WTT_Binh bridge

7 16 0
  • Loading ...
1/7 trang

Thông tin tài liệu

Ngày đăng: 25/11/2018, 18:17

Vo l 39 N o A u g u s t 2006 Wind Tunnel Test for “Binh Bridge” in Vietnam UEJIMA Hidesaku : P.E.Jp., Manager, Heat & Fluid Dynamics Department, Research Laboratory, Research & Development MATSUDA Kazutoshi : Doctor of Engineering, P.Eng., Manager, Heat & Fluid dynamics Department, Research Laboratory, Research & Development YABUNO Masashi : P.E.Jp., Manager, Design Department, Bridge & Road Construction Division, Logistics Systems & Structures The “Binh Bridge,” located in Haiphong city in Vietnam, consists of a cable-stayed bridge and its approach bridges The results of wind tunnel tests confirmed that the bridge has adequate stability for torsional flutter On the other hand, vortex shedding excitation may occur below the design wind velocity in smooth flow conditions However, according to statistical analyses for wind conditions at the bridge site, the wind velocity at which vortex shedding excitation would occur is extremely rare, and mitigation effects by turbulent flows can be expected as well Therefore, it was decided that no countermeasure for vibration control should be installed for now Introduction The “Binh Bridge” is located in Haiphong city and is the first long-span bridge in Vietnam This bridge consists of a cable-stayed main bridge and approach bridges, and stretches for a total length of 280 meters The main bridge crosses over the Cam River and is a composite cable-stayed bridge with a span of 260 meters Its main structural feature is the adoption of a composite bridge deck consisting of double steel I-girders and an RC slab, and concrete towers There are no cases where this double I-girder cable-stayed bridge has been constructed in Japan; however, there are many cases where it has been adopted overseas because of its economic advantage On the other hand, many studies are reported on the aerodynamic stability of this type of bridge.(1)-(4) According to these reports, some bridges employed relatively large-scale aerodynamic countermeasures after vortex shedding excitation became conspicuous and torsional flutter was apt to occur in wind tunnel tests Thus, the double I-girder type hardly has a good aerodynamic stability In consideration of the above, this bridge was subjected to a detailed aerodynamic stability check by a wind tunnel test This test was conducted based on the wind resistant design criteria of Japan.(5)-(7) This article reports the test details Examination method 2.1 Evaluation and modeling of vibration characteristics Figure shows a general view of the cable-stayed portion of the “Binh Bridge,” and Fig shows a cross section of the bridge deck Because not only the cable-stayed portion of the spans shown, but also a total of 17 spans, including the approach bridges of the spans on each side are structurally continuous, all these spans must be considered when evaluating the vibration characteristics First, a 3D space frame model of all the spans was created to carry out eigenvalue analyses In this type of structure, not Saint Venant’s torsion stiffness, but bending torsion stiffness dominates, because the bridge deck has a π-shaped open section In this examination, a technique similar to Hoshino’s(8) was used as a simple modeling technique that could consider bending torsion stiffness Figure shows the modeled bridge deck Figure shows the mode shape of the first vertical bending mode of free vibration as an example of eigenvalue analysis results of the real bridge In this mode shape, the vibration of the spans of the cable-stayed bridge from S7 to S12 is dominant, and the approach bridge segments from S1 to S7 and from S12 to S18 hardly vibrate This is a common tendency among all the modes considered in the wind tunnel test, that is, the 1st to 5th vertical bending modes and torsional 1st mode and 1st horizontal bending mode The aerodynamic forces acting on the approach bridges can be negligible, as the vibration modes are very small there in comparison to those of the full bridge, which are directly related to the generalized aerodynamic forces having effects