m, 75 m, 81 m and 99 m).
Fig. 11 summarizes the cable areas, Ac, (Fig. 11(A)), the envelope of increments of axial
stresses in the stays, Δσc, (Fig. 11(B)), the envelope of bending moments in the deck, MD, (Fig.
11(C)), the envelope of vertical deflections in the deck, vD, (Fig. 11(D)), the envelope of bending
moments in the pylon, MP, (Fig. 11(E)), and the envelope of horizontal deflections in the pylon, uP,
(Fig. 11(F)). In this figure, continuous and dotted lines are used to identify the results of the
analyses with and without P-δ effects, respectively.

4.2 Results

In this section, the results of the parametric analysis presented in the preceding section are
discussed in detail.

4.2.1 Study of the back span length

Fig. 4(A) shows that the stay cable areas are not influenced by the back span length, L1, since
these areas only depend on the stay inclination. Obviously, the higher the inclination the lesser
efficient the cable and therefore, the higher the area. As illustrated in Fig. 4(B), the back span
length does not play an important role in the oscillation of the stay stresses. Nevertheless, this
figure shows that L1 does influence the maximum and the minimum values, especially in the least
efficient stays (i.e., stays located near the abutment of the main span). In this case, the smaller L1 is
the higher the minimum increments of axial stresses and the lower the maximum ones are. For
example, in the models with α = -20° the maximum values of the stress oscillations in stays vary
from +0 MPa for L1 = 10 m to +60 MPa for L1 = 99 m. In the case of the minimum values, they
vary from -320 MPa for L1 = 10 m to +105 MPa for L1 = 99 m.
Fig. 5 shows that the back span length plays an important role in the structural behavior of both
the deck and the pylon. Regardless of the pylon inclination, the smaller the L1 the higher the
sagging moments (and the corresponding deflections) in the deck. Similarly in the pylon, higher
 Structural behavior of non-symmetrical steel cable-stayed bridges

values of bending moments, especially at its bottom, and higher deflections, at its top, appear
when L1 decreases. The lack of symmetry of the bending moment diagram at the bottom of the
pylon is a consequence of the lack of symmetry of the span distribution. This behavior is
illustrated in the models with α = -20°, where the maximum sagging moments in the deck vary
from 11 MNm for L1 = 10 m to 6 MNm for L1 = 99 m, and in the pylon, these values vary from 42
MNm for L1 = 10 m to 21 MNm for L1 = 99 m.

4.2.2 Study of the pylon inclination

Fig. 6(A) shows that the pylon inclination defines the area of the stay cables according to the
criterion presented in Section 3, where it was also established that the higher the stay inclinations
the lesser efficient the stay and therefore, the higher the area required. As illustrated in Fig. 6(B),
the pylon inclination also influences the oscillation of stresses in the stay cables, especially in
those elements with the highest inclinations (located at the proximities of the abutment of the main
span). In this stay cable, the minimum values of the stresses are increased when the pylon is
inclined towards the back span (α < 0°) while the maximum ones are not significantly changed.
For example, the minimum increment of stress in this stay varies from -50 MPa for α = -20° to -
130 MPa for α = 10°.
Fig. 6(C) shows that inclining the pylon towards the back span (α < 0°) results in slightly
higher bending moment envelopes in the deck at the proximities of the pylon. For example, in the
main span for x = 20 m the maximum hogging bending moment changes from 2 MNm for α = 10°
to 3 MNm for α = -20°. The opposite happens when the pylon is inclined towards the main span (α
> 0°) For example, in the main span for x = 80 m the maximum hogging bending moment changes
from 5 MNm in α = -20° to 7 MNm in α = 10°.
It is noteworthy that when α < 0°, the required area for the backstay is smaller because of the
favorable weight of the pylon. Since this stay is not very efficient due to its small area, higher
horizontal deflections in the pylon are obtained than in a structure where α > 0°. This is
appreciable in Fig. 6(E), where the maximum pylon deflections vary from 125 mm for α = 10° to
100 mm for α = -20°.

