This action might not be possible to undo. Are you sure you want to continue?

# th

The 14 World Conference on Earthquake Engineering October 12-17, 2008, Beijing, China

Nonlinear Seismic Response Analysis and Seismic Performance Evaluation of A Reinforced-Concrete Cable-Stayed Bridge Tower Using IDA Method

Jiao chi-yu,

1 2

12

Li jian-zhong, Fan li-chu,

3

3

PhD, Dept. of Bridge Engineering, Tongji University, Shanghai, CHINA Lecturer, College. of Civil Engineering, Beijing University of Civil Engineering and Architecture, Beijing, 3 Professor, Dept. of Bridge Engineering, Tongji University, Shanghai, CHINA Email: jiaochiyu2000@yahoo.com.cn

ABSTRACT : Nonlinear Seismic Response analysis and Seismic performance evaluation for cable-stayed bridge are very important, especially for the tower of it. Based on the recent researches, A simplified single tower finite element model was developed to modelling the seismic response under the longitudinal excitation in this paper. In order to investigate the nonlinear seismic response of the tower, a fiber flexural element was adopted and the P-delta Effect was considered in present study. On the other hand, the incremental dynamic analysis (IDA) of this Tower has been done herein, in order to evaluate the seismic performance of the tower more reasonably. The results show the damage zone (or the plastic zone) for this tower mainly concentrated in three regions. According to IDA analysis and the summary curves of it, seismic damage state evaluation can be more reasonable which has more probabilistic characteristic. Finally, some suggestions on the definition of the global ductility factor for a complex structure like the tower of cable-stayed bridge were given herein. KEYWORDS: 1. INTRODUCTION In 1990’s, The 3D nonlinear seismic response analysis of cable-stayed bridge was studied by Aly S. Nazmy and Ahmed M. Abdel-ghaffar [1990]. In that paper the author paid more attention to the geometric nonlinearities, such as cable-sag effect, axial force-bending moment interaction and the large displacements. The conclusion showed that for the cable-stayed bridge with the longer center span (>=610m) including the geometrical nonlinearities would reduce the response significantly under uniform earthquake excitation. It also showed the dead load deformed state was the basis of the reasonable nonlinear seismic response analysis. With the development of the computer science, more possible effects of nonlinearities can be included in the analysis. For instance, Wei-Xin Ren [1999] has considered the material nonlinearity due to the stiffening steel girder yielding. The 2D Plane beam element was adopted in order to reduce the computing time, and the maximum equivalent plastic strain ratio was proposed to evaluate the elastic-plastic seismic damage for the local element in his research. However, his research still assumed the cable, and reinforced-concrete tower remained in elastic state under the earthquake action. As the performance-based design philosophy were followed, the seismic performance of the tower of long-span cable supported bridges need evaluating and designing in details, as well as, the damage states for different performance levels from fully operational to collapse should be clarified. Kazuo Endo, [2004] and Shehata Eldabie ABDEL RAHEEM, [2003] have done some researches at the tower considering the material nonlinearity. In their researches the steel tower was modeled by shell and fiber elements, the strength and damage progress characteristics were obtained, and the acceptable ductility capacity for the steel tower structure exceeding the elastic limit was proposed. A fiber beam-column element is a good tool to investigate the spread of plastic zone in the tower under the great earthquake ground motion. So far, the nonlinear seismic response of RC tower in cable supported bridge has not been carried out in details, which will be studied in this paper. On the other hand, the incremental dynamic analysis (IDA) which was carried out by Dimitrios Vamvatsikos Nonlinear Seismic Response, Seismic Performance Evaluation, Cable-Stayed Bridge, Reinforced-Concrete Tower, IDA

th

The 14 World Conference on Earthquake Engineering October 12-17, 2008, Beijing, China

