Automation in Construction 37 (2014) 196–202

Contents lists available at ScienceDirect

Automation in Construction
journal homepage: www.elsevier.com/locate/autcon

Experiment and analysis of flexural capacity and behaviors of pre-stressed

composite beams
Hu Shaowei ⁎
Department of Materials and Structural Engineering, Nanjing Hydraulic Research Institute, Nanjing 210029, China

a r t i c l e i n f o a b s t r a c t

Article history: At present, there is no design specification for the pre-stressed composite beam. To research the bending mechanical
Accepted 20 August 2013 behaviors of pre-stressed steel–concrete composite box beams and provide corresponding design specification and
Available online 20 September 2013 requirements, the tests of ten pre-stressed composite box beams were finished, and the effects on flexural behaviors
were given for different initial pre-stress levels, the forms of reinforcement placement and loading method and so
Keywords:
on. Design calculation methods under serviceability and bearing capacity limits were mainly discussed. Test results
Pre-stressed composite box beam
Load deflection behaviors
indicated that the pre-stressed composite beam displays a lot of advantages compared with ordinary composite
Serviceability limit state beams. The flexural capacity of the composite beams through using pre-stressed technology was greatly increased.
Ultimate flexural capacity The elastic-plastic analysis of the pre-stressed composite beams was finished; what's more, the equations of deflec-
Interface slip tion and elastic bearing capacity with consideration of interface slip were derived out. Calculating equations for
Experiments research on bending behaviors ultimate banding capacity were derived with consideration of pre-stress increments. Finally, the calculated values
Calculation of deflection met precise requirements when compared with the test results. This study provides advisory opinions for the
Bearing capacity designer to analysis and design of the pre-stressed composite beams.
© 2013 Elsevier B.V. All rights reserved.

1. Research significance experimental results were compared according to the stress characteris-
tics of beams. Results from this paper provide the foundation for further
In recent years, the use of pre-stressed composite beam structures improvements in design theory and analysis of pre-stressed composite
has increased significantly with various types of bridges being used in box beams.
urban overpasses, and railway and highway bridges [1]. Advantages of
the pre-stressed composite box beams include large flexural rigidity,
2. Pre-stressed composite box beams experiment
torsion rigidity, good integral performance and high stability and so
on. In 2005, Liu, Hanbing [2] provided the structural behavior of eight
2.1. Test design
external pre-stressed steel–concrete composite beams with 4.0 m
span from symmetric loading and bending tests. Nie, Hu (1999)[3]
Ten pre-stressed steel–concrete composite box beams with number
presented the experimental studies of the flexural behavior of steel–
15 to 24 were designed. The specimen cross-sectional structure,
concrete composite beams, and four simply supported composite
parameters and the loading apparatus were shown in Fig. 1, Table 2
beams subjected to two-point concentrated loads were tested and com-
and Fig. 2, respectively. The cross-sectional size was determined mainly
pared to investigate the effect of high strength engineering materials.
by its elastic neutral axis location. Generally, the neutral axis was set in
However, there is no design provision for the pre-stressed composite
the precast slab above the interface and as close as possible to the steel
beam in the current standards [4,5]. In the past, the study on the flexural
beam to promote the favorable attributes of the two kinds of material.
behavior of composite beams had focused mainly on non-prestressed
The hydraulic servo machine was used, and the force values of two hy-
section beams ([6], [7], [8]). In this paper, to understand the flexural be-
draulic jacks were the same using the bleeder (Table 1).
havior of pre-stressed steel–concrete composite box beams, the tests of
ten full-scale pre-stressed composite box beams were conducted.
Furthermore, real-time measurement and loading of the pre-stressed 2.2. Test results and analysis
tension process were performed, mainly measuring changes in strain,
interface slip and deflection of feature positions. The practical calculation The bearing capacity decreases with concrete crushing at the section
formula for deflection and bearing capacity under serviceability of maximum bending (see Fig. 3). In Table 2, My, δy and ΔPy are the mid-
limit state and bearing capacity were derived out; theoretical and span moment, mid-span deflection and pre-stressed reinforcement
force increment, respectively. When the steel beam bottom yielded,
⁎ Tel.: +86 2585829601. Mu, δu and ΔPu is the mid-span moment, mid-span deflection and pre-
E-mail address: hushaowei@nhri.cn. stressed reinforcement force increment in the limit state of the bearing