4.2.3 Study of the pylon stiffness

Since the cable areas are obtained only by considering its geometry and tributary length, the
pylon stiffness does not influence them at all, as shown in Fig. 7(A). The increment of cable
stresses is therefore only due to the pylon stiffness, and it is virtually negligible, as illustrated in
Fig. 7(B). In the case of the deck bending moment envelopes presented in Fig. 7(C) slight
differences appear at the proximities of both abutments. These differences are clearly observed
when the envelopes of deflections in the deck of Fig. 7(D) are analyzed. As expected, higher
envelopes of deflection are obtained in structures with flexible pylons. This figure also shows the
importance of the back span length (L1). In highly asymmetrical structures (short L1) the
differences between both envelopes are limited to positive deflections in the main span while in
the more symmetrical structures (high L1) these differences appear in both positive and negative
deflections. As illustrated in Fig. 7(F), the stiffness is a factor of major importance in the structural
response of the pylon. In fact, this figure shows that main differences between the models with
stiff and flexible pylons are obtained at intermediate heights. The differences of deflections at the
deck are a consequence of the sensitivity of flexible pylons to horizontal loads. These differences
are found in both the positive and the negative pylon deflections in structures with high L1 and
limited to the positive deflections in structures with small L1.
 Juan J. Jorquera-Lucerga, Jose A. Lozano-Galant and Jose Turmo

4.2.4 Study of the deck stiffness

For the same reason that in the previous study, the deck stiffness does not affect the definition
of the cable areas, as shown in Fig. 8(A). In Fig. 8(B) it is presented that the deck stiffness does
not play an important role in the stay stresses. Regarding the bending moment envelopes in the
deck presented in Fig. 8(C) slight differences appear at the proximities of abutments (hogging
moments in the main span abutment and sagging moments in the back span one). As presented in
Fig. 8(D), the higher the deck stiffness the lower the deck deflections. However, the reduction of
the deck deflections is not proportional to the increase of the deck stiffness. It seems, as happens
with symmetrical bridges, that increasing the deck stiffness is not the most efficient way to reduce
its deflections. Similar behavior is found at the pylon (Figs. 8(E) and (F)), as both bending
moments and deflections increase with the deck stiffness.

4.2.5 Study of the backstay stiffness

Fig. 9 shows the benefits, in both the deck and the pylon, of increasing the stiffness of the
backstay. As presented in Section 4.5.1, the backstay area strictly obtained corresponds to γBS = 1.
For example, maximum hogging moments in the deck are reduced from 3 MNm for γBS = 1 to 2.5
MNm for γBS = 2. Nevertheless, the favorable effect of stiffening the stay is successively less
advantageous when γBS is increased. In fact, the maximum hogging moments in the deck only
change from 2.2 MNm for γBS = 5 to 2.1 MNm for γBS = 10. However, as illustrated in Fig. 9(C),
the backstay stiffness is a design parameter that can be efficiently used to reduce the pylon
deflections.

4.2.6 Study of the stay configuration

Since the angle of the cables is the same for each span in bridges with harp configuration, the
areas of the cables in the same span are equal (Fig. 10(A)). As presented in Fig 10(B), main
increment of cable stresses are found at the vicinities of the pylon. Figs. 10(E) and (F) show how,
in bridges with semi-fan configuration, the need for additional intermediate supports at the
backstay span is evident. This fact is illustrated by the high deflections at intermediate zones of the
pylon. As a consequence of the lack of additional fixed points, the deck deflections for harp
configuration are also very high (Fig. 10(D)). Regarding this point, there is not significant
variation with respect to symmetrical bridges.

4.2.7 Study of the geometrical nonlinearity

Fig. 11 shows that the effect of geometrical nonlinearity is very small and its consideration does
not affect the previously described responses.

4.3 Discussion

4.3.1 Initial design of the cable areas

A primary criterion to define the cable area consists of limiting its maximum stress to a
percentage of the cable strength (usually the 45%, see Svensson 2012) assuming a tributary area in
the deck for both permanent and live loads. A second criterion defines the cable areas limiting the
oscillation of axial stresses to 200 MPa when a given percentage of the live loads is applied (in this
paper this percentage has been assumed as 100%).
In the simulations presented in this paper, only distributed live loads, and not heavy vehicles
are considered. As presented in Figs. 4(B) and 5(B), these assumptions are on the unsafety side,
 Structural behavior of non-symmetrical steel cable-stayed bridges

since when heavy vehicles are not considered the area of the cables are underestimated. For these
reasons, when only distributed loads are considered, it seems reasonably to reduce the percentage
of the cable strength from 45% to 35% or a similar value. On the other hand, the oscillation of stay
stresses criterion might be the determining criterion in light cable-stayed bridges. In these
structures, the ratio of the live load considered in the oscillation is defined by the structural code.
This ratio is usually lower than the 100% assumed in the presented analyses.