and Allin Cornell [2003] gave us a good thought to study the structure performance under seismic loads, especially for the nonlinear model, which can’t be predicted well by Nonlinear static pushover analysis. Therefore, in this paper fiber elements were used to model the tower and the IDA method was adopted to investigate the seismic performance of the tower. 2.COMPUTING MODEL AND NOLINEARITY CONSIDERATION 2.1. Computing Model The tower of a cable-stayed bridge can be divided into to two parts, the anchorage zone and the pylons. Due to the high circumferential compression stress and the complex constitutive model for the anchorage zone of the tower, the elastic beam elements were used in this study. For the pylons, which made from high strength concrete reinforced by steel bars, the fiber beam-Column elements supported by the Open System for Earthquake Engineering Simulation (Opensees) [2007] were used. Three different material constitutive models were used in a section, the confined Mander model [1988] (modeling confined concrete), the unconfined Mander model (modeling cover concrete), and the Menegotto Pinto model (modeling the longitudinal steel bars). A typical cable-stayed bridge tower with the height of 235m and its critical section used in this paper are shown in Fig.2.1. The constitutive model of Mander model and the constitutive model of the steel bar are illustrated in Fig. 2.2. The parameters of the model are shown in Table 2.1. In longitudinal seismic response analysis, a simplified tower model suggested by Yan Hai-quan [2007] has been modified and applied in this paper. In order to consider the inertia force of the main girder more reasonably, the main girder and the cables are modeled as an oscillator linked to the highest anchorage point of the tower. The weight of the main girder is taken as the mass of the oscillator, the longitudinal stiffness of the oscillator is calculated by the period of the floating mode of the system made up of cables and the beam. Eqn. 2.1 is used to calculate this stiffness K1.

⎛ 2π ⎞ K1 = ⎜ ⎟ × mb ⎝ TF ⎠

B

2

(2.1)

For the soil is stiff enough, and the tower is fixed at the bottom, so the soil-structure interaction is not included herein.

A SectionG

SectionF

SectionE SectionD

SectionC

Fig. 2.1 FEA model division of the tower

Fig. 2.2 Constitutive model of the confined concrete and steel bar

Material Confine concrete Cover concrete

Table 2.1 Material parameters of section 15 f’y (kPa) E B ε’cc f’cu(kPa) ε’cu Material Confine 3.838E+04 0.004 3.374E+04 0.011 3.35E+05 2.00e8 0.0003 concrete f’cc(kPa)

3.240E+04 0.002 2.668E+04 0.004

th

The 14 World Conference on Earthquake Engineering October 12-17, 2008, Beijing, China

2.2 Nonlinearity Consideration The material nonlinearities are considered as above. In order to consider the geometric nonlinearity of the tower, the nonlinear history analysis including P-delta effect has been done after the dead load analysis in this paper. 3. NONLINEAR SEIMIC DEMAND ANALYSIS To investigate the seismic performance of the tower, ten ground motion records have been adopted, which are taken out from the pacific earthquake engineering research center (PEER) national ground acceleration (NGA) database. To reduce dispersion of the seismic response, the fields where these earthquakes occurred all belong to the type I, according to the Specifications of Earthquake Resistant Design for Highway Engineering in China. The epicenter distance of these waves are less than 20kM as shown in table 3.1 The elastic acceleration response spectra and displacement spectra of the ten waves can be seen in Fig. 3.1. On the other hand, for IDA analysis, recent researches have shown that taking the Sa(T1,ξ) as an Intensity Measure(IM) for scaling, can be more effective than PGA as an IM, in reducing the dispersion of the seismic response of the structure. Therefore, for this study the Sa (T1, 0.03) has been scaled from 0.001g to 0.035g, and the incremental step of 0.0029g is adopted. The damping ratio 0.03 is used here mainly considering that the main girder and cables are made of steel. Table 3.1 Ground motion records used for IDA analysis Record Record/ PGA Sa EpiD Earthquake Station Tp M (g) (km) ID Component (T1,0.03) Helena, Montana 2022 Carroll HELENA/ wave1 0.15 0.140 0.0003 6.000 6.310 A-HMC180 1935/10/31 18:38 College Helena, Montana 2022 Carroll HELENA/ wave2 0.173 0.280 0.00065 6.000 6.310 A-HMC270 1935/10/31 18:38 College Helena, Montana 2229 Helena HELENA/ wave3 0.047 0.080 0.00002 6.000 6.310 1935/10/31 19:18 Fed Bldg B-FEB000 Helena, Montana 2229 Helena HELENA/ wave4 0.041 0.060 0.00002 6.000 6.310 Fed Bldg B-FEB090 1935/10/31 19:18 San Francisco 1117 Golden SANFRAN/ wave5 0.095 0.260 0.0001 5.280 11.130 Gate Park GGP010 1957/03/22 19:44 San Francisco 1117 Golden SANFRAN/ wave6 0.112 0.220 0.0002 5.280 11.130 1957/03/22 19:44 Gate Park GGP100 Central Calif 1028 Hollister CTRCALIF/ wave7 0.041 0.300 0.00012 5.000 8.010 City Hall B-HCH181 1960/01/20 03:26 Central Calif 1028 Hollister CTRCALIF/ wave8 0.063 0.260 0.00016 5.000 8.010 City Hall B-HCH271 1960/01/20 03:26 Hollister 1028 Hollister HOLLISTR/ wave9 0.072 0.320 0.0002 5.500 18.920 1961/04/09 07:25 City Hall C-HCH181 Hollister 1028 Hollister HOLLISTR/ wave10 0.075 0.420 0.0004 5.500 18.920 City Hall C-HCH271 1961/04/09 07:25