0926-5805/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.autcon.2013.08.004
 H. Shaowei / Automation in Construction 37 (2014) 196–202 197

capacity, respectively. Pu is the pre-stressed reinforcement force in the rigidity reduction coefficient ζ and the composite functioning coefficient
limit state of the bearing capacity. ϕ are related to the shear rigidity k of the stud connector. Various shear–
From Table 2, the pre-stress expands the elastic scope of the beam slip curves have been proposed for the stud connector. The proposed
and increases bearing capacity when compared with an ordinary com- model in the paper is comprehensively applied as follows:
posite beam. For the linear-type composite PCB-15 and PCB-17 without
−ns m
limit block, the load increases from 247.23 to 262.31 kN m when the V ¼ V u 1−e ð1Þ
steel beam bottom yields. Similarly, as for the fold-line type pre-stressed
reinforcement placement, the pre-stressed reinforcement of PCB-20 is Vu is the limit bearing capacity of the studs, s is the slip, and m and n
two times as that of PCB-19. Although initial effective pre-stress is rela- are parameters achieved from the test.
tively low, its limit bearing capacity is relatively high. As for the linear- The anti-slip rigidity of the stud can be said to equal its limit i.e.
type with limit block such as PCB-23, the increment in pre-stress rein-
c
forcement force is 167.8 kN due to its increase in steering block function k ¼ Nv;k : ð2Þ
and eccentricity. Comparing PCB-19 with PCB-23 shows that when the
effective pre-stress is the same as the eccentricity, fold-line type My In this case, Ncv,k is the standard value of the bearing capacity rather
with reinforcement placement is relatively large. For the pre-stressed than the bearing capacity of the studs.
composite beam without limit block at the limit state, the yield ratio ECCS1981 Composite Structures provides an equation to calculate
and ductility ratio are lower than the ordinary composite beam, but as the standard value of the stud shear bearing capacity
for the pre-stressed composite beam with limit block or steering block,
qffiffiffiffiffiffiffiffiffiffiffiffiffi
the yield and ductility ratios are higher than other beams. Figs. 4 and 5 c
Nv;k ¼ 0:46As f ′ ck Ec ≤0:7f u As ð3Þ
show the slip distribution in longitudinal direction and the slip increase
for PCB-15 and CB-16.
The stud bearing capacity equation as given by European standard
Fig. 4 shows that the slip distribution is similar in both the pre-
EC4 has been changed into a design value expression equation
stressed and ordinary composite beams. From Fig. 5 it can be seen that
the load–slip curve of the pre-stressed composite beam can be divided qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
into three stages. Initially, no slip occurs at the interface. In the second c
0:37As f ′ ck Ecm 0:8f u As
Nv ¼ ≤ ð4Þ
stage, the increase of slip is relatively slow. In the third stage, concrete γv γv
surrounding the studs is gradually drawn to the crack, and bending of
the stud and crack of the concrete is the main sources of slip and the China's Code for Design of Steel Structures (GB50017-2003) states
slip increases fast relatively. that the design value of the stud bearing capacity is
qffiffiffiffiffiffiffiffiffiffi
c
3. Serviceability and bearing capacity limit states Nv ¼ 0:43As f c Ec ≤0:7fγAs ¼ 1:169f As ð5Þ

3.1. Determination of the shear rigidity of the stud
In China's current standard, Code for Design of Steel Structures
(GB 50017-2003), the influence of slip on the flexural rigidity of com-
posite beam is considered with the reduced rigidity methods. Both the
′
f ck is the characteristic value of the compressive strength of a
150 × 300 mm concrete cylinder; Ecm is the average elastic modulus
of the concrete; γv is the resistance factor of the stud bearing capacity
(1.25 is suggested for EC4); As is the cross-sectional area of the stud; fu
is the limit tensile strength of the stud material; f is the design value