on the vibration Thus, for the three-dimensional aeroelastic model, only the spans of the cable-stayed bridge were modeled 2.2 Design criteria wind speed Vietnam in general is under a similar wind environment to that of Japan, and strong winds there are generally limited to monsoons and typhoons The design values for wind 65 Vo l 39 N o A u g u s t 2006 100 000 60 000 S7 100 000 260 000 S8 S9 S10 Fig 60 000 S11 S12 General view of “ Binh Bridge” (unit : mm) Girder Girder Girder 23 400 750 20 500 2% 2% Rigid bar element (Note) Girder : Gives horizontal bending stiffness and Saint Venant’s torsional stiffness Girder : Gives vertical bending stiffness to ensure the matching of bending torsional stiffness Vibration mode Fig 1.0 0.8 0.6 0.4 0.2 0.0 −0.2 −0.4 −0.6 −0.8 −1.0 −700 Cross section of bridge deck (unit : mm) S1 S2 −600 S3 −500 S4 S5 −400 S6 S7 −300 S8 −200 Fig S9 S10 −100 100 Modeling of bridge deck as a beam element(8) S11 200 S12 S13 300 S14 400 S15 S16 500 S17 S18 600 700 Distance from center (m) (Note) Vibration mode : First vertical bending mode Natural frequency : 0.453 Hz Fig Vibration mode of the first vertical bending mode speed are set as follows:(7) Basic wind speed After completion V10 = 30 m/s During erection V10 = 20 m/s Considering the bridge deck elevation of z = 30 m and surface roughness category II, the design criteria wind speed of the bridge deck is: After completion Vd = 35.4 m/s During erection Vd = 23.6 m/s Considering the correction value, safety factor, etc., based on wind speed fluctuations for the design criteria wind speed, the check wind speed for flutter and other divergent vibrations was set to 48.9 m/s 66 Wind tunnel test A check of aerodynamic stability was performed on this bridge by two kinds of response tests, that is, a sectional model test using a bridge deck section model and a full bridge model test using a three-dimensional aeroelastic model The check covered the aerodynamic stability of the bridge not only after completion, but also at an erection stage, just before closing of the center span 3.1 Bridge deck sectional model test The sectional model test was conducted using the structure stability wind tunnel (test section’s sectional area size: 1.5 m wide by 2.5 m high) at IHI Research Laboratory (Yokohama) Figure shows the sectional Vo l 39 N o A u g u s t 2006 model (model scale = 1/60) of the bridge deck This was spring-supported in two degrees of freedom of the vertical bending and torsional mode and was subjected to a response test The structural damping (logarithmic decrement) d was set to approximately 0.02 End plate Railing Concrete block Main girder (Note) Model scale : 1/60 Fig Sectional model of bridge deck Fig Figure shows the response for the completed bridge in smooth flow as a result of the bridge deck sectional model test Vortex shedding excitation of the vertical bending mode occurs in the equivalent full-scale wind speed range from 10 to 30 m/s In Japan, 100 Gal(7) is used as the allowable acceleration in many cases, and if converted into an amplitude, the allowable amplitude is 0.123 m The response value observed in the test was more than twice the allowable amplitude, so this was problematic from a design point of view In the torsional mode, torsional flutter occurred at an equivalent full-scale wind speed of about 60 m/s However, because this onset wind speed was sufficiently high with respect to the reference wind speed, there was no problem with torsional flutter In this bridge, a torsional mode got coupled with a horizontal bending mode, and the center span and side spans were different in bridge deck depth These and other factors could not be considered in the bridge deck sectional model test Therefore, it was thought necessary to evaluate the aerodynamic stability based on the results of both the sectional model and full bridge model tests Thus, a full Sectional model test results (at completion, in smooth flow) 67 Vo l 39 N o A u g u s t 2006 bridge model test was conducted using the threedimensional aeroelastic model in smooth flow and turbulent boundary layer flow 3.2 Full bridge model test The full bridge model