T3=0.83s T2=2.33s T1=13.53s w ave1 w ave2 w ave3 w ave4 w ave5 w ave6 w ave7 w ave8 w ave9 w ave10 10 T3=0.83s T2=2.33s T1=13.53s w e1 av w e2 av w e3 av w e4 av w e5 av w e6 av w e7 av w e8 av w e9 av w e10 av

0.1

1

0.01

Sd(cm)

Sa(g)

0.1

1E-3 0.1

Period T(s)

1

10

0.01 0.1

Period T(s)

1

10

Fig. 3.1 Elastic response spectra of the ten ground motion records

th

The 14 World Conference on Earthquake Engineering October 12-17, 2008, Beijing, China

For evaluating the damage of the tower, deformations of the critical position are recorded during every time history analysis. The curvature of each integrator point of the element is obtained. Due to the space limitation, only seismic responses excited by wave2, wave3, wave7, wave9 are shown in Fig. 3.2. The displacements in the longitudinal direction of each point are recorded as shown in Fig. 3.3. It should be stated that, in order to show the increasing trend of the curvature of the tower as the intensity level increases, the curvatures are normalized by insuring tower bottom curvature to be positive. The displacements of the tower are normalized by insuring tower top displacement to be positive.

180 160 180 160

140 120 100 80 60 40 20 0

wave2-0.001 wave2-0.004 wave2-0.007 wave2-0.01 wave2-0.013 wave2-0.016 wave2-0.02 wave2-0.023 wave2-0.026 wave2-0.029 wave2-0.032( NOT CONV) wave2-0.035

140 120 100 80 60 40 20 0

wave3-0.001 wave3-0.004 wave3-0.007 wave3-0.01 wave3-0.013 wave3-0.016 wave3-0.02 wave3-0.023 wave3-0.026 wave3-0.029 wave3-0.032 wave3-0.035( NOT CONV)

Height of the tower(m)

-0.01 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10 0.11

Height of the tower(m)

-0.005 0.000 0.005 0.010 0.015 0.020 0.025 0.030 0.035 0.040 0.045

**curvature distribution at the maximum curvature time( 1/m)
**

180 160 140 120 100 80 60 40 20 0 -0.12 -0.10 -0.08 -0.06 -0.04 -0.02 0.00 0.02 0.04 180 160

Curvature distribution at the maximum curvarure time( 1/m)

wave7-0.001 wave7-0.004 wave7-0.007 wave7-0.01 wave7-0.013 wave7-0.016 wave7-0.02 wave7-0.023 wave7-0.026 wave7-0.029(NOT CONV) wave7-0.032(NOT CONV) wave7-0.035(NOT CONV)

140 120 100 80 60 40 20 0 -0.01 0.00 0.01 0.02 0.03

wave9-0.001 wave9-0.004 wave9-0.007 wave9-0.01 wave9-0.013 wave9-0.016 wave9-0.02 wave9-0.023 wave9-0.026 wave9-0.029(NOT CONV) wave9-0.032(NOT CONV) wave9-0.035(NOT CONV)

Height of the tower(m)

Height of the tower(m)

0.04

0.05

Curvature distribution at the maximum curvarure time(1/m)

Curvature distribution at the maximum curvarure time(1/m)

**Fig. 3.2 Curvature distribution of the tower at the maximum curvature time
**

250 250

200

150

Height ot the tower( m)

100

50

wave2-0.001-2.92s wave2-0.004-2.93s wave2-0.007-2.97s wave2-0.01-3.02s wave2-0.013-3.06s wave2-0.016-3.09s wave2-0.02-3.13s wave2-0.023-3.15s wave2-0.026-3.21s wave2-0.029-3.25s wave2-0.032-3.17s(NOT CONV) wave2-0.035-26.1s

