Failure Mode Prognosis of Highway Grillage Girder Bridge Considering Shear

Resistance of Diaphragms
Zhi SUN
Research Fellow
Tongji University
Shanghai, China

Yi ZHANG
Former Master Student
Tongji University
Shanghai, China

Yue AN
Master Student
Tongji University
Shanghai, China

sunzhi1@tongji.edu.cn

tarky2132@sina.com

ayxjlz@163.com

Zhi SUN, born 1977, received his PhD

degree from the Hong Kong Univ. of
Science and Technology. He majors in
structure condition monitoring and
performance assessment, and etc.

Yi ZHANG, born 1983, received his

Master degree from Tongji Univ. His
research interest is reliability analysis
of bridge safety considering strength.

Yue AN, born 1987, received his

Bachelor degree from Sichuan Univ.
He majors in reliability analysis of
bridge safety considering ductility

Summary
This paper presents a study on failure mode prognosis for a highway grillage girder bridge considering the shear resistance capacity of diaphragms. The bridge concerned is composed of 5 main girders and 3 diaphragms. The plastic yielding of the main girders considering their moment capacity
and the diaphragms considering their shear capacity is concerned. The incremental analyses are
conducted to evaluate bridge vehicle-carrying capacity. During the incremental analysis, the ductile
or brittle shear failures of the diaphragm are considered. The results tell that the shear resistance
capacity of diaphragm will induce the variation of structure failure mode under the traffic loads and
should be designed to be of more redundancy to avoid the local failure, especially if uncertainty is
considered.
Keywords: grillage, integrity, shear failure, uncertainty, ductility

1.

Introduction

Grillage girder bridge is a generally adopted structure type for short or medium-span highway
bridges. Diaphragms, which transversely connect the main girders to form the grillage structure, are
important components to enhance structure capacity to carry the traffic loads [1, 2, 3]. Recent
damage survey on highway bridges showed that the diaphragm crack and fracture was frequently
observed. The reason may be due to that the length/depth ratios of the diaphragms are generally
around 1~3 and thus the shear effect will dominate its failure but was not appropriately concerned
of in design. This will induce unpredictable structural global and local failure mode and should be
investigated.
This paper investigates the influence of shear-resistance capacity of diaphragms on the load
carrying capacity of a highway grillage girder bridge. The bridge is composed of 5 main girders
with T-shape cross section and 3 diaphragms. A finite element model of the bridge is established
and the plastic yielding of the main girders considering their moment capacity and the diaphragms
considering of the shear capacity is concerned.

2.

Considered Engineering Structural System

2.1

The Bridge and Its Modelling

The bridge considered in this study is a simply-supported reinforced concrete grillage girder bridge
as shown in fig.1. The bridge is composed of 5 girders and 3 diaphragms. Each girder is 20 meters
long, 1.6 meters wide and 1.3 meters high. The cross section of each girder is T shape. The girder
cross section and girder reinforcement are shown in fig. 1. The girder is fabricated using concrete

C30 and rebar HRB335. Bridge pavement is asphalt concrete with an average thickness of 11cm.
The design vehicle load is highway Class-II vehicle load.
7. 0

1. 6

1. 6

1. 6

1.3

1. 3

0. 5

220
832

1. 6

0.18

Fig.1 Bridge cross section dimension and reinforcement layout

For structure analysis, a 25-nodes grillage girder model is setup (as shown in fig. 2). The main
girders and diaphragms are all modeled to be grid elements. The grid element is a kind of
degenerated beam element with 3 degrees of freedom at each node. The three degrees of freedom
are vertical displacement, vertical bending angle and torsion angle.

Fig. 2 Grid girder analytical model of the structure

2.2

The Lane Load and Its Modelling

For a highway bridge, its loads include dead load, live load, impact load and other special loads.
Dead load, which is the load due to self weight of main girders and subsidiary structures, is
modelled as uniformly distributed forces along the bridge. The nominal gravity density for
reinforced concrete is set to be 25.0kN/m3 in this study for self-weight calculation. Live load refers
to the load due to moving vehicles. In this study, according to Chinese Code [4], the live load is
simplified to be a uniformly distributed force with the magnitude of qk and a concentrated force
with the magnitude of Pk (as shown in fig. 3). Since the bridge is designed for highway Class-II
vehicle load, the nominal value of qk and Pk are 7.875 kN/m and 180kN, respectively. To calculate
the maximum load effect due to live load, the longitudinal and transverse influence lines for the
bridge structure are computed. The most detrimental loading positions in two directions are thus
determined accordingly. Figure 3 shows the live load placement for the worst load effect for
structural collapse.
50