800 800
130

130

stud
stirrup stud
170

longitudinal bar stirrup

170

longitudinal bar
open steel steel strand
40

box girder open steel steel strand

40

35 130 35 box girder

200 subplate

a) Cross-sectional of the mid span b) Cross-sectional of the support

concrete wing
open steel box girder
steel strand
subplate

c) side view of the specimen

Fig. 1. Cross-sectional details of beam specimen. (a) Cross-sectional of the mid span. (b) Cross-sectional of the support. (c) Side view of the specimen.
198 H. Shaowei / Automation in Construction 37 (2014) 196–202

Jack
Support
Jack

Support

Fig. 2. Styleonme and sketch of loading apparatus.

of the tensile strength of the stud material; and γ is the ratio between In the elastic-plastic stage:
tensile and yield strength of the stud material. According to the national
standard Cheese Head Studs for Stud Welding (GB/T 10433-2002), E s A0 pEs A0 EA
when the stud material level is equal to 4.6, γ = 1.67. ξ0:75 ¼ ¼ ¼ 1:2 s 02 ¼ 1:2ξ
KL2 ns k0:75 L2 ns kL ð10Þ
If Eq. (3) is expressed using China's standard, the following is 1
obtained ϕ0:75 ¼
1 þ 12ξ
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 
s
N v;k ¼ 0:37As f ′ ck Ecm ¼ 0:37As 1:675 e1:779 f c Ecm
  qffiffiffiffiffiffiffiffiffiffiffiffiffi ð6Þ
3.2. Calculation of deflection in serviceability limit state
¼ 0:479 e 0:494 As f c Ecm :
While calculating the deflection of the pre-stressed composite beam,
The standard value of the stud shear rigidity as an average value of its adverse effect should be taken into consideration. The loading process
the above-mentioned standard was used as namely can be divided into two stages: a pre-stressed tensioning stage and an
external load stage. In the pre-stressed tensioning stage, there will be
qffiffiffiffiffiffiffiffiffiffi
an inverted arch δ1 in the composite beam, which considers the pre-
k0:5 ¼ 1:25  0:43As f c Ec : ð7Þ
stress as an external force according to an equivalent load, which can
be calculated using a structural mechanical equation.
This value is 1.25 times the design value of China's stud bearing
capacity, with units of N/mm. At the elastic-plastic stage of the stud Z Z
l M ðxÞM p ðxÞ l MðxÞM p0 ðxÞ
force, namely, when the stud bears 70 to 75% of the limit load, the con- δ1 ¼ dx ¼ dx ð11Þ
crete surrounding the stud cracks and the rigidity of the stud decreases: 0 B 0 Es I0 þ ϕA0 d2c

qffiffiffiffiffiffiffiffiffiffi
1 Mp(x) is the moment produced by the equivalent load of the initial
k0:75 ¼  0:43As f c Ec ð8Þ
1:2 pre-stress P0, and M ðxÞ is the moment produced by the function unit
load installed at the displacement position.
The composite action coefficient φ is calculated according to the In the external load functioning stage, the increment of the pre-
rigidity of the stud at the elastic and elastic-plastic stages. In the elastic stressed reinforcement is not taken into consideration. Meanwhile, the
stage: force increment ΔP produced by the pre-stressed reinforcement, and
inverted arch δΔp is produced in the composite beam caused by the
Es A0 EA pre-stressed increment. The elastic deflection is
ξ0:5 ¼ ¼ 0:8 s 02 ¼ 0:8ξ
KL2 ns kL ð9Þ
1
ϕ0:5 ¼ δ2 ¼ δload −δΔp : ð12Þ
1 þ 8ξ

Table 1
The design parameters of specimen.

Test no. Span/mm Sectional dimension Reinforcement Reinforcement cover/mm Stud layout

Steel/mm Web/mm Longitudinal bar Stirrup

PCB-15
CB-16
PCB-17
SCB-18 Top flange: 80 × 10
PCB-19 4000 Web: 150 × 8 800 × 130 ϕ10 at 187.5 ϕ8 at 200 20 2ϕ16 at 140
PCB-20 Bottom flange: 240 × 10
SCB-21
SCB-22
PCB-23
PCB-24 2000
 H. Shaowei / Automation in Construction 37 (2014) 196–202 199

Fig. 3. Failure of specimens with bending.

