SEISMIC VIBRATION CONTROL OF IZADKHAST BRIDGE USING VISCOUS DAMPERS

Abstract

Present study addresses the effectiveness of viscous dampers (VDs) in reducing the
seismic responses of Izadkhast Bridge under earthquake ground motions. With the length of 485 m,
Izadkhast Bridge is the longest box girder bridge in Iran and is located in Isfahan-Shiraz
railway. The bridge is installed with VDs at the two ends. The Finite element model of the
bridge is developed. Five pairs of representative earthquake records are selected and scaled
using the earthquake code and applied to the model. Nonlinear seismic analyses of the structure
without and with VD's are performed. Comparison of the
results clarifies VD's effectiveness on the seismic response reduction of the bridge. Sensitivity
analyses are performed to demonstrate the effects of damper parameters on structural responses.

Full text

H.R. Anajafi, A.K. Ghorbani-Tanha and M. Rahimian
1
COMPDYN 2011
3rd ECCOMAS Thematic Conference on
Computational Methods in Structural Dynamics and Earthquake Engineering
M. Papadrakakis, M. Fragiadakis, V. Plevris (eds.)
Corfu, Greece, 25-28 May 2011
SEISMIC VIBRATION CONTROL OF IZADKHAST BRIDGE USING
VISCOUS DAMPERS
H.R. Anajafi1, A.K. Ghorbani-Tanha1, and M. Rahimian1
1School of Civil Engineering, University of Tehran
P.O. Box 11155-4563, Tehran, Iran
E-mail: hamidanajafi@Tazand.com, {ghtanha, rahimian}@ut.ac.ir
Keywords: Vibration control, Izadkhast Bridge, Viscous damper, Passive control, Earthquake
Abstract. Present study addresses the effectiveness of viscous dampers (VDs) in reducing the
response of Izadkhast Bridge under earthquake ground motions. With the length of 485 m,
Izadkhast Bridge is the longest box girder bridge in Iran and is located in Isfahan-Shiraz
railway. The bridge is installed with VDs at the two ends. The Finite element models of the
bridge are developed. Five pairs of representative earthquake records are selected and
scaled using the earthquake code and applied to the models. Nonlinear seismic analyses of
the structure without and with VDs are performed and the results are reported. Comparison
of the results clarifies VDs effectiveness on seismic response reduction of the bridge.
Sensitivity analyses are performed to demonstrate the effects of damper parameters on
structural response.

H.R. Anajafi, A.K. Ghorbani-Tanha and M. Rahimian
2
1 INTRODUCTION
The basic function of passive energy dissipation devices when added to a structure is to
absorb and dissipate a portion of the input energy, thereby reducing energy dissipation
demand on primary structural members and minimizing possible structural damage. Serious
efforts have been devoted to the development and utilization of passive energy dissipation
devices [1]. Viscous dampers (VDs) are a kind of these devices which significant efforts have
been directed toward their application for structural vibration control. In VDs, energy
dissipation occurs via conversion of mechanical energy to heat as a piston deforms a thick,
highly viscous substance [1]. In present study, the effectiveness of VDs on response
reduction of Izadkhast Bridge under earthquake ground motions is investigated.
With the length of 485 m, Izadkhast Bridge is the longest box girder bridge in Iran. This
bridge is located in the central part of Iran, spanning Izadkhast valley on the railway line
between Isfahan and Shiraz. The bridge is composed of five 77 m spans in the middle, two
side spans having the length of 60 m and 40 m, six piers and two abutments, as shown in Fig.
1. The cross section of piers is shown in Fig. 2. With the width of 6.6 m, the deck is
composed of two 4.5 m high box girders (Fig. 3). Seismic considerations were considered in
the design of the bridge due to its location. The bridge is equipped with four 1000 KN VDs
(stroke 250 mm) longitudinally directed at both ends (Fig. 4). According to the manufacturer
catalogs and design documents, the governing equation of the dampers is a
cvf , where f is
force, c is the damping coefficient, v is the velocity, and 15.0

a [2, 3]. Four 800×800×219
mm rubber bearings are placed at the top of each pier and two 800×800×357 mm rubber
bearings at the top of each abutment (Fig. 5).
The Finite element model of the bridge is developed. Five pairs of representative
earthquake records are selected and scaled using earthquake code and applied to the model.
Nonlinear seismic analyses of the structure without and with VDs are performed and the
results are reported. Comparison of the results clarifies VDs effectiveness on seismic
response reduction of the structures.
44
Figure 1: Izadkhast Bridge and its support conditions on the piers and abutments [2]