180

130

180

pk
qk

1#

Longitudinal placement of live load

2#

3#

4#

5#

Transverse placement of live load

Fig. 3 Live load position in two directions

The impact load is a pseudo-static load to concern the impact effect during the vehicle passes over
the bridge. It depends on bridge dynamic characteristics, vehicle dynamic characteristics, vehicle
speed, and road roughness. According to Chinese Code [4], the nominal impact load can be

calculated as a product of the nominal live load times an impact factor. In this study, the impact
factor is set to be 1.1776 according to the fundamental vertical bending frequency of the bridge.

3.

Resistance Model

Considering the geometric layout of the bridge and the dimension of its compoennts, the main
girders are regarded to be bending-dominant components and the diaphragms are treated to be
shear-dominant components. The moment resistance of the cross section of the main girder and the
shear resistance of the diaphragm were computed. No force interaction is concered for the yiedling
analysis of the cross-sections in this study.
3.1

Main Girder Resistance Model Considering Bending

As presented above, the materials used for the bridge girder are concrete C30 and rebar HRB335
and their layout is shown in fig. 1. The adopted material constitutive models for the concrete and
rebar are the models recommended by the Chinese Code [4]. The moment-curvature relationship is
calculated using the strain incremental approach under the basic assumption of plain cross section
after deformation. The result (as shown in Figure 4) tells that the ultimate bending moment of the
reinforced concrete main girder is 2358.13KNm.
2,400

Moment(kNm)

2,000
1,600
1200
800
400
0
0

Curvature(rad/km)

Fig. 4 Moment curvature curve of the main girder

3.2

Diaphragm Resistance Model Considering Shear

The shear resistance of the the diaphragm is computed according to the configuration of the
diaphram (as shown in figure 5) and the shear strength of the connecting steel plate. Considering
that the shear force-displacement curve is of a plastic yielding platform but the length of the
platform is quite short, two idealized shear failure models for the diaphragm are considered during
structure incremental analysis. The models adopted are the shear plastic hinge model and the brittle
shear failure model. The computed ultimate shear force of the diaphragm is 204.9 MN.

Fig. 5 Detailing of the diaphragm

4.

Bridge Capacity and Failure Mode Analysis

Bridge ultimate load carrying capacity is analyzed using the incremental method [5]. The nominal
live load defined in the code is applied on the bridge model with a variable load multiplier .
Structure limit states are defined to be that some local or global mechanisms are formed. The load
multiplier reflects the load carrying capacity of the bridge system, and shows the generalized
bridge structure resistance.

(a)
(b)
Fig. 6: Bridge failure mode considering a) no shear yielding and b) ductile shear yielding of the
diaphragms;
Employing the incremental method, the load multiplier considering the idealized plastic yielding of
the main girder with infinity ductility is computed. The result is 2.952. Figure 6a is the structure
failure mode concerning the formation of global mechanism. If the plastic shear yielding of the
diaphragm with infinity ductility is concerned, the computed load multiplies is 2.951. Figure 6b
shows the structure failure mode. As the structure failure is still a global failure and no local
fracture and mechanism occur, bridge load carrying capacity is not reduced.
If the brittle shear failure of the diaphragm is concerned, the load carrying capacity of the bridge
will vary with the shear resistance strength of the diaphragms. Figure 7 shows the variation of the
load multiplier along with the change of the ultimate shear foce of the diaphragm. As shown, for the
F

3.00
B

(a) OA ( Pv

109.32 KN)

Load Multiplier

2.50

D
C

2.00
1.50
1.00
0.50
0.00

(b) BC ( 109.33

(c) DE ( 149.53

Pv

149.52 KN)

O
0

50

100

150
200
250
300
350
400
Shear Resistance of Diaphgram (KN)

450

500

Pv 174.25 KN) (d) EF ( 174.26 Pv 305.85 KN)