test was conducted using the industrial aerodynamics wind tunnel (test section’s sectional area size: 6.0 m wide and 3.0 m high) at IHI Research Laboratory (Yokohama) Figure shows the three-dimensional aeroelastic model (model scale = 1/120) used for the test In contrast to the sectional model test taking up the bridge deck alone, the full bridge model test uses a three-dimensional aeroelastic model which is (a) Bridge in complete state similar to the real bridge in its entire shape and vibration characteristics, including the towers, cables and others Therefore, the latter offers response characteristics that are closer to the real bridge For the vibration modes considered in the model design, their natural frequencies and natural vibration mode shapes were investigated by analyzing the real bridge and measuring the full bridge model Table and Fig show these analyzed and measured values for comparison For the first and second vertical bending and torsional modes, one can see that the vibration mode shapes were well reproduced For high-order vertical bending modes, Table (b) Bridge during erection Comparison of natural frequencies between real bridge and full bridge model (at completion) Analysis value of Required value of Measured value of full bridge model*2 full bridge model real bridge*1 (Hz) (Hz) (Hz) Vibration mode (Note) Model scale : 1/120 Fig First vertical bending 0.453 4.96 5.03 Second vertical bending 0.526 5.76 6.02 Third vertical bending 0.659 7.22 7.38 Fourth vertical bending 0.961 10.53 First torsional 0.969 10.6 (Note) *1 : fp *2 : fm' = Full bridge model − (b) Second vertical bending mode Vibration mode Vibration mode (a) First vertical bending mode − − − − − − − − − Distance from center (m) Distance from center (m) − (d) Fourth vertical bending mode Vibration mode Vibration mode (c) Third vertical bending mode − − − − − − 9.45 10.6 − − − Distance from center (m) Distance from center (m) Vibration mode (e) First torsional mode (Note) : Analysis value of real bridge : Measured value of full bridge model − − − − − Distance from center (m) Fig Comparison of vibration modes between real bridge and full bridge model (at completion) 68 120 × fp Vo l 39 N o A u g u s t 2006 however, discrepancies were ascertained in vibration mode shape between the real bridge and its model Therefore, their effects were corrected when evaluating the response As described above, the test was performed not only in smooth flow, but also in turbulent boundary layer flow For the wind characteristics at the bridge site, no adequate data based on field observation was available, so the turbulent boundary layer characteristics were established based on the standards(7) of Japan The measured values of the turbulent boundary layer characteristics used for the wind tunnel test are shown in Table Table Turbulent flow characteristics in the generated boundary layer Unit Item Turbulence intensity Turbulence scale*3 3.2.1 Response of completed bridge Figure shows the bridge response in relation to wind speed in smooth flow The sectional model test results shown in Fig also show the bridge response under the same conditions, that is, the response of the completed bridge in smooth flow Table shows a comparison of the onset wind speed and response amplitude of vibration obtained from these two kinds of tests One can see a good agreement of torsional flutter onset wind speed between the two On the other hand, a discrepancy was ascertained in the vortex shedding excitation amplitude of the two In general, response amplitudes of the vortex shedding excitation of the sectional model test and full bridge Target value of turbulent Measured value of employed boundary layer*1 turbulent boundary layer*2 Streamwise component Iu % 15.5 15.3 Vertical component Iw % 7.5 8.1 Streamwise component Lu /B — 3.8 3.8 Vertical component Lw /B — 1.7 1.3 (Note) *1 : Shows the value based on the Wind Resistant Design Manual for Highway Bridges.