4.3.2 Design of the backstays

Fig. 12(A) shows the area of the backstays in structures with different pylon inclinations (α = -
20°, -10°, 0°, 10° and 20°) for a back span length L2 ranging from 10 to 100 m. The geometries
and mechanical properties of these models are defined according to the procedures presented in
Sections 1 and 3, respectively.
This figure shows: (1) Role of the pylon inclination: The higher the pylon inclination towards
the main span (α < 0°), the higher the backstay area as this element has also to resist the
unfavorable weight of the pylon. The opposite effect appears when the pylon is inclined towards
the back span (α > 0°). (2) Role of the back span length: Independently of the pylon inclination, a
minimal backstay area is obtained when the back span length represents a ratio between the 20 and
50% of the main span length.

4.3.3 Axial forces and stresses in the backstays

The axial force for permanent loads in the backstay has been defined to set to zero the bending
moments at the pylon foundation. In a symmetrical bridge (L1/L2 = 1) the axial force in the
backstay is very low. As L1/L2 reduces, the axial force of the backstay for permanent loads
increases. This force also depends on the pylon inclination. In fact, when α > 0°, the backstay force
is reduced because of the favorable effect of the pylon weight.
The axial force in the backstays oscillates due to the live loads. This is appreciable in Fig.
12(B), where the variation of axial forces in the backstays of structures with two pylon inclinations
(α = -0° and 10°) and different back span lengths (L2 from 10 to 100 m) are compared. This figure
shows the axial forces due to permanent loads, and their maximum and minimum axial forces,
when, besides, live loads act upon the bridge. This figure shows that in structures with a reduced
length L2 the axial forces for permanent loads are practically equal to the minimum ones. On the
other hand, when L2 is near L1, the axial forces for permanent loads present a medium value
between the maximum and the minimum lines. This situation can be easily explained by the fact
that the live loads in the back span reduce the tensile stresses of the backstay, while the live loads
in the main span present the opposite effect.
In the vast majority of bridges analyzed, the stress oscillation criterion has been the most
adverse criterion to define the backstay area. This is appreciable in Fig. 12(C), where the
increments of axial stresses in the backstays of structures with two pylon inclinations (α = 0° and
10°) and different back span lengths (L2 from 10 to 100 m) are compared, and where the maximum
and minimum of their stresses and their oscillation are shown. Fig. 12(C) shows that in most of the
structures the stress oscillations in stays are near to 200 MPa (design threshold assumed in the
codes).

4.3.4 Weight of steel in cables

The weights of steel in cables in cable-stayed bridges with five pylon inclinations (α = -20°, -
10°, 0°, 10° and 20°) and different back span lengths (L2 from 10 to 100 m) are summarized in Fig.
 Juan J. Jorquera-Lucerga, Jose A. Lozano-Galant and Jose Turmo

12(D), where the weight of the backstay cables, the rest of the cables and of all the cables are
compared. Fig. 12(D) can be expressed in terms of kg/m2 as presented in Fig. 12(E). In this figure
the minimum steel weight is obtained when L2 is near 40 m (L2/L1 = 0.40).
For all the pylon inclinations the minimum value is obtained for α = -20°, although the
differences are not significant. In symmetrical bridges the optimal pylon inclination is α = 10°. On
the other hand, when 0.55 ≤ L2/L1 ≤ 0.75 the optimal pylon corresponds with α = 0°. It is remarked
that differences between the different cases are not significant enough to advice one or another
pylon inclination. This is not the case in structures with a short L2 span as in this structures the
steel weight is significantly reduced when α < 0°.