Similarly, for the function unit load installed at the displacement The additive moment of the section moment reduction caused by
position, the moment distribution produced is M ðxÞ. ΔNs is

Z Z ΔM ¼ Es εsl hs ðhAs1 þ 2hc As3 Þ=6h: ð17Þ

l MðxÞMload ðxÞ l M ðxÞM load ðxÞ
δload ¼ dx ¼ dx ð13Þ
0 B 0 Es I 0 þ ϕA0 d2c
As1 and As3 are the cross-sectional area of the web and support
Z Z plates, respectively.
l MðxÞM ΔP ðxÞ l MðxÞMΔP ðxÞ
δΔP ¼ dx ¼ dx ð14Þ According to the additive deformation method,
0 s I 0 þ ϕA0 dc
B 2
0 E
′ ′
Δφ ¼ εsl =h ¼ ζ M =EI h : ð18Þ
The pre-stressed increment ΔP can be calculated according to the
equation provided in Ref. [9]. While calculating the moment MΔp(x)
Substitution of Eq. (18) into Eq. (17), yields
produced by an equivalent load of pre-stressed increment, the decrease
in eccentricity of the pre-stressed reinforcement is not taken into
ζ ′ M′ hs
consideration. ΔM ¼ ðhAs1 þ 2hc As3 Þ ð19Þ
6Ieq
3.3. Elastic bearing capacity of pre-stressed composite beam
Therefore, in the elastic limit state, the composite beam moment at
The additive deformation method to calculate the elastic bearing ca- the corresponding steel beam yield is:
pacity of the pre-stressed composite beam was used with calculation " #
model as shown in Fig. 6. The calculation steps of the additive deforma- ζ ′ hs ′
My ¼ 1− ðhAs1 þ 2hc As2 Þ M y : ð20Þ
tion method are the transformed section method is used without consid- 6Ieq
eration of the elastic bearing capacity M′ y of the pre-stressed composite
beam with slip effect; another is that the slip effect is considered to calcu- M′ y is the yield moment of the pre-stressed composite beam calcu-
late the reduction coefficient of the elastic moment. lated by the transformed sectional method; ζ′ is the rigidity reduction
It is assumed that the moment is M when the composite beam does coefficient of the pre-stressed composite beam, which can be calculated
not consider slip. At this time, the relative slip strain is εsl, whereas the from Ref. [10]. Results from the theoretical equation were compared to
additive strain at the top of the steel beam is experimental data from six tested beams [11]. The yield load My of the
pre-stressed composite beam and deflection δy at the serviceability
εssl ¼ εsl hs =h: ð15Þ limit state were calculated according to Eq. (20) and Eq. (12) and com-
pared to experimental results, as shown in Table 3 (j represents the cal-
The additive pressure on the steel beam section is culated values and t represents the measured values.)
Data from Table 3 show that the yield load increases as long as the ini-
ΔNs ≈Es εsl hs ð0:5As1 þ As3 Þ=h ð16Þ tial effective pre-stress increases. In general, the theoretical calculated

Table 2
Main parameters of beam specimen.

Test no. Grouping Pre-stressed Valid pre-stressed Eccentricity/mm Loading conditions Parameter comparison
reinforcement quantity strength/kN

PCB-15 Linear-type non-limit block 2 204.6 156.8 Pure bending

B-16 Stress free – – – Pure bending Degree of effective prestress and
reinforcing effect
PCB-17 Linear-type non-limit block 2 230.73 156.8 Pure bending
CB-18 Linear-type non-limit block 2 232.03 156.8 Pure bending
PCB-19 Fold-line type 2 191.65 236.8 Pure bending Influenced of rebar and degree of
effective prestress on reinforcing effect
PCB-20 Fold-line type 4 183.40 236.8 Pure bending
CB-21 Fold-line type 2 199.88 236.8 Pure bending
CB-22 Fold-line type 2 242.45 236.8 Pure bending
PCB-23 Linear-type limit block 2 198.07 236.8 Pure bending Eccentricity and limit block
PCB-24 Fold-line type 2 191.43 236.8 Bending shear Loading condition
200 H. Shaowei / Automation in Construction 37 (2014) 196–202