200

150

100

50

0

0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0

wave3-0.001-2.56s wave3-0.004-2.56s wave3-0.007-2.58s wave3-0.01-2.60s wave3-0.013-2.62s wave3-0.016-2.63s wave3-0.02-2.64s wave3-0.023-2.66s wave3-0.026-8.23s wave3-0.029-8.32s wave3-0.032-13.76s wave3-0.035-1.87s(NOT CONV)

Height of the tower(m)

-0.5

Longitudinal displacement distribution at the maxmimum disp. time(m)

-0.2 -0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3

Longitudinal displacement at the maximum disp. time (m)

250

250

200

Height ot the tower( m)

150

100

50

0

wave7-0.001-5.215s wave7-0.004-17.605s wave7-0.007-3.055s wave7-0.01-5.635s wave7-0.013-23.205s wave7-0.016-23.3s wave7-0.02-38.44s wave7-0.023-40s( Too large) wave7-0.026-40s( Too large) wave7-0.029-37.905s(NOT CONV) wave7-0.032-35.375s(NOT CONV) wave7-0.035-37.96s(NOT CONV)

-0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2

200

150

100

50

0

wave9-0.001-11.6s wave9-0.004-11.345s wave9-0.007-4.705s wave9-0.01-21.265s wave9-0.013-25.17s wave9-0.016-28.13s wave9-0.02-40s( TOO LARGE) wave9-0.023-40s( TOO LARGE) wave9-0.026-40s( TOO LARGE) wave9-0.029-39.655s( NOT CONV) wave9-0.032-35.105s( NOT CONV) wave9-0.035-36.5s( NOT CONV)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0

Height ot the tower( m)

Longitudinal displacement distribution at the maxmimum disp. time(m)

-0.5

Longitudinal displacement distribution at the maxmimum disp. time(m)

Fig. 3.3 Longitudinal displacement distribution of the tower at the maximum disp. time

th

The 14 World Conference on Earthquake Engineering October 12-17, 2008, Beijing, China

From above graphs the conclusion can be obtained that, for different ground motion records or for different scales for a single ground motion record, the damage zone (plastic zone) does not occur in a fixed place. But there still exist some rules as the following. 1． Under longitudinal ground motion excitation, the damages of the tower mainly concentrate on three zones: the two pylon legs’ bottoms (SectionC), the two pylon legs’ tops (SectionG), the middle zones of the two pylon legs’ upper parts (SectionF). 2． The maximum displacement occurs in two critical positions, the intersection point of the two pylons (PointA), and the top of the tower (PointB). 3． The damages caused by Wave1，wave2, mainly concentrates on the pylon leg bottom (SectionC), when the Sa(T1) of the ground motion is below 0.013g. As the intensity increases, the maximum curvatures of the tower occur at section C, section D section E and section G because of the apparent increasing contribution of high order modes. When the Sa is above 0.023g, the plastic curvature mainly concentrates on the SectionC and SectionG, and the value is nearly the ultimate curvature, which may cause the collapse of the tower. 4． The damages caused by Wave3，wave4，wave5，wave6 mainly concentrates on the pylon leg bottoms, even when the Sa(T1) is above 0.01g, which will cause the yield of critical section in some important part. The reason may be that the spectra shapes of these four ground motion are smoother than other ground motions. 5． Under Wave7，Wave8，wave9 and wave10 with IMs below 0.01g, the damages mainly occurs at sectionC and sectionG. When IMs are above 0.02g, (this intensity would cause some section exceeding their ultimate curvatures), the maximum curvature occurred sometimes at the pylon leg middle, because this curvature level is much higher than the one caused by the lower intensity level. This kind of damage can be thought as the dynamic instability.

