Available online at www.sciencedirect.com Available online at www.sciencedirect.

com

Procedia Engineering
www.elsevier.com/locate/procedia

Procedia Engineering 931938 Procedia Engineering 14 (2011)00 (2011) 000000

The Twelfth East Asia-Pacific Conference on Structural Engineering and Construction

Shake Table Test of Cable-Stayed Bridge Subjected to Non-Uniform Excitation

Cheng-Yu YANG1a, Moe M. S. CHEUNG2
1 2

State Key Laboratory of Disaster Reduction in Civil Engineering, Tongji University, China Department of Civil Engineering, Hong Kong University of Science and Technology, China

Abstract The shake table test of a long span cable-stayed bridge model subjected to non-uniform excitation is presented in this paper. The prototype is a double-deck cable-stayed bridge with a main span of 430m and side spans of 160m on either side. An innovative dual shake table testing system is designed and constructed for the purpose of this study. The design of 1:120 model of the prototype bridge based on laws of similitude for dynamic testing is also presented in this paper. In order to study the effect of non-uniform excitation on different structural form, two types of tower-girder connection are considered in the testing. The earthquake records from SMART-1 are used in this study. Testing result of cable-stayed bridge model subjected to identical input and varying input is reported and the corresponding discussion is given.

2011 Published by Elsevier Ltd.

Keywords: shake table test, cable-stayed bridge, non-uniform excitation.

1. Introduction Observations from instrument arrays of earthquake strong-motion have shown visible differences among the records from accelerometers at different locations within the dimensions of typical long span structures (Zerva and Zhang 1997; Santa-Cruz et al. 2000). In spite of earthquake ground motions spatially varying nature, the uniform input mode is widely adopted and can provide credible results for seismic analysis of normal structures which have limited space or span. While for long span structures, as bridges, dams, stadiums, and pipe network systems, which are long with respect to the wavelengths of the

Corresponding author and Presenter: Email: seismicyang@tongji.edu.cn

18777058 2011 Published by Elsevier Ltd. doi:10.1016/j.proeng.2011.07.117

932

Cheng-Yu YANG and Moe M.S. CHEUNG / Procedia Engineering 14 (2011) 931938

Author name / Procedia Engineering 00 (2011) 000000

input earthquake ground motions, the effects of non-uniform excitation may have significant influence on seismic response of structures. During the last 30 years, many scholars have been working on the numerical analysis of seismic behavior of long span bridges subjected to non-uniform ground motion (Nazmy and Abdel-Ghaffar 1992; Mylonakis et al.2001; Sextos et al. 2003; Soyluk et al. 2004); however, few experimental studies have been carried out in this area. Experimental method plays an important role in civil engineering research. Both the full scale field test and the reduced scale model test are regarded as strong research techniques which can be applied to study the real behavior of the structure under different loadings and to verify the theoretical analysis. However, its very difficult to perform experimental study of long span bridges subjected to earthquake loading. The space and payload limitation of the shake table and the severe similitude requirements for dynamic testing significantly increase the difficulty in this research. Several shake table tests of full bridge model of long span bridges has been reported in the past few decades (Godden and Aslam 1978; Garevski et al. 1991; Wang et al. 2006), while none of them include the effect of non-uniform excitation. Shake table tests of two girder bridge model, including the non-uniform cases, were accomplished at the University of Nevada, Reno, by using a multiple shake table system (Carden et al. 2006; Saiidi et al. 2007). The shake table test of a long span cable-stayed bridge model subjected to non-uniform excitation is performed in this study. The prototype is a double-deck cable-stayed bridge with a main span of 430 m and side spans of 160 m on either side. In order to study the effect of non-uniform excitation on different structural form, two types of tower-girder connection are considered in the testing. The earthquake records from the Strong Motion Array in Taiwan, Phase I (SMART-1) are used in this study. Due to space limitation, only part of the testing result is reported in this paper and the corresponding discussion are given. 2. Multiple Shake Table Testing System
Table 1: Multiple shake table testing systems in use and under construction Size (m) 3.6 3.6 4.3 4.5 Max. Payload (ton) Frequency Range (Hz) 0.1-100 Max. Travel (mm) Max. Acceleration (g)

Location University at Buffalo, The State University of New York, USA University of Nevada, Reno, USA Chongqing Communications Research and Design Institute, China Pusan National University, Korea* Fuzhou University, China* Tongji University, China*