1.8 2.0
PCB-15 M/Mu=0.6 CB-16 M/Mu=0.6
0.7
1.8 0.7
1.6 0.8
0.85 0.8
0.91 1.6 0.85
1.4 0.96 0.90
0.95
1.4
1.2 1

1.2
Shear strain

Shear strain
1.0
1.0
0.8
0.8
0.6
0.6

0.4
0.4

0.2 0.2

0.0 0.0
0 25 50 75 100 125 150 175 200 225 0 25 50 75 100 125 150 175 200 225
The distance from the mid-range (cm) The distance from the mid-range (cm)

PCB-15 CB-16
Fig. 4. Slip distribution at interface.

values match the measured values well, which implies that the theoretical
calculations can be considered for real applications.
3.4. Calculation of ultimate flexural capacity according to the plastic theory
Test results show that the limit strain of ordinary reinforcements is
small, and it can be neglected as compared with a steel beam [12,17].
The stress and strain distribution of the beam section at the limit state
is shown in Fig. 7.
For the full shear connected composite beam, the shear connector can
effectively transmit shear between the composite beam interfaces from
Ref. [13][14]. Therefore, the height c in the equivalent rectangular compres-
sion region of the concrete is given according to equilibrium conditions and
is
f c Bc c ¼ As1 f y1 þ As2 f y1 þ ðP 0 þ ΔP Þ:
fc is the concrete compressive strength; Bc is the effective width of the con-
crete flange plate; As1 and As2 are the cross-sectional areas of the web plate
and bottom plate of the steel beam respectively; fy1 and fy2 are the yield
strengths of the flange and bottom plates respectively; and P0 and ΔP are
the initial force of the pre-stressed reinforcement and the increment of
the pre-stressed reinforcement force, respectively. The height of the con-
crete equivalent compression zone of the flange plate from Ref. [15][16] is
 
f y1 As1 þ f y2 As2 þ f p0 þ Δf p Ap
c¼ : ð22Þ
f c Bc
The pre-stressed increment Δfp is obtained from Ref. [17], and it can
be shown that the pre-stressed force increment ΔP is in relation to the
regional height of the concrete flange plate in the compression region
which can be calculated by iteration using the following steps
(1) Set Δfp0 = 0 and achieve height c1 of concrete compression zone;
(2) Substitute height c1 into related equation to obtain Δfp1;
ð21Þ (3) Substitute Δfp1 into equation to obtain height c2 of the concrete
compression zone;
(4) Repeat steps (1) to (3) until Δfpn − Δfpn − 1 is relatively small.
This will yield the height c of the concrete equivalent compression
zone and the pre-stressed force increment Δfp.

400
CB-16
450
PCB-15 350

400
300 The distance from the mid-range
350 The distance from 0cm
mid-range: 62.5cm
Moment M / kN.m
Moment M / kN.m

0cm 250
300 62.5cm 93.75cm
93.75cm 125cm
125cm 156.25cm
250 156.25cm 200 187.5cm
187.5cm
200
150

150
100
100

50
50

0 0
-0.3 0.0 0.3 0.6 0.9 1.2 1.5 1.8 2.1 2.4 2.7 3.0 3.3 3.6 3.9 4.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Strain / mm Strain / mm

PCB-15 CB-16
Fig. 5. Load–slip at interface.
 H. Shaowei / Automation in Construction 37 (2014) 196–202 201

Bc

fc BcC

hs As fy
1 1

P+ P
he As fy
2 2

The section of composite beam Section strain Section stress

Fig. 6. Model of additional deformation method.