4.DAMAGE STATE DEFINITION To analysis the seismic performance of the tower, different damage state or (limit state) can be defined with two major measures. One is the flexural curvature at the critical section. The other is the displacement of the tower at the critical point. For the former, the axial forces as the maximum curvature occurred has been adopted to calculate the different damage state of the different section, which are decided by the material strain levels in different states. The material strain level defined by Arzoumanidis S [2005] in Taoma new suspension bridge tower is modified in consideration of some recent research result of bridge concrete piers. The material strain levels for different states are listed in table 4.1.The critical section, includes SectionC, SectionD, SectionE, SectionF, SectionG, as shown in Fig. 2.1. Although the different axial forces are used to calculate the curvature for different intensity level of different ground motion records, only 50% fractile curvature are shown in figures for clarity. Table 4.1 Damage states and strain levels for the critical sections Damage states Damage description Damage measures of strain levels there's no crack on the concrete, the steel ε s ≤ ε y = 0.001675 No damage bars don’t yield the spalling of the cover concrete does 0.001675 < ε s ≤ ε h = 0.015; Slight damage not occur, and the cracks were less than ε c ≤ 2ε co = 0.004 1cm strain. the damage on the section were limited 0.015 < ε s ≤ 0.55ε su = 0.0495; Repairable damage to the repairable state economically and 0.004 < ε cc ≤ 0.75ε ccu = 0.007875 technically the damage on the section cannot be 0.55ε su < ε s ≤ ε su = 0.09; Extensive damage repaired; the tower doesn’t lose bearing 0.75ε ccu < ε cc ≤ ε ccu = 0.0105; capacity.

Complete damage the tower collapsed

ε s > ε su = 0.09; ε cc > ε ccu = 0.0105;

The drift ratio measures of two critical points (as shown in Fig.2.1) are adopted as defined by Ahmed Ghobarah

th

The 14 World Conference on Earthquake Engineering October 12-17, 2008, Beijing, China

[2001], which is show in Table 4.2. Table 4.2 Damage states measured by drift ratio for the critical node Drift ratio at point B Drift ratio at point A Damage state At the bottom of the anchorage zone At the top of the tower No damage UA<=0.2% UB<=0.2% Slight damage UA<=0.5% UB<=0.5% Repairable damage UA<=1.5% UB<=1.5% Extensive damage UA<=2.5% UB<=2.5% Complete damage UA>2.5% UB>2.5%

5. SEIMIC PERFORMANCE EVALUATION USING IDA METHOD 5.1 SEISMIC DAMAGE STATE EVALUATION USING IDA METHOD Based on the above definition of damage states, seismic performance was studied by IDA analysis as shown in Fig.5.1 ~ 5.2.

0.040 0.035 0.030

0.090 0.085 0.080 0.075 0.070 0.065 0.060 0.055 0.050 0.045 0.040 0.035 0.030 0.025 0.020 0.015 0.010 0.005 0.000 0.000 0.002 0.004 0.006 0.008

0.090 0.085 0.080 0.075 0.070 0.065 0.060 0.055 0.050 0.045 0.040 0.035 0.030 0.025 0.020 0.015 0.010 0.005 0.000 0.000

Sa(T1,3% )(g)

0.025 0.020 0.015 0.010 0.005 0.000 0.000 0.002 0.004 0.006 0.008 0.010

wave1 wave2 wave3 wave4 wave5 wave6 wave7 wave8 wave9 wave10

The threshold of Slight damage The threshold of Repairable damage The threshold of Extensive damage The threshold of Complete damage

wave1 wave2 wave3 wave4 wave5 wave6 wave7 wave8 wave9 wave10

Sa(T1,3% )(g)

Sa(T1,3% )(g)

**wave1 wave2 wave3 wave4 wave5 wave6 wave7 wave8 wave9 wave10
**

0.002 0.004 0.006 0.008 0.010

0.010

Curvature of SectionC at the maximum curvature time(1/m)

Curvature of SectionF at the maximum curvature time(1/m)

Curvature of SectionG at the maximum curvature time(1/m)

**Fig. 5.1 IDA curves and the capacity of critical section at the time of the maximum curvature
**

0.055 0.050 0.045 0.040 0.055 0.050 0.045 0.040

0.035 0.030 0.025 0.020 0.015 0.010 0.005 0.000 0.000

**wave1 wave2 wave3 wave4 wave5 wave6 wave7 wave8 wave9 wave10
**

0.005 0.010 0.015 0.020 0.025 0.030

0.035 0.030 0.025 0.020 0.015 0.010 0.005 0.000 0.000

wave1 wave2 wave3 wave4 wave5 wave6 wave7 wave8 wave9 wave10

Sa(T1,ξ=0.03)

Sa(T1,ξ=0.03)