(e) FG( Pv 305.86 KN)
Fig. 7: Bridge load carrying capacity and failure mode considering brittle shear failure of diaphragm

bridge studied, if the ultimate shear force of the diaphragm is small, the brittle shear failure of the
diaphragm will occur. This will destroy structure integrity and reduce structural capacity under
other possible loading cases. If the ultimate shear force is greater than 305.86 kN, no local fracture
will occur and the bridge will finally lose its load carrying capacity after the formation of a global
mechanism.
Table 1 Statistical Property of Load, Material, and Dimension Variables
Random variables
Self weight
Maximum live load
Impact factor
Concrete C30 strength
Rebar HRB335 strength
Rebar area
Girder width
Girder effective depth
Girder ultimate load

Distribution type

Mean/nominal ( ) Variation ( )
1.0212
0.0462
0.6861
0.1569
1.1776 (mean)
0.0428
1.5012
0.1773
1.0849
0.0719
1.0000
0.0350
1.0013
0.0081
1.0124
0.0229
2358.13 (mean) 0.0740

Normal
Extreme I
Normal
Normal
Normal
Normal
Normal
Normal
Normal

If the variable uncertainty during the numerical analysis is considered, the vehicle carrying capacity
of the bridge is evaluated using the reliability index based on Monte Carlo simulations (MCS). The
random variables considered are shown in Table 1 and their statistical properties are determined
according to the Chinese Standard [6]. The reliability index for bridge vehicle carrying safety can
then be computed by checking the statistical distribution of the MCS computed load multipliers.
Figure 8 shows the MCS computed results considering the ductile shear yielding and brittle shear
fracture of the diaphragms. For the ductile shear yielding of the diaphragms, the uncertainty of
random variables will induce the variation of the failure mode. However, as no local mechanisms
will be formed, structure integrity is maintained. The Kolmogorov-Smirnov (K-S) test tells that the
statistical distribution of in this case can still be treated as a normal distribution. For the brittle
shear fracture of the diaphragms, the possibility of forming local mechanisms makes the vehicle
carrying capacity of the bridge unreliable. The normality of the statistical distribution of is
rejected by the K-S test.
700

600

MCS results
500

500

Number of Samples

Number of Samples

MCS results

Fitted curve

600

400
300
200

300

200

100

100
0
2.0

400

2.5

3.0

3.5
Load Multiplier

4.0

4.5

5.0

0
2.0

2.5

3.0
3.5
Load Multiplier

4.0

4.5

(a)
(b)
Fig. 8: Statistical distribution of load multiplier for the grillage bridge considering uncertainty with
a) ductile shear yielding and b) brittle shear fracture of the diaphragms;

5.

Conclusions

This paper presents a study on failure mode prognosis for a highway grillage girder bridge consider-

ing the shear resistance capacity of diaphragms. The ductile and brittle shear failures of the diaphragm are considered. The results tell that the variation of shear resistance capacity of diaphragm
will induce the variation of failure mode of the grillage girder bridge under traffic loads. If the ductility of the diaphragms after the yielding cannot be ensured, strong shear resistance strength of the
diaphragm must be ensured to avoid the local fracture and the loss of structure integrity. Further
study to check the ductility requirement and considering the force interaction during the failure is in
need.

Acknowledgement
This research was supported by the 973 program of Ministry of Science and Technology of China
(Grant No. 2013CB036305).

References
[1]
[2]
[3]
[4]

[5]
[6]

EAMON C D, NOWAK A S. (2004). Effect of Secondary Elements on Bridge Structural

System Reliability Considering Moment Capacity. Structural Safety, 26:1, 29-47.
EAMON C D, NOWAK A S. (2005). Effect of edge stiffening and diaphragms on the
reliability of bridge girders. Journal of Bridge Engineering, 10(2): 206-214.
TIN-LOI F. (1995). Plastic limit analysis of plane frames and grids using GAMS. Computers
& Structures, 54(1): 15-25.
CHINESE MINISTRY OF COMMUNICATIONS. (2004). Code for Design of Highway
Reinforced Concrete and Prestressed Concrete Bridges and Culverts, China Communications
Press, Beijing, China.
JIRASEK, M., BAZANT, Z. P., Inelastic Analysis of Structures, John Wiley & Sons, Ltd,
2002.
CHINESE MINISTRY OF CONSTRUCTION. (1999). Unified Standard for Reliability
Design of Highway Engineering structures, China Planning Press, Beijing, China.