Table 3
Main experimental results from bending tests.
Mu δu
Beam no. My/kN m δy/mm ΔPy/kN Mu/kN m δu/mm ΔPu/kN Pu/kN My δy

PCB-15 247.23 13.74 29.5 393.02 47.74 90.66 295.26 1.59 3.47
CB-16 216.3 12.55 – 366.32 49.58 – – 1.69 3.95
PCB-17 262.31 14.15 27.1 396.27 45 95.87 326.6 1.51 3.18
SCB-18 – – – 389.4 48.94 80.16 313.25 – –
PCB-19 280.25 14.11 34.54 454.38 62.99 162.24 353.89 1.62 4.46
PCB-20 272 14.1 75.81 482 58.86 269.86 453.26 1.77 4.17
SCB-21 215/290 10.22/13 12.98 472.81 61.54 166.35 366.23 –/1.63 –/4.73
SCB-22 210.1/310 11.15/13.75 9.36 494.6 58.53 156.43 398.88 –/1.59 –/4.25
PCB-23 270.13 14.5 36.12 488.75 67.08 195.4 393.47 1.8 4.63
PCB-24 259 4.95 20.79 436.05 28.21 130 321.43 1.68 5.7

To calculate the moment at the center of the concrete equivalent
compression zone, the ultimate flexural capacity of pre-stressed com-
posite beam can be obtained:
 
Mu ¼ f y1 As1 y1 þ f y2 As2 y2 þ f p0 þ Δf p AP y3
Y1 and Y2 are distances from the web plate section center and the
bottom plate of the steel beam in tensile region to the section center
of the concrete flange plate in the compression zone, respectively; Y3
is the distance from the section center of the pre-stressed reinforcement
to that of the concrete flange plate in the compression zone when con-
sidering the quadratic effect.
At the limit state, eccentricity will decrease. The distance Y3 from the
section center of the pre-stressed reinforcement to the section center of
flange plate in the compression zone can then be calculated as
c conducted for ten pre-stressed composite box beams: six via bending
y3 ¼ h−he − −δu :
2 test, three via pure bending test with analog pre-stressed technology
δu is the limit deflection of the pre-stressed composite beam mid-
span, which can be calculated according to Eqs. (3–40) in Ref. [17]. For
the fold-line type composite beam or composite beam with steering
block, there is a relatively small loss in eccentricity and the influence
ð23Þ of quadratic effect can be omitted.
A comparison of the calculation of the limit flexural bearing capacity
Mu and the limit deflection δu of the testing beam according to Eq. (23)
with the test results is shown in Table 4 (j represents the calculated and
t represents the measured values).
From Table 4 it can be seen that the theoretical calculated values
match the test results. The calculated values for the deflection are
large, whereas the limit bearing capacity is small.
4. Conclusion
To determine the mechanical behavior of pre-stressed steel–concrete
composite box beams under bending, bending capacity tests were
ð24Þ

Bc

fc BcC

hs As1 fy1
P+ P
he As2 fy2
The section of composite beam Section strain Section stress

Fig. 7. Stress and strain distribution of the section in the limit state.
202 H. Shaowei / Automation in Construction 37 (2014) 196–202

Table 4 Table 5
Comparison of measured and calculated values of yield deflection and load. Comparison of measured and calculated values of bending capacity.

Testing beam no. My,j/kN m δj/mm My,t/kN m δt/mm My,j/My,t δj/δt Beam no. Mu,j/kN m Mu,t/kN m Mu,j/Mu,t

PCB-15 243.52 14.76 247.23 13.74 0.98 1.07 PCB-15 360 393.02 0.92
PCB-17 250.43 15.02 262.31 14.15 0.95 1.06 PCB-17 365.83 396.27 0.923
PCB-19 266.4 14.59 280.25 14.11 0.95 1.03 PCB-19 429.07 454.38 0.944
PCB-20 255.35 13.63 272 14.1 0.938 0.96 PCB-20 453.4 482 0.941
PCB-23 253.81 14.91 270.13 14.5 0.94 1.03 PCB-23 441.6 488.75 0.90
PCB-24 244.88 5.21 259 4.95 0.94 1.05 PCB-24 418.31 436.05 0.96