4.3.5 Deflections limits

An important aspect to take into account is the fact that no deflection limitations are introduced

Fig. 12 Results backstay areas (A); variation of stay forces due to live loads (B); variation of stay
stresses due to live loads (C); steel consumption (D); and steel in kg/m2 (E)
 Structural behavior of non-symmetrical steel cable-stayed bridges

in Fig. 12(D) and 12(E). A reference value might be considered as L2/1000 for the frequent live
load [IAP 11], which corresponds to the Load Model 4 (5 kN/m2) defined at Eurocode 0, factored
by ψ = 0,4. Nevertheless, this ratio might be reduced in agreement with the client. For example, in
the Sunniberg Bridge this value was as low as L2/400 (Marchetti and Lecinq 1999).

4.3.6 Effect of the backspan length

The back span length plays an important role in the structural response of the bridge. The
shorter this length, the higher both the bending moments and the deflections in the main span. In
very short back spans, the pylon only leans towards the main span and the bending moments due
to live loads use to have a predominant sign.
The tensile forces in the cables are usually concentrated at the places with maximum hogging
moments and are reduced significantly at the proximities of the abutments. In the proximities of
the pylon, the maximum stresses do not vary significantly between adjacent stays.

4.3.7 Effect of the pylon inclination

When the pylon lends towards the back span for a certain back span length, both the bending
moments and the deflections of the main span slightly increase. The bending moments in the pylon
are practically the same but the maximum values are located at different heights. It is to be
remarked that the pylons leaning towards the back span present higher horizontal deflections.

4.3.8 Effect of the deck stiffness

The analyzed structures show that increasing the deck stiffness is not recommendable as the
bending moments and deflections are not reduced in the same proportion. For example, when the
deck inertia is increased 11.35 times (that is to say, when the deck is changed from flexible to stiff)
the maximum bending moments and deflections in this element are only reduced the 3,84% and
the 9,72%, respectively.

4.3.9 Effect of the pylon-back stay stiffness

Vertical deflections at the deck are counteracted by two complementary structural systems: the
axial stiffness of the backstay and the bending stiffness of the pylon. Increasing the backstay
stiffness is a much more efficient way to reduce the deflections and internal forces at the deck.

4.3.10 Effect of the stay configuration

The bending moments in both spans are higher for the harp configuration than for the semi-fan
one.
As expected, major differences appear in the pylon when the two configurations are compared.
As the harp configuration is more sensitive to horizontal loads, higher horizontal deflections
throughout the pylon are obtained. These horizontal deflections increase the vertical deflections in
the deck. From the results obtained it might be concluded that the harp system is less efficient than
the semi-fan one. To increase the efficiency of the harp system intermediate supports can be
introduced to anchor the deck (as in the Knie Bridge).

4.3.11 Effect of the geometrical nonlinearity

For the cases studied, the effect of the geometrical nonlinearity is not significant.
 Juan J. Jorquera-Lucerga, Jose A. Lozano-Galant and Jose Turmo

5. Conclusions

This paper presents a parametric study addressing the structural response of asymmetrical
cable-stayed bridges. In this analysis, the effects of the main design parameters (the pylon
inclination, the back span length, the stiffness of the deck, pylon and backstay and the stay
configuration) are compared. This study includes the role of the geometrical nonlinearity. Despite
of the growing number of built examples, this structural type has deserved little attention from
researchers.
The parametric analyses carried out show that: (1) In the vast majority of analyzed bridges, the
stress oscillation criterion stands as the most adverse criterion to define the backstay area. On the
contrary, the rest of the stays are usually defined by limiting its axial strength. (2) The higher the
pylon inclination towards the main span, the higher the backstay area. The minimal backstay area
is obtained when the back span length represents a ratio between the 20 and 50% of the main span
length. (3) In most analyzed structures, variations in steel weight between structures with different
pylon inclinations are not significant enough. Nevertheless, this is not the case in structures with a
short backspan. In these structures, the steel weight is significantly reduced when the pylon is lent
towards the backspan. (4) The backspan length plays an important role in the structural response of
asymmetrical bridges. The shorter this length the worse the structural response in the main span is.
(5) Leaning the pylon towards the back span for a certain back span length, worsen the structural
behavior of the main span. (6) Deck deflections are not reduced in the same proportion when the
deck bending stiffness is increased. For this reason, the use of stiffer decks is not advised. (7)
Increasing the backstay stiffness is more efficient to improve the deck behavior than increasing the
bending stiffness of the pylon. (8) Generally speaking, the harp configuration is less efficient than
the semi-fan one. (9) The geometrical nonlinearity does not play an important role in the structural
response of the asymmetrical bridges analyzed in this study.
As a final conclusion, generally speaking, asymmetrical bridges present a less efficient
structural behavior, since the disadvantages of symmetrical bridges are enhanced, especially for
very short back spans. As showed above, factors like the area and inclination of the backstay cable,
the cable arrangement (harp or semi-fan), the stiffness and inclination of the pylon, and the
stiffness of the deck, are of major importance. With no doubts, this kind of bridges need further
studies in order to, for example, establishing criteria to design the area of the backstay cables.
These and other problems will be addressed in the near future by the authors.

