Influence of Vibration Modes On Flutter Analysis For Long-Span Suspension Bridges
Influence of Vibration Modes On Flutter Analysis For Long-Span Suspension Bridges
Influence of Vibration Modes On Flutter Analysis For Long-Span Suspension Bridges
ABSTRACT
U
Since the aerodynamic stability is one of the most important issues in the wind resistance design for the long-span bridges, various studies have been carried out. The studies on the Akashi-Kaikyo Bridge, the worlds longest suspension bridge with the center span of 1,991m, revealed that not only the three-dimensional effect of the structure and the wind characteristics but also the influence of multi-vibration modes have to be considered in checking the aerodynamic stability1). If the characteristics of dominant vibration mode changes, the critical flutter velocity also may change. This paper describes the results of flutter analysis for an assumed long-span bridge, which has the center span is approximately 1500m or 2300m, and the influence of vibration mode for each bridge.
P P
1. INTRODUCTION
U
There are several plans or ideas of strait crossing road projects in Japan (Fig.1)2). In these projects, super long-span bridges, which would be longer than the Akashi-Kaikyo Bridge, are included. In order to make these super long-span bridges to come true, the aerodynamic stability is one of the most important issues. In the advanced studies on the Akashi-Kaikyo Bridge, special attentions were paid to the following considerations in order to ensure the aerodynamic stability for a long-span bridge. 1) The three-dimensional effect of the structure Honshu-Shikoku Bridges and the wind 2) The influence of multi-vibration modes According to the later consideration, it can be Kanmon Strait Kitan Strait possible to improve the aerodynamic stability of long span bridges by controlling the dominant vibration modes. Therefore, the influence of the vibration mode on flutter characteristics of assumed suspension bridges was examined. This Tokyo Bay Mouth study is base on the Wind resistant design code Ise Bay Mouth for Honshu-Shikoku Bridges (2001)3). Hoyo Strait
P P P P
Honshu-Shikoku Bridges
Fig.1
1 Sub-Leader, Wind and Structural Engineering Group, Long-span Bridge Engineering Center, Honshu-Shikoku Bridge Expressway Co., Ltd. (former Honshu-Shikoku Bridge Authority), Japan 2 Leader, ditto 3 Engineer, ditto
Start The procedure of the flutter analysis in this paper is shown in Fig.2 Fig.3 shows two suspension bridges assumed in the flutter analysis. These bridges are designed for the above-mentioned strait crossing road projects. For simplicity, a two-span suspension bridge with the center span of 1,480m is called Bridge-A, and a three-span suspension bridge with the center span of 2,250m is called Bridge-B. The cross sections of girders are shown in Fig.4. One-box girder is applied to Bridge-A and slotted-box girder with superior aerodynamic stability is applied to Bridge-B. Assumption of Suspension Bridges Choose the Cross section of girder (Spring-Supported test) Coefficients of Three Components Forces and Unsteady Aerodynamic Forces Flutter Analysis Evaluation of Aerodynamic Stability End Fig.2
360 1480
1A
860
2P (a) BridgeA
2250
3P
4A
860
1A
3P
4A
(Unit: m)
4.0
3.0
26.1
26.7
The results of the two-dimensional spring supported test at the angles of attack of -3, 0 and +3 degrees are shown in Table 1. Both of the two cross sections showed good aerodynamic stabilities except at +3 degree (one-box girder) and -3 degree (slotted-box girder). Fig.5 shows coefficients of three components forces. The coefficients of unsteady aerodynamic forces, which coordinate system is defined by Fig.6, is shown in Fig.7. Coefficients of the unsteady aerodynamic forces were defined as follows: L = B 2 LZR 2 z + LZI z + B 3 LR 2 + LI (1) (2) where, L:lift, M:aerodynamic moment, z:vertical displacement, :torsional displacement, :circular frequency, ( )':d( )/dt, Lxx or Mxx: coefficients of unsteady aerodynamic forces (Z: caused by vertical vibration, :caused by torsional vibration, R:in phase with displacement, I:in phase with velocity)
ZR
B B B B B B B B
{ ( M = {B (M
3
z + M ZI z ) + B (M R + M I )}
2 4 2
)}
Table 1 Results of the two-dimensional spring-supported test Angle of Attack (deg.) -3 0 +3 Bridge-A (One-box) 82 m/s+ 80 m/s 76 m/s Bridge-B (Slotted box) 81 m/s 100 m/s+ 100 m/s+
Note: + indicates that the critical flutter velocity is larger than the tabulated value
CD , CL
3.0 2.0 1.0 0.0 -1.0 -2.0 -3.0 -15 -10 -5 0 5 CD CL CM 10
CM
0.3 0.2 0.1 0.0 -0.1 -0.2 -0.3 15
CD, CL
3.0 2.0 1.0 0.0 -1.0 -2.0 -3.0 -15 -10 -5 0 5 CD CL CM 10
CM
0.3 0.2 0.1 0.0 -0.1 -0.2 -0.3 15
Angle of attack(deg.)
Angle of attack(deg.)