in reinforced beams and one via pure bending test without pre-stressing
(Table 5). Test results indicate that pre-stressed composite beam displays
a number of advantages compared with ordinary composite beams. The
flexural capacity of the composite beams was greatly increased using
pre-stressed technology. Equations for calculating the deflection and
elastic bearing capacity of the pre-stressed composite beams under a
serviceability limit state with consideration of interface slip were de-
rived. When considering slip, simplified plastic theory is used to derive
an equation of ultimate flexural capacity of the pre-stressed composite
beam, which can reflect the stress characteristics of the composite
beam.
Results from this derived theoretical equation are compared with test,
and results match experimental data well [17]. This indicates that the
equation can be used to predict real behavior. Calculating equations
were derived for ultimate flexural capacity with consideration of the
pre-stress increment that reflected the stress characteristics of pre-
stressed composite beam ([17]; S.W. [17]). Finally, the calculated values
met precise requirements when compared with the test results. This
study provides a basis for the design theory of pre-stressed steel–concrete
composite box beams, and also provides the foundation for further im-
provements in design theory and analysis of pre-stressed composite box
beams.
Acknowledgement
The authors gratefully acknowledge that the work is supported by
China National Funds for Distinguished Young Scientists (No. 513250
276) and the National Natural Science Foundation of China (No. 51279
111). Thanks for the help.
References
[1] J. Brozzetti, Design development of steel–concrete composite bridges in France,
J. Constr. Steel Res. 55 (2000) 229–243.
[2] Liu Hanbing, Liu Wenhui, Zhang Yunlong, Calculation analysis of shearing slip for
steel–concrete composite beam under concentrated load, Appl. Math. Mech. 26
(6) (2005) 677–682.
[3] Nie Jianguo, Hu Shaowei, Experiment on steel–concrete composite T-beams under
simple torsion, J. Harbin Univ. Archit. Civ. Eng. (5) (1999) 101–105.
[4] T.C. Hsu, Nonlinear analysis of concrete torsional members, Struct. J. 1 (1991)
105–116.
[5] M. Nakashima, T. Matsumiya, K. Suita, Full-scaletest of composite frame under large
cyclic loading, J. Struct. Eng. 133 (2) (2007) 297–304.
[6] Andrea Dall'Asta, Luigino Dezi, Nonlinear behavior of externally prestressed composite
beams: analytical model, J. Struct. Div., ASCE 124 (5) (1998) 588–597.
[7] M. Samaan, K. Sennah, J. Kennedy, Distribution factors for curved continuous composite
box-grider bridges, J. Bridg. Eng. 10 (6) (2005) 678–692.
[8] Nie Jianguo, Tao Muxuan, Huang Yuan, Research advances of steel–concrete composite
structural system, J. Build. Struct. 31 (5) (2010) 71–80.
[9] Zhao Huiling, Yuan Yong, Experimental studies on composite beams with high-
strength steel and concrete, Steel Compos. Struct. 10 (5) (2010) 1–3.
[10] Chen Shiming, Gu Ping, Load carrying capacity of composite beams pre-stressed
with external tendons under positive moment, J. Constr. Steel Res. (61) (2005)
515–553.
[11] Hu Shaowei, Nie Jianguo, Study on the torsional behaviors of steel–concrete composite
beams, Building Structure 4 (182) (1999) 38–40.
[12] A. Mendelson, Plasticity Theory and Application, 5Pergamon Press, 1975. (4).
[13] Hu Shaowei, Nie Jianguo, Simpled analysis of composite grider under loadings—by
the progressor method, Mech. Pract. 21 (3) (1999) 16–18.
[14] Hu Shaowei, Nie Jianguo, Xiong Hui, Torsional experiment and analysis on composite
steel–concrete beams, J. Build. Struct. 27 (4) (2006) 103–109.
[15] Hu Shaowei, Nie Jianguo, Zhu Linsen, Luo Ling, Complete history analysis on torsional
performance of composite steel–concrete beams, Chin. J. Comput. Mech. 21 (4) (2004)
435–454.
[16] shaowei Hu, Wang Hongxia, Study on 3D finite bodies containing cracks using the
finite element method of lines, Acta Mech. Solid. Sin. 19 (4) (2003) 1–12.
[17] S.W. Hu, Research on the Torsional Behavior of Composite Steel–Concrete Beams,
Tsinghua University, 1999.