No. of Tables

Shaking Direction

50

X,Y,Z

+150

1.15

45

0.1-50

X,Y

+300

1.0

3 6 4 4 5 5 2.5 2.5 4 4

35

0.1-50

X,Y,Z

+150

1.0

30 60 22 10 30 70

0.1-60

X,Y,Z

+300

1.2

3 4

0.1-50 0.1-50

X,Y X,Y

+250 +500

1.5 1.5

4 6

Cheng-Yu YANG and Moe M.S. CHEUNG / Procedia Engineering 14 (2011) 931938

933

under construction In doing dynamic test of long span structures subjected to spatially varying earthquake ground motion, one prerequisite is a multiple shake table system. Supported by advancing of test technology and keep-rising interest in this research topic, several large multiple shake table testing systems have recently been developed or under construction. Details of these multiple shake table systems can be found in Table 1. An innovative dual shake table testing system is designed and constructed for the purpose of this study in the Hong Kong University of Science and Technology. The system is composed of 2 shake tables. Each table is 1.5m by 1.5m, with a maximum payload of 2 ton and providing a maximum acceleration of 1g and maximum stroke of +75mm in X-axis. Each table is driven by an MTS 244.22 hydraulic actuator, which is supported by a steel-made reaction stand with high rigidity. Since the table and reaction stand are anchored to the strong floor at grid holes, it can be easily relocated and rotated. Therefore, not only the center-to-center distance can be changed, but also the dual shake table system can provide both in-line excitation and parallel excitation to satisfy different testing purposes, as shown in Figure 1.

(a)

In-line excitation mode Figure 1: Dual shake table testing system

(b) Parallel excitation mode

3. Model Design and Input Motions 3.1. Prototype and two types of tower-girder connection The prototype adopted in this study is the Kap Shui Mun Bridge in Hong Kong, which is a double-deck cable-stayed bridge with a main span of 430 m and side spans of 160 m on either side (see Figure 2). Most of the main span is made of steel/concrete composite girder while the rest of main span and both side spans are PC box girders. The two H-shape concrete towers are 150m high, consisting of 2 shafts and 3 cross beams. The bridge consists of 176 stay cables, arranged in two vertical planes. The bridge girder is vertically supported by 4 piers and 2 towers (LT Tower and MW Tower). In order to study the effect of non-uniform excitation on different structural form, two types of tower-girder connection are considered in the testing. The first type, named as Fixed Model, adopts the same tower-girder connection as real bridge, i.e., the bridge girder is movable on the cross beam of LT tower while fixed on the cross beam of MW tower along the longitudinal direction. The second type, named as Floating Model, adopts longitudinal moveable tower-girder connection for both two main towers. The longitudinal floating system is a widely adopted structural form for cable-stayed bridges constructed in Mainland China. Therefore, studying the effects of spatially varying earthquake ground motion on seismic response of these two types of structural form is of great significance.

934

Cheng-Yu YANG and Moe M.S. CHEUNG / Procedia Engineering 14 (2011) 931938

Author name / Procedia Engineering 00 (2011) 000000

Figure 2: Front evaluation of the Kap Shui Mun Bridge

3.2. Reduced scale model design A 1:120 model of the prototype bridge is designed based on laws of similitude for dynamic testing. The geometric scale of 1:120 is determined by the length of the prototype and the dimensions of the dual shake table. Acrylic is selected to fabricate the model tower. Aluminum plate with rectangular cross-section is selected to model the girder with satisfaction of ratio of similitude on vertical and transversal bending stiffness. High-strength stainless steel wire is used for modeling cables and machined iron blocks are attached to the model as artificial mass. A summary of ratio of similitude adopted in the proposed model can be found in Table 2. For detailed description of model design, one may refer to Yang 2007.
Table 2: Ratio of similitude for reduced scale model Quantity Scale factor Length 1:120 Acceleration 1:1 Time 1:10.954 Frequency 1:0.0913 Modulus 1:11.182 Strain 1:1 Model weight 1:161022

3.3. Input ground motion Earthquake records from SMART-1 seismograph array are used in this study. The SMART-1 array consists of 36 stations which are configured in three concentric circles, with radii of 200 m (inner ring), 1000 m (middle ring) and 2000 m (outer ring). In view of the fact that each pair of stations (along the diametrical direction) in the inner ring are spaced at a distance of 400 m, which is quite close to the span length of the example bridge (430m); several pairs of selected records from the inner ring stations are used in this study. Table 3 lists a typical group of cases for the shake table testing. For each group of cases, one pair of ground motion records from the inner ring station of SMART-1 is used. In the identical input cases, both tables have the same input wave; while in the varying input case, different input waves are applied to different tables. The wave propagation effect is studied by changing the wave passage velocity, including five cases, i.e. 400 m/s, 600 m/s, 1000 m/s, 2000 m/s and infinite. The distance between the two main towers (430 m) is used to calculate the difference of the arrival time for the two tables. Only the longitudinal input is considered in this study.