Acknowledgments

The authors would like to thank the former students D. Javier Sánchez Mateos Enrique, D.
Mario Pérez Sánchez y D. Alberto de la Torre Palencia for contributing to the initial studies of this
paper. The authors thank the Spanish Ministerio de Economia y Competitividad and the FEDER
funds for the funding provided through the research grant BIA2013-47290-R directed by Jose
Turmo.

References

Agrawal, T.P. (1997), “Cable-stayed bridges-parametric study”, J. Bridge Eng., 2(2), 61-67.
Atmaca, B. and Ates, S. (2012), “Construction stage analysis of three-dimensional cable-stayed bridges”,
 Structural behavior of non-symmetrical steel cable-stayed bridges

Steel Compos. Struct., Int. J., 12(5), 413-426.

Baldomir, A., Hernandez, S., Nieto, F. and Jurado, A. (2010), “Cable optimization of a Long Span Cable-
Stayed Bridge in la Coruna (Spain)”, J. Adv. Eng. Software, 41(7-8), 931-938.
Bazant, Z.P. (1972), “Prediction of concrete creep effects using age-adjusted effective modulus method”,
ACI Journal, 69(4), 212-219.
Casas, J.R. and Aparicio, A.C. (1998), “Monitoring of the Alamillo cable-stayed bridge during construction”,
Experim. Mech., 38(1), 24-28.
Chen, D.W., Au, F.T.K., Tham, L.G. and Lee, P.K.K. (2000), “Determination of initial cable forces in
prestressed concrete cable-stayed bridge for given design deck profiles using the force equilibrium
method”, Comput. Struct., 74(1), 1-9.
Du, G.H. (1989), “Optimal cable tension and construction tensioning of cable-stayed bridges”, Bridge
Construct., 74, 18-22. [In Chinese]
Eurocode 3 (2005), Design of Steel Structures.
Gimsing, N.J. (1997), Cable Supported Bridges, Concept and Design, John Wiley and Sons, Chichester, UK.
Guan, H. (2000), “Construction control of cable-stayed bridges”, Ph.D. Thesis; The Hong Kong University
of Science and Technology, Hong Kong, China.
Gunaydin, M., Adanur, S., Altunisik, A.C., Sevim, B. and Turker, E. (2014), “Determination of structural
behavior of Bosporus suspension bridge considering construction stages and different soil conditions”,
Steel Compos. Struct., Int. J., 17(4), 405-429.
Hassan, M.M. (2013), “Optimization of stay cables in cable-stayed bridges using Finite element, genetic
algorithm, and b-spline combined technique”, Eng. Struct., 49, 643-654.
Hassan, M.M., Nassef, A.O. and El Damatty, A.A (2012), “Determination of optimum post-tensioning cable
forces of cable-stayed bridges”, Eng. Struct., 44, 248-259.
Janjic, D., Pircher, M. and Pircher, H. (2002), “The unit load method - some recent applications”, Adv. Steel
Struct., II, 831-837.
Janjic, D., Pircher, M. and Pircher, H. (2003), “Optimization of cable tensioning in cable-stayed bridges”, J.
Bridge Eng., 8(3), 131-137.
Jorquera-Lucerga, J.J. (2013), “Understanding Calatrava‟s bridges: A conceptual approach to the „La
Devesa-type‟ footbridges”, Eng. Struct., 56, 2083-2097.
Lazar, B.E., Troitsky, M.S. and Douglas, M.C. (1972), “Load analysis balancing of cable-stayed bridges”,
ASCE J. Struct. Eng., 92(8), 1725-1740.
Lozano-Galant, J.A., Paya-Zaforteza, I., Xu, D. and Turmo, J. (2012a), “Analysis of the construction process
of cable-stayed bridges built on temporary supports”, Eng. Struct., 40, 95-106.
Lozano-Galant, J.A., Paya-Zaforteza, I., Xu, D. and Turmo, J. (2012b), “Forward Algorithm for the
construction control of cable-stayed bridges built on temporary supports”, Eng. Struct., 40, 119-130.
Lozano-Galant, J.A., Xu, D., Paya-Zaforteza, I. and Turmo, J. (2013), “Direct simulation of the tensioning
process of cable-stayed bridges”, Comput. Struct., 121, 64-75.
Lozano-Galant, J.A. and Turmo, J. (2014a), “An algorithm for simulation of concrete cable-stayed bridges
built on temporary supports and considering time dependent effects”, Eng. Struct., 79, 341-353.
Lozano-Galant, J.A. and Turmo, J. (2014b), “Creep and shrinkage effects in service stresses of concrete
cable-stayed bridges”, Comput. Concrete, 13(4), 483-499.
Lozano-Galant, J.A., Ruiz-Ripoll, L., Paya-Zaforteza, I. and Turmo, J. (2014), “Modifications of the stress-
state of cable-stayed bridges due to staggered erection of their superstructure”, The Baltic J. Road Bridge
Eng., 9(4), 241-250.
Lozano-Galant, J.A., Paya-Zaforteza, I. and Turmo, J. (2015), “Effects in service of the staggered construction
of cable-stayed bridges built on temporary supports”, The Baltic J. Road Bridge Eng., 10(3), 247-254.
Lute, V., Upadhyay, A. and Singh, K.K. (2009), “Computationally efficient analysis of cable-stayed bridge
for GA-based optimization”, Eng. Appl. Artif. Intel., 22(4-5), 750-758.
Manterola, J., Siegrist, C. and Gil, M.A. (2006), Puentes, Escuela Técnica Superior de Ingenieros de
Caminos, Canales y Puertos de Madrid, Madrid, Spain. [In Spanish]
Marchetti, M. and Lecinq, B. (1999), “Stay adjustment: From design perspective to on site practice”,
 Juan J. Jorquera-Lucerga, Jose A. Lozano-Galant and Jose Turmo