H.R. Anajafi, A.K. Ghorbani-Tanha and M. Rahimian
3
11.9 to 44 m2.5 m
Figure 2: View and Cross section of piers [2]
Figure 3: Cross section of bridge deck at piers[2]

H.R. Anajafi, A.K. Ghorbani-Tanha and M. Rahimian
4
Figure 4: VDs placement [2]
(a) (b)
Fi
g
ure 5: Details of rubber bearin
g
s (a) at the top of piers, (b) at the top of abutments [2]
2 MODELLING AND ANALYSES
2.1 Finite element model of the bridge
A finite element model of the bridge is developed in PERFORM-3D [4]. Constraints are
applied to restrict the deck from moving horizontally at Piers 2, 3, 4, and 5 and laterally at all
piers. The displacement capacity of the rubber bearings are calculated as following
TtgVx


(1a)
a
T
tg
a
T
tg
a
T 9.0:2.0;7.0:2.0

(1b)
Where T is the effective thickness of the rubber bearing which according to the
manufacturer catalogs is 144 mm for piers and 252 mm for abutments; and a is the minimum

H.R. Anajafi, A.K. Ghorbani-Tanha and M. Rahimian
5
dimension of the cross section of rubber bearings. Under seismic loading, an increase of 50%
is applied to displacement capacity. As a result, the displacement capacity for the abutment
bearings is 220 mm while this value for the pier bearings is 150 mm. The rubber bearing are
modeled as non-linear springs whose initial stiffness are TGAKs

, where G is shear
modulus, A is the area and T is the effective thickness of bearings. Under seismic loading, the
shear module of bearings is considered two times greater than its common value. For the
definition of plastic hinge in piers, the famous available models are employed. The length of
plastic hinge, Priestley relation is used [5,6]
),(044.0022.008.0 MPammdfdfLL blyeblyep 
(2)
Where LP is the distance between critical section and inflection point of the member; db is the
diameter of longitudinal bars and fy is the yield stress. According to design documents, the
expansion joint between deck and abutments has a width of 250 mm which is taken into
account in the model. If the longitudinal displacement of the deck is greater than this value,
the deck knocks the abutments.
2.2 Earthquake records used
For the excitation of the bridge, five pairs of earthquake records are chosen (Table 1). These
records are scaled according to UBC [7] and then applied to the structure.
Table 1: Earthquake records used
No. Record Name Date PGA(g)
1 IZMIT1 17/8/1999 0.2195
2 IZMIT2 17/8/1999 0.1521
3 ELCENTRO1 19/5/1940 0.2148
4 ELCENTRO2 19/5/1940 0.3129
5 K0BE1 16/1/1995 0.5985
6 K0BE2 16/1/1995 0.8213
7 NORTHRIDGE1 17/1/1994 0.493
8 NORTHRIDGE2 17/1/1994 0.8283
9 SANFRANCISCO1 18/10/1989 0.056
10 SANFRANCISCO1 18/10/1989 0.105
3 RESULTS
3.1 Uncontrolled bridge
Dynamic analyses of the bridge without VDs under scaled earthquake time histories are
conducted. The first natural period of the bridge is 2.5 sec. The results show that the
maximum longitudinal displacement of the deck occurs under scaled Izmit earthquake
(PGA=0.546g) which is 510 mm. The time histories of the longitudinal displacement
responses under two components of Izmit (Izmit 1 & 2) are shown in Fig. 6. The moment-

H.R. Anajafi, A.K. Ghorbani-Tanha and M. Rahimian
6
curvature diagram for pier P6, as critical pier, is shown in Fig. 7. It is clear that the piers will
exhibit nonlinear behavior. Fig. 8 shows the energy response of the bridge. A significant
portion of the energy input to the structure is dissipated with both inelastic hysteretic
mechanisms and inherent viscous damping. In this case, the expansion joint will be closed
and the deck will knock the abutments which can cause serious damages.
DIS P. R E S P ONSE
-30
-20
-10
0
10
20
30
0 2 4 6 8 10121416
Tim e( s ec )
Dis
p
.
(
cm
)
Figure 6: Longitudinal displacement response of the uncontrolled bridge under Izmit earthquake records
Mome nt-C u rva tur e
-100000
-80000
-60000
-40000
-20000
0
20000
40000
60000
80000
100000
-0.01 -0.005 0 0.005 0.01 0.015
(rad/m)
M(KN.M)
Figure 7: Moment-curvature diagram for pier P6 (shortest pier with the height of 11.9 m) for the uncontrolled
bridge
WI THOUT DAMPER
0
5000
10000
15000
20000
25000
30000
0 2 4 6 8 10 12 14 16
Tim e( s ec)
Energy(KN .M)
dissipated inelastic energy
beta-K viscous energy
alpha-M viscous energy
strain energy
kinetic energy
total energy
DIS P .R E S P ONS E
-30
-20
-10
0
10
20
30
40
50
60
0 2 4 6 8 10121416
Tim e(s e c)
Disp.
(
cm
)