Cheng-Yu YANG and Moe M.S. CHEUNG / Procedia Engineering 14 (2011) 931938

935

Table 3: One typical group of test cases Case No. 1 2 3 4 5 6 7 8 9 10 Varying Input Identical Input Test Type Input at Pier1,2 and LT Tower Wave No.1 Wave No.1 Wave No.1 Wave No.1 Wave No.1 Wave No.1 Wave No.1 Wave No.1 Wave No.1 Wave No.1 Input at Pier 3,4 and MW Tower Wave No.1 Wave No.1 Wave No.1 Wave No.1 Wave No.1 Wave No.2 Wave No.2 Wave No.2 Wave No.2 Wave No.2 Arrival Time delay(sec) 0 0.215 0.430 0.717 1.075 0 0.215 0.430 0.717 1.075 Wave velocity(m/sec) Infinite 2000 1000 600 400 Infinite 2000 1000 600 400

4. Test Results and Discussion Due to space limitation, one group of input cases (including Identical Input and Varying Input) for both structural forms (i.e. Fixed Model and Floating Model) are presented in this paper.

Peakverticalacc.(g)atmidspan
0.4 0.3 0.2 0.1 0 Infinite 2000m/s 1000m/s 600m/s 400m/s 0.4 0.3 0.2 0.1 0

Peakverticalacc.(g)atmidspan

Infinite 2000m/s 1000m/s 600m/s 400m/s

Peakverticalacc.(g)atquarterspan
0.4 0.3 0.2 0.1 0 Infinite 2000m/s 1000m/s 600m/s 400m/s
(a) Identical Input

Peakverticalacc.(g)atquarterspan
0.4 0.3 0.2 0.1 0 Infinite 2000m/s 1000m/s 600m/s 400m/s
(b) Varying Input

Figure 3: Peak vertical acceleration along the girder, Fixed Model

936

Cheng-Yu YANG and Moe M.S. CHEUNG / Procedia Engineering 14 (2011) 931938

Author name / Procedia Engineering 00 (2011) 000000

Peakstrainatmiddletower
0.0002 0.00015 0.0001 0.00005 0 Infinite 2000m/s 1000m/s 600m/s 400m/s 0.0002 0.00015 0.0001 0.00005 0

Peakstrainatmiddletower

Infinite

2000m/s 1000m/s 600m/s

400m/s

Peakstrainatlowertower
0.0004 0.0003 0.0002 0.0001 0 Infinite 2000m/s 1000m/s 600m/s 400m/s 0.0004 0.0003 0.0002 0.0001 0

Peakstrainatlowertower

Infinite

2000m/s 1000m/s

600m/s

400m/s

(a)

Identical Input

(b) Varying Input

Figure 4: Peak longitudinal strain along the MW Tower, Fixed Model

Peakverticalacc.(g)atmidspan
0.4 0.3 0.2 0.1 0 Infinite 2000m/s 1000m/s 600m/s 400m/s 0.4 0.3 0.2 0.1 0

Peakverticalacc.(g)atmidspan

Infinite 2000m/s 1000m/s 600m/s 400m/s

Peakverticalacc.(g)atquarterspan
0.4 0.3 0.2 0.1 0 Infinite 2000m/s 1000m/s 600m/s 400m/s
(a) Identical Input

Peakverticalacc.(g)atquarterspan
0.4 0.3 0.2 0.1 0 Infinite 2000m/s 1000m/s 600m/s 400m/s
(b) Varying Input

Figure 5: Peak vertical acceleration along the girder, Floating Model

Cheng-Yu YANG and Moe M.S. CHEUNG / Procedia Engineering 14 (2011) 931938

937

Peakstrainatmiddletower
0.00015 0.0001 0.00005 0 Infinite 2000m/s 1000m/s 600m/s 400m/s 0.00015 0.0001 0.00005 0

Peakstrainatmiddletower

Infinite

2000m/s 1000m/s 600m/s

400m/s

Peakstrainatlowertower
0.0002 0.00015 0.0001 0.00005 0 Infinite 2000m/s 1000m/s 600m/s 400m/s 0.0002 0.00015 0.0001 0.00005 0