0.005

0.010

0.015

0.020

0.025

0.030

Drift ratio of Point A at the maximum displacement time

Drift ratio of Point B at the maximum displacement time

Fig. 5.2 IDA curves and the capacity of critical Point at the time of the maximum displacement From figure, some remarkable characteristics of the seismic performance of the whole tower can be conclude: 1. As the increasing of the seismic intensity, either for curvature or for displacement of the critical point, the dispersion grows larger. 2. At the time of maximum curvature occurs in the tower, the main plastic zone mainly concentrates on sectionC, section F, sectionG. 3. The displacement curve of PointA and PointB also indicate some plastic performance of the tower, the slope of them is gentler than that of the curvature. To better summarize the seismic performance of the tower, the statistic analysis method has been used herein. The 50%, 84%, 16% fractile of the IDA curves of the maximum curvature of the whole tower and the maximum drift ratio of the whole tower are illustrated in Fig. 5.3.

th

The 14 World Conference on Earthquake Engineering October 12-17, 2008, Beijing, China

**0.035 0.030 0.025 0.05
**

wave1 wave2 wave3 wave4 wave5 wave6 wave7 wave8 wave9 wave10 50% fractile 16% fractile 84% fractile wave1 wave2 wave3 wave4 wave5 wave6 wave7 wave8 wave9 wave10 50% fractile 16% fractile 84% fractile

Sa(T1,3% )(g)

0.020 0.015 0.010 0.005 0.000 0.000 0.005 0.010 0.015

Sa(T1,3% )(g)

0.04 0.03 0.02 0.01 0.00 0.000 0.005 0.010 0.015 0.020 0.025 0.030

Maximum curvature of the whole tower( 1/m)

Maximum drift ratio of the whole tower

Fig. 5.3 Summary of IDA curves of the Maximum drift ratios and Maximum curvatures of the tower From the point view of the section damage of the tower, the figures shows below Sa=0.003g tower will keep in no damage state at a probability of 50%, below Sa=0.0119g the tower will keep in slight damage state at a probability of 50%, above Sa=0.0144g the tower will go into extensive damage, and the tower will collapse when Sa is over 0.0155g at a probability of 50%. The same intensity measures for different damage states of the drift ratios of the tower are 0.005g, 0.0118g, 0.03g and 0.035g with the same order. Compared with the drift ratio as the damage measure, taken the curvature as the damage measure to evaluate the damage state will be more rigorous, especially for the state of repairable damage state and extensive damage state. Because the curvature damage measure has more supporting proof from experiment of bridge piers, it has more reliability than the drift ratio. 5.2 DUCTILIY CAPACITY DICUSSION To investigate the ductility of the whole tower under different ground motion excitations based on summary of the maximum curvature IDA curves, the yield curvature are assigned by the turning point of the 50% fractile lines. At this point, the curvature equals to 0.001(1/m), earthquake intensity equals to 0.01g, normalized by the two threshold value. The R-μ relationships is plotted in Fig. 3.5., from which it can be seen if the tower goes into plastic state at large ground motion level, the maximum available ductility factor can be thought as 8, the strength reduction factor of the tower can be thought as 1.5. Using the same earthquake intensity level, which is 0.01g to normalizing the yield drift ratio, the yield value of the drift ratio will be 0.004. It can be seen the maximum available ductility factor of the drift ratio will be only 1.25. It is clear that the drift ratio ductility (or displacement ductility) is defined by the yield state and ultimate state of the section. This method still needs discussion in details. So for a complex structure system, the global displacement ductility capacity is difficult to define, the relationship between this ductility and the curvature ductility is difficult to define as well. So the section curvature ductility is more appropriate to be the measure of global ductility than the displacement ductility, because it has more reliability than the latter. On the other hand, the curvature ductility may be more useful for seismic design.

yield

3.0

threshold of slight damage

Strength reduction,R=S a/S a

2.5 2.0 1.5 1.0 0.5 0.0 0 2 4 6

yield

Strength reduction,R=S a/S a

wave2 wave3 wave5 wave6 wave8 wave9 wave11 wave12 wave14 wave15 50% fractile 16% fractile 84% fractile