H.R. Anajafi, A.K. Ghorbani-Tanha and M. Rahimian
7
Figure 8: Energy response of the bridge under scaled Izmit earthquake
3.2 Controlled bridge
As mentioned before, the bridge is equipped with four 1000 KN VDs at both ends.
Dynamic time-history analyses of the controlled bridge show that the maximum displacement
of the bridge reduces to 380 mm but it is still greater than the width of expansion joints and
the deck knocks the abutments (Fig. 9). However, a significant portion of the energy input is
absorbed and dissipated by the dampers which reduce the nonlinear deformation of the
structure. The moment-curvature diagram for pier P6 is shown in Fig. 10.
DIS P .R E S P ONS E
-30
-20
-10
0
10
20
30
40
0 2 4 6 8 10 12 14 16
Tim e (s e c )
Disp.(cm)
Figure 9: Longitudinal displacement response of the bridge fitted with four 1000 KN VDs under Izmit
earthquake
Moment-Curvature
-100000
-80000
-60000
-40000
-20000
0
20000
40000
60000
80000
100000
-0.003 -0.002 -0.001 0 0.001 0. 002 0.003 0.004 0.005 0.006 0. 007
(rad/ m)
M(KN.M)
Figure 10: Moment-curvature diagram for pier P6 for the case that the bridge is equipped with four 1000 KN
VDs

H.R. Anajafi, A.K. Ghorbani-Tanha and M. Rahimian
8
DIS P .R E S P ONS E
-2 0
-1 0
0
10
20
30
0 2 4 6 8 10121416
Tim e (s e c )
Disp.(cm)
Figure 11: Longitudinal displacement response of the bridge fitted with four 2500 KN VDs under Izmit
earthquake
Moment-Curvature
-60000
-40000
-20000
0
20000
40000
60000
80000
100000
-0. 0015 -0.001 -0.0005 0 0.0005 0. 001 0. 0 015 0. 002 0. 0025
(rad/m)
M(KN.M)
Figure 12: Moment-curvature diagram for pier P6 for the case that the bridge is equipped with four 2500 KN
VDs
Proper VDs will prevent structural damages and does not let the deck knock the
abutments. A trial and error procedure employed and finally it was concluded that if 1000 KN
dampers are replaced by 2500 KN dampers, then the maximum longitudinal displacement
reduces to 240 mm and piers will remain elastic (Figs. 11 and 12). These dampers are more
expensive and off course apply higher values of reaction forces to the abutments which
should be taken into consideration in design procedure.
4 SUMMARY AND CONCLUSIONS

H.R. Anajafi, A.K. Ghorbani-Tanha and M. Rahimian
9
 The inclusion of VDs enhances the seismic behavior of the bridge and the dampers
dissipate a significant portion of the energy input to the bridge. This reduces the hysteretic
energy dissipated by the sub-structural members and decreases damage to the structure and
is favorable for earthquake resistant design.
 For the case that the bridge is fitted by four 1000 KN VDs, the maximum longitudinal
displacement under design earthquake is 380 mm which is more than the displacement
capacity of the damper and expansion joint width and the deck knocks the abutments.
 To overcome the above-mentioned problems and improve seismic performance of the
bridge, 2500 KN dampers are recommended to be used. In this case the maximum
displacement reduces to 240 mm.
ACKNOWLEDGEMENT
The authors are grateful to Tazand Co. for providing them with the bridge design documents
and data.
REFERENCES
[1] T.T. Soong, G.F. Dargush, Passive energy dissipation systems in structural engineering,
John Wiley & Sons, 1997
[2] Tazand Consulting Engineers, Design documents of Izadkhast Bridge, 2006 (In Persian).
[3] http://www.fip-group.it
[4] CSI PERFORM-3D, V4.0.1, Computers and Structures Inc, Berkeley, California, Release
2006, Components and Elements.
[5] Caltrans, Seismic Design Criteria, V1.4, June 2006.
[6] T. Paulay, M.J.N. Priestley, Seismic design of reinforced concrete and masonry buildings,
John Wiley & Sons, 1992.
[7] Unified Building Code, 2006.