Peakstrainatlowertower

Infinite

2000m/s 1000m/s 600m/s

400m/s

(a) Identical Input

(b) Varying Input

Figure 6: Peak longitudinal strain along the MW Tower, Floating Model

Typical results of Fixed Model subjected to Identical Input and Varying Input are shown in Figure 3 and 4. From preliminary comparison, one finds that non-uniform excitation strongly influences the dynamic response of the bridge model. Comparing the results at the same wave velocity, in most cases the varying input excitation results in a visibly larger response than the identical input. In terms of the wave propagation effect, the influence is inconsistent with different structural components. As shown in Figure 3 and 4, the wave propagation influences the vertical response of the girder while has relatively less effect on the tower response. Typical results of Floating Model subjected to Identical Input and Varying Input are shown in Figure 5 and 6. Same as Fixed Model, the wave propagation effect has relatively larger impact on girder than tower for Floating Model. However, the finding from Fixed Model doesnt work completely on Floating Model. Comparing the results of identical input and varying input in Figure 5 and 6, one finds that the varying input has a beneficial effect on the structure in general, which is adverse as what we find in Fixed Model. In view of the complexity and lack of regularity in these results, more test investigation and numerical analysis for comparison will be introduced in this study. 5. Concluding remarks Shake table test of a long span cable-stayed bridge model subjected to non-uniform excitation is presented in this paper. The following concluding remarks can be drawn: A dual shake table testing system and a 1:120 scaled cable-stayed bridge model is designed and constructed for this study. The test is well performed and provides some credible results, which help to improve the understanding of dynamic response of long span bridges subjected to spatially varying earthquake ground motion. Comparing the results of Fixed Model and Floating Model, one finds the effect of non-uniform excitation has a complex and inconsistent influence on different structural form and different structural components. To further study the topic, more test data should be analyzed and numerical simulation for comparison should be included.

938
8

Cheng-Yu YANG and Moe M.S. CHEUNG / Procedia Engineering 14 (2011) 931938
Author name / Procedia Engineering 00 (2011) 000000

Acknowledgements The authors wish to gratefully acknowledge the financial support of the Hong Kong Research Grant Council RGC No. 612106 for the development of finite strip analysis of spatial vibration of multi-support, long span bridges. The authors are also grateful to the Institute of Earth Sciences (Taiwan) for providing the seismic records of SMART-1. References
[1] [2] [3] [4] [5] [6] Carden LP, Itani AM and Buckle IG (2006). Seismic performance of steel girder bridges with ductile cross frames using single angle X braces. ASCE Journal of Structural Engineering, 132(3), pp. 329-337. Garevski MA, Brownjohn JMW, Blackborough A and Severn RT (1991). Resonance-search tests on a small-scale model of a cable-stayed bridge. Engineering Structures, 13, pp. 59-66. Godden WG and Aslam M (1978). Dynamic model studies of Ruck-A-Chucky Bridge. Journal of the Structural Division, ASCE, 104(ST12), pp. 1827-1844. Mylonakis G, Papastamatiou D, Psycharis J and Mahmoud K (2001). Simplified modeling of bridge response on soft soil to nonuniform seismic excitation. ASCE Journal of Bridge Engineering, 6(6), pp. 587-597. Nazmy AS and Abdel-Ghaffar AM (1992). Effects of ground motion spatial variability on the response of cable-stayed bridges. Earthquake Engineering and Structural Dynamics, 21, pp. 1-20. Saiidi MS, Nelson R, Sadrossadat-Zadeh M and Buckle, IG (2007). Shake table studies of a 4-span reinforced concrete bridge model. Proceedings 2007 ANCER Meeting, Earthquake Engineering Research: From Strong Seismic Regions to Regions of Moderate Seismicity, HK, China. [7] Santa-Cruz S, Heredia-Zavoni E and Harichandran RS (2000). Low-frequency behavior of coherency for strong ground motions in Mexico City and Japan. Proceedings 12th World Conference on Earthquake Engineering, Auckland, New Zealand, Paper No. 0076. [8] Sextos AG, Pitilakis KD and Kappos AJ (2003). Inelastic dynamic analysis of RC bridges accounting for spatial variability of ground motion, site effects and soil-structure interaction phenomena. Part 1: methodology and analytical tools. Earthquake Engineering and Structural Dynamics, 32, pp. 607-627. [9] Soyluk K, Dumanoglu AA and Tuna ME (2004). Random vibration and deterministic analyses of cable-stayed bridges to asynchronous ground motion. Structural Engineering and Mechanics, 18(2), pp. 231-246. [10] Wang JJ, Zhang XT, Fan LC, Wang ZQ, Chen H, Zhou M, Li SY, Mo HL and Ni ZJ (2006). A brief introduction to the shaking-table test of Liede Bridge. Proceedings 4th PRC-US Workshop on Seismic Analysis and Design of Special Bridges: Advancing Bridge Technologies in Research, Design, Construction and Preservation, Chongqing, China. [11] Yang CY (2007). Seismic analysis of long span bridges including the effects of spatial variation of seismic waves on bridges. Ph. D. thesis. Department of Civil Engineering, Hong Kong University of Science and Technology. [12] Zerva A and Zhang O (1997). Correlation patterns in characteristics of spatially variable seismic ground motions. Earthquake Engineering and Structural Dynamics, 26, pp. 19-30.