5 4 3 2 1 0 0 1 2 3 4 5

yield

wave1 wave2 wave3 wave4 wave5 wave6 wave7 wave8 wave9 wave10 50% fractile 16% fractile 84% fractile

threshold of repairable damage threshold of extensive damage threshold of complete damage threshold of tower section yield state threshold of tower section complete damage state

yield

8

10

6

7

8

Ductility, μ=φ/φ

Ductility, μ=δ/δ

Fig. 5.4 R-μ relationship of the Maximum drift ratio and Maximum curvature of the tower

th

The 14 World Conference on Earthquake Engineering October 12-17, 2008, Beijing, China

6. CONCLUSIONS From all the IDA analyses of the reinforced concrete tower of a cable-stayed bridge in above, nonlinear seismic demand and capacity are investigated, some conclusions can be drew as follows: 1. Under longitudinal ground motion excitation, the damages of the tower mainly concentrate on three regions: the two pylon legs’ bottoms (Section C), the two pylon legs’ tops (Section G), the middle zones of the two pylon legs’ upper parts (Section F). The maximum displacement will occurred in two critical positions, the intersection point of the two pylons (Point A) and the top of the tower (Point B). 2. As the increasing of the seismic intensity, either for curvature or for displacement of the critical point, the dispersion grows larger. At the time of maximum curvature occurs in the tower, the plastic zone mainly concentrates on sectionC, section F, sectionG. The displacements of PointA and PointB also contain some plastic performance of the tower, the slope of which is gentler than that of the curvature. 3. The IDA method is a precise tool for seismic damage state evaluation which has more probabilistic characteristic. From the point view of the section damage of the tower, when Sa is below 0.003g, the tower will keep in no damage state at the probability of 50%, below 0.0119g the tower will keep in slight damage state at a probability of 50%, above 0.0144g the tower will go into the extensive damage state, the tower will collapse when Sa is over 0.0155 at a probability of 50%. The same intensity measures for different damage states of the drift ratios of the tower are 0.005g, 0.0118g, 0.03g, 0.035g at the same order. 4. From the view of ductility, the section curvature ductility is more appropriate to be the measure of global ductility than the displacement ductility, because it has more reliability than the latter, especially for such complex structure as the cable-stayed bridge tower. For the tower studied herein, the maximum available ductility factor reaches 8, the strength reduction factor of the tower is 1.5. Acknowledgements This research is supported by Science & Technology Program Foundation For Transportation Development In West China under Grant No. 200631882225. REFERENCES Aly S. Nazmy, Ahmed m. Abdel-Ghaffar. (1990) Non-linear earthquake-response analysis of long-span cable-stayed bridge: theory, Earthquake Engineering & Structural Dynamics, 36: p. 45-62. W.-X. Ren and M. Obata. (1997) Elastic-Plastic Seismic Behavior Of Long Span cable-Stayed Bridges, Journal of Bridge Engineering, ASCE, 4: 3,194-203. Kazuo ENDO, Chihiro KAWATOH, Shigeki UNJOH. (2004) Analytical Study On Seismic Performance Evaluation Of Long-Span Suspension Bridge Steel Tower, 13th World conference on earthquake engineering Vancouver, B.C., Canada, No.944. Shehata Eldabie ABDEL RAHEEM. (2003) Tower Nonlinear Dynamic Response Of Cable-Stayed Bridges Under Great Earthquake Ground Motion, http:// www.civil.hokudai.ac.jp/egpsee/alumni/abstracts/Shehata.pdf D. Vamvatsikos. (2002) Seismic Performance, Capacity And Reliability Of Structures As Seen Through Incremental Dynamic Analysis, Stanford University, Stanford. Silvia Mazzoni, Frank McKenna, Michael H. Scott, Gregory L. Fenves,et al.(2007) OpenSees Command Language Manual http: //opensees.berkeley.edu/OpenSees/user/primer.php Mander JB, Priestley MJN, Park R (1988) Observed stress–strain behavior of confined concrete. J Struct Eng ASCE 114:8, 1827–1849. YAN Haiquan WANG Junjie. (2007) A tower model for seismic response prediction of floating cable-stayed bridge in longitudinal direction. Journal Of Earthquake Engineering And Engineering Vibration 27:4 80-86 Arzoumanidis S, Shama AA, Marlow SJ, Orsolini GO. (2005) The New Tacoma Narrows Suspension Bridge: Critical Issues in Seismic Analysis and Design. Structures 2005.New York, ASCE. p. 21-21. Ahmed Ghobarah, Performance-based design in earthquake engineering: state of development. (2001) Engineering Structures 23, 878–884.