(a)
Bridge-A Fig.5
D L
Fig.6
2.0 1.5 1.0 One Box Girder Slotted Box Girder Theodorsen
LZR
0.5 0.0 -0.5 0.00 0.05 0.10 0.15 0.20 0.25 fB/V
-6.0 -8.0 -10.0 0.00 0.05 0.10 0.15 0.20 0.25 fB/V One Box Girder Slotted Box Girder Theodorsen
(a) LZR
B B
(b) LZI
B B
M ZR
-0.2 -0.3 -0.4 0.00 0.05 0.10 0.15 0.20 0.25 fB/V One Box Girder Slotted Box Girder Theodorsen
-4.0 -6.0 -8.0 0.00 0.05 0.10 0.15 0.20 0.25 fB/V One Box Girder Slotted Box Girder Theodorsen
(c) MZR
B B
(d) MZI
B B
4.0
20.0 10.0 0.0 0.00 0.05 0.10 0.15 0.20 0.25 fB/V
0.00
0.05
0.10
0.15
0.20
0.25 fB/V
(e) LR
B B
(f) LI
B B
0.5
One Box Girder Slotted Box Girder Theodorsen
0.0 -0.5
I
1.0 0.0 -1.0 0.00 0.05 0.10 0.15 0.20 0.25 fB/V
-1.0 -1.5 -2.0 0.00 0.05 0.10 0.15 0.20 0.25 fB/V One Box Girder Slotted Box Girder Theodorsen
(g) MR
B B
(h) MI
B B
Fig.7
1) Influence of high mode Before selecting the important modes in flutter analysis, the influence of high natural modes was examined in order to evaluate the number of modes in the analysis. The critical flutter velocities were calculated for Bridge-A with various combinations of natural vibration modes. These calculations applied a multi-mode flutter analysis, using the mode combination method. Static displacements by wind load, which were calculated from the measured three component forces (Fig.5), were also considered. Table 2 shows the assumption in the flutter analysis. The preliminary analyses by using the lowest 20, 30, 40, and 50 modes were carried out. The analysis concluded that the right solution was equivalent to the approximation by considering at least the lowest 30 modes. In addition, a mode, which had an influence on the flutter characteristics, was expected to exist between the 20th mode and the 30th mode. Therefore, additional analyses were carried out to identify such mode. The conclusion was that the 21st mode affected largely, and the analysis including modes up to the 22nd mode or more could obtain a good approximation of flutter characteristics of Bridge-A as shown in Fig.8. Table 2 Assumptions in flutter analysis
U
Item Analytical method Air density Structural damping Static deformation in wind condition
Analysis condition Mode combination method.( Using lower 50 modes.) 1.23 kg/m3 =0.02 for all modes Considered Main girder Direction Vertical Torsional Horizontal Lift Moment Drag ;Unsteady aerodynamic forces ;Quasi-steady aerodynamic forces Cable: Quasi-steady drag force and lift force (CD=0.7) Tower: Not considered
P P
Forces
150
Critical flutter velocity (m/s)
100
85 59 57 57 57
50
Fig.8
However, the critical flutter velocities derived from these flutter analyses (Fig.8) were significantly smaller than the results based on the two-dimensional spring-supported test (Table 1). Consequently, it might be difficult for the long-span suspension bridge with center span of 1480m to evaluate the aerodynamic stability by the results of the ordinary two-dimensional wind test. 2) Selection of dominant modes Since it had been predicted that dominant modes existed between the 1st to the 21st of natural vibration modes, a series of analyses were carried out with various combinations of modes. The analysis resulted that the 2nd (1st symmetric vertical mode), the 8th (2nd symmetric vertical mode), and the 21st (1st symmetric torsional mode) dominate the flutter characteristics of Bridge-A. The combination of these three modes is identified with the combination from the 1st to the 21st mode in the flutter characteristics (shown in Fig.9). Results of natural vibration analysis without wind load are shown in Table 3. According to the table, both the 20th and the 21st modes are symmetric torsional mode having close frequencies to each other. Based on previous experiences, the 20th mode had been applied to spring-supported test condition because of its lower frequency and smaller equivalent mass. Though, the result indicated that the 21st mode affected more than the 20th mode. Dominant vibration modes are shown in Fig.10. The 21st mode, which is basically a torsional mode, was combined with not only horizontal mode but also 2nd symmetric vertical mode. Consequently, the 21st mode is considered to have a relatively large contribution to the flutter characteristics. In addition, the proportion of center span length and side span length is exactly 2 to 1 in Bridge-A. This was expected to excite the vertical vibration in high modes.