(7) *2 : Shows the value at the design criteria wind speed of Vd = 35.4 m/s *3 : Shows the value made dimensionless by B (bridge deck width) (a) Angle of attack a = 0°, Wind direction b = 0° 2.5 0.8 2.0 Design criteria wind speed Vd = 35.4 m/s 0.6 1.5 0.4 1.0 0.2 0.5 0.0 10 20 30 40 50 60 70 80 Vertical bending amplitude of real bridge (m) 1.0 : First vertical bending mode (d = 0.019) : Second vertical bending mode (d = 0.023) : Third vertical bending mode (d = 0.017) : Fourth vertical bending mode (d = 0.024) : First torsional mode (d = 0.022) 1.2 0.0 90 1.0 Table 2.5 0.8 2.0 Design criteria wind speed Vd = 35.4 m/s 0.6 1.5 0.4 1.0 0.2 0.5 0.0 10 Full-scale wind speed (m/s) Fig 3.0 20 30 40 50 60 70 80 Torsional amplitude (degrees) 3.0 : First vertical bending mode (d = 0.022) : Second vertical bending mode (d = 0.023) : Third vertical bending mode (d = 0.018) : Fourth vertical bending mode (d = 0.018) : First torsional mode (d = 0.023) Torsional amplitude (degrees) Vertical bending amplitude of real bridge (m) 1.2 (b) Angle of attack a = +3°, Wind direction b = 0° 0.0 90 Full-scale wind speed (m/s) Full bridge model test results (at completion, in smooth flow) Comparison between sectional model test results and full bridge model test results Wind direction with respect to bridge Angle of attack : a = +3° Wind direction : b = 0° Vortex shedding excitation Torsional flutter Vibration phenomenon Wind speed at occurrence of maximum response (First vertical bending mode) Maximum response amplitude Onset wind speed (Wind speed at which amplitude reaches degree) Unit Sectional model test results Full bridge model test results m/s 24.9 20.5 m 0.381 0.438 m/s 68.0 70.0 (Note) Values after conversion into full-scale values are shown 69 0.8 : Static displacement : RMS*2 amplitude : Peak amplitude 0.6 Design criteria wind speed Vd = 35.4 m/s 0.4 0.2 0.0 −0.2 10 20 30 40 50 60 70 Full-scale wind speed (m/s) (Note) *1 : Span center *2 : Root mean square Angle of attack : a = 0° Wind direction : b = 0° Fig 11 Gust response amplitude of real bridge*1 (m) model test not agree with each other completely As a cause for this in the case of this bridge, one can cite the fact that, in the sectional model, the bridge deck changes in its sectional shape along the bridge axis, and the aerodynamic forces acting on the towers, cables and other structural members other than the bridge deck cannot be taken into account From Fig 9, the response amplitude of vortex shedding excitation tended to be larger for upward flow (angle of attack, a = +3°) than for horizontal flow (angle of attack, a = 0°) In the full bridge model test, the influence of the wind direction was also examined The wind direction is a value to define the wind direction in a horizontal plane, and the transverse direction of the bridge axis is a wind direction of b = 0° From Fig 10, one sees that the amplitude of vortex shedding excitation decreased in the situation of a quartering wind of b = 30° In turbulent boundary layer flow, the vortex shedding excitation response disappeared and was replaced by vibration characteristics in which the gust response in the vertical bending mode was dominant, as shown in Fig 11 Although not shown in this paper, the gust response in the horizontal bending mode had a small amplitude, that is, the response level was about one seventh of that in the vertical bending mode As shown in Fig 12, wind direction had a small influence on gust response 3.2.2 Response of bridge during erection Figure 13 shows the response characteristics of the bridge at an erection stage The structural system under consideration was the bridge just before closing of the center span, which was deemed most uneasy in view of its aerodynamic stability Its bridge deck section was the section of the completed bridge deprived of the railings, curbing concrete blocks and inspection cart rails On the full bridge model, scaffoldings under the bridge deck and erection cranes, temporary facilities on the deck, were also reproduced Vertical bending response of real bridge*1 (m) Vo l 39 N o A u g u s t 2006 Dynamic response in turburent boundary layer flow (at completion) : Vertical bending RMS amplitude : Vertical bending peak amplitude : Horizontal bending RMS amplitude : Horizontal bending peak amplitude 0.25 0.20 0.15 0.10 0.05 0.00 10 15 Wind