Proceedings of IABSE, Malmö, Sweden, June.

Menn, C. (1990), Prestressed Concrete Bridges, Springer Science & Business Media.
Negrao, J.H.O. and Simoes, L.M.C. (1997), “Optimization of cable-stayed bridges with three dimensional
modeling”, Comput. Struct., 64(1-4), 741-758.
Scotti, A. (2003), “Long term behavior of cable-stayed bridges”, Master‟s Thesis; Polititecnico di Milano,
Italy.
SETRA (2001), Recommandations de la Commission Interministrielle de la Precontrainte, Service d‟Etudes
Techniques des Routes et Autoroutes, France.
Simoes, L.M.C. and Negra, J.H.J.O. (2000), “Optimization of cable-stayed bridges with box-girder decks”,
Adv. Eng. Software, 31(6), 417-423.
Svensson, H. (2012), Cable-Stayed Bridges: 40 years of Experience Worldwide, Wiley.
Virlogeux, M. (1994), “Erection of cable-stayed bridges: the control of the designed geometry”, Proceedings
of International Conference A.I.P.C., F.I.P., Cable-Stayed and Suspension Bridges, Volume 2, Deauville,
France, October, pp. 321-350.
Walther, R., Houriet, B., Isler, W., Moïa, P. and Klein, J.F. (1999), Cable-Stayed Bridges, Thomas Telford.
Wang, P.H., Tseng, T.C. and Yang, C.G. (1993), “Initial shape of cable-stayed bridges”, Comput. Struct.,
46(6), 1095-1106.
Wang, Y.C., Vlahinos, A.S. and Shu, H.S. (1997), “Optimization of cable preloading on cable-stayed
bridges”, SPIE, 3043, 248-259.

CC