U
0.15
Logarithmic decrement
0.1
0.05
-0.05 0 20 40 60 80 100
Wind velocity(m/s)
Fig.9
Mode 2
(f=0.078Hz)
Mode 8
(f=0.175Hz)
Mode 20
(f=0.294Hz)
Fig.10 Table 3
Mode 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Frequency (Hz) 0.046 0.078 0.083 0.092 0.115 0.116 0.132 0.175 0.182 0.186 0.193 0.220 0.228 0.243 0.245 0.252 0.263 0.265 0.272 0.294 0.295 0.295 0.305 0.324 0.334 Period (sec) 21.589 12.806 12.025 10.893 8.684 8.592 7.569 5.711 5.485 5.370 5.177 4.552 4.395 4.113 4.081 3.974 3.800 3.769 3.674 3.401 3.391 3.389 3.277 3.082 2.993
353.0
264.8 4020.7
SH-2 (2) 18485.5 38079.2 ST-1 (1) ST-1 (2) AV-3 (1) AV-3 (2)
1) Influence of higher mode The influence of high natural modes was examined in order to evaluate how many number of modes in the analysis for Bridge-B in the same way as Bridge-A. The analysis including modes up to the 20th mode could obtain a good approximation of flutter characteristics of Bridge-B as shown in Fig.11. Therefore, the influence of the lowest 20 modes of Bridge-B was investigated.
U
150 130
114
50
Fig.11
U
2) Selection of dominant modes Flutter analyses with various combinations of vibration modes of Bridge-B were carried out by the same way as Bridge-A. Fig.12 shows the result of analysis. The combination of the 4th mode (1st symmetric vertical mode) and the 13th mode (1st symmetric torsional mode), which usually applied to spring supported test, did not excite the flutter. Furthermore, the result of flutter analysis with the lowest 20 modes can be represented by the analysis with four dominant modes, 4th, 11th (2nd symmetric vertical mode-a), 13th and 14th (2nd symmetric vertical mode-b). The critical flutter velocity, calculated by removing the 11th mode from these four modes, was lower than that calculated with these four modes. As mentioned above, it was found that the 11th mode is the important mode which suppresses the excitation of flutter characteristics.
0.30
Logarithmic decrement
0.20
0.10
0.00
Wind velocity(m/s)
Fig.12
Mode 4
(f=0.073Hz)
Mode 11
(f=0.117Hz)
Mode 13
(f=0.145Hz)
(f=0.147Hz)
2226.1
SH-2 (2) AT-1 (1) AT-1 (2) SH-2 (3) AH-2 (2) SV-3
6. EVALUATION ON ENERGY
U
Besides three-dimentional analysis, the investigation was carried out in order to identify the part of span exciting the flutter. The energy of aerodynamic forces which work on the girder was examined. The energy excited by aerodynamic forces are defined in the following equation.
WL = LR dy R WM
B B
= M
B B B
R d R
(3)
vibration, LR and MR are the real part of the lift and pitching moment of unsteady aerodynamic forces, yR and and R are the real part of the vertical and torsional displacements, respectively. The aerodynamic stability of the bridge depends on plus or minus of Eq.(4), the integral of energy on each nodal points.
span
Wdl =
(W
span 0
+ WM )dl
(4)
Fig.14 shows the spanwise distributions of energy on girders of Bridge-A and Bridge-B at the critical flutter velocity. It is assumed that the spanwise distribution of energy is strongly associated with the flutter mode shape (shown in Fig.15). When we focus the energy distribution in the center span, Bridge-A indicates a high symmetric mode (the 3rd symmetric mode). On the other hand, Bridge-B is the 1st symmetric mode shape. Therefore, each bridge is evaluated to have a different characteristic. Furthermore, the distribution of energy in the side span, as shown in Fig.14, seems to be minus, and it means that the side span has a damping effect regularly. Bridge-A has only single side span, and this fact seems to reduce the aerodynamic stability of the Bridge-A.
()
Lift Moment Drag Total Energy
Energy
2P
-750
-250
250
750
3P
1250
()
()
Energy
2P
-1985
3P
1985
()
Akashi-Kaikyo
T/8
Bridge-A
Bridge-B
7. CONCLUSIONS
U
The flutter analyses were carried out for the assumed long-span suspension bridges with the center spans of 1,500m and 2,300m. The results of the flutter analyses are summarized as follows, 1) The combination of the lowest flexural mode and torsional mode, which usually used in the two-dimensional spring supported test, might have the possibility of underestimating the flutter instability for super long-span bridges. 2) The characteristics of multi-mode flutter with dozens of vibration modes can be represented by the analysis with only three or four dominant modes in flutter analysis. 3) Some of the dominant modes can control the excitation of flutter, which implies that the adjustment of vibration modes can enhance the critical flutter velocity. It is necessary to obtain the influence of high natural modes in more detail by using the same analytical method for other suspension bridges.
REFERENCES 1) Honshu-Shikoku Bridge Authority: The Akashi-Kaikyo Bridge: Design and Construction of the Worlds Longest Bridge, pp.97-100, 1999.8 2) H. Sato, N. Hirahara, K. Fumoto, S. Hirano and S. Kusuhara: Full Aeroelastic Model Test of a Super Long-span Bridge with Slotted Box Girder, Journal of Wind Engineering and Industrial Aerodynamics, No.90, pp.2023-2032, 2002 3) Honshu-Shikoku Bridge Authority: Wind resistant design code for Honshu-Shikoku Bridges (2001), 2001.8 4) N. Toyama, K. Hata and S. Kusuhara: The Influence of the Vibration Mode on Flutter Characteristics of Long-span Bridges, The 6th Asia-Pacific Conference on Wind Engineering, 2005,9
U U