direction b (degrees) (Notes) *1 : Span center Angle of attack : a = 0° Response amplitude of vortex shedding excitation of real bridge (m) Fig 12 Relationship between wind direction and gust response (at completion) : First vertical bending mode : Second vertical bending mode : Third vertical bending mode 0.5 0.4 0.3 0.2 0.1 0.0 15 30 Wind direction b (degrees) (Note) Angle of attack : a = 0° Fig 10 70 Relationship between wind direction and response amplitude of vortex shedding excitation (full bridge model test results, in smooth flow) In the bridge during erection, the vortex shedding excitation found in the completed bridge did not occur In the deck erection stage of a general long cable-stayed bridge, gust response often becomes a problem in the design For this bridge, however, the test in the turbulent boundary layer flow offered the result that, at the design criteria wind speed during erection of 23.6 m/s, the vertical gust response caused a maximum amplitude of about 60 mm at the bridge deck end and the horizontal gust response caused a maximum amplitude of about 20 mm, these values ensuring that there is no structural problem Vo l 39 N o A u g u s t 2006 1.0 : First vertical bending mode (d = 0.017) : First horizontal bending mode (d = 0.020) : First torsional mode (d = 0.042) 0.8 2.5 2.0 Design criteria wind speed during erection Vd = 23.6 m/s 0.6 1.5 0.4 1.0 0.2 0.5 0.0 0.0 10 20 30 40 50 60 70 80 Torsional amplitude*1 (degrees) Vertical bending amplitude of real bridge*1 (m) Horizontal bending amplitude of real bridge*1 (m) (a) Dynamic response in smooth flow 90 Full-scale wind speed (m/s) (Note) *1 : Cantilever erection end Vertical bending response of real bridge*1 (m) (b) Dynamic response in turbulent boundary layer flow : Static displacement : RMS*2 amplitude Design criteria wind speed during erection Vd = 23.6 m/s : Peak amplitude 1.0 0.8 0.6 0.4 0.2 0.0 −0.2 10 20 30 40 50 60 70 Full-scale wind speed (m/s) (Notes) *1 : Cantilever erection end Angle of attack : a = 0° Wind direction : b = 0° Fig 13 Dynamic response of bridge during erection Conclusion The aerodynamic stability of the “Binh Bridge” at completion and during erection was evaluated by a wind tunnel test As a result, it was found that the safety is sufficient against torsional flutter and other divergent vibrations; however, there is a possibility of vortex shedding excitation occurring in smooth flow Considering the wind characteristics at the site, however, a vibration-suppressing effect can be expected from the wind turbulence, and this was verified by a test in turbulent boundary layer flow According to the statistical wind speed data at and around the bridge site, the occurrence frequency of the wind speed causing vortex shedding excitation is as low as 225 days in 125 years of bridge life Considering this, it was decided not to apply damping measures at present — Acknowledgments — This wind tunnel test was conducted as a part of the fabrication and erection work of the “Binh Bridge.” We sincerely thank BPMU (Haiphong Bridge Projects Management Unit) and Chodai Co., Ltd., for their giving us generous guidance in conducting the test REFERENCES (1) P A Irwin : Wind Tunnel Tests of Long Span Bridge, Proc 12th Congress IABSE (1984) (2) R L Wardlaw : The Improvement of Aerodynamic Performance, Aerodynamics of Large Bridges (1992) (3) A A C Wallace : Wind Influence on Kessock Bridge, Engineering Structures Vol.7 (1985) (4) Y Kubo, K Sadashima, E Yamaguchi, K Kato, Y Okamoto and T Koga : Improvement of Aeroelastic Instability of Shallow π Section, Journal of Wind Engineering and Industrial Aerodynamics 89 (2001) pp.1 445-1 457 (5) Honshu-Shikoku Bridge Authority : Wind Resistant Design Criteria (1976) (6) Honshu-Shikoku Bridge Authority : Wind Tunnel Test Guideline for Akashi-Kaikyo Bridge 1991 (7) Japan Road Association : Wind Resistant Design Manual for Highway Bridges 1991 (8) M Hoshino : Creep Analysis and Torsional Vibration Analysis of Cable-Stayed Bridges with Two Edge Composite, Journal of the Japan Society of Civil Engineers No.543 I-36 July 1996 pp.239246 71 ... model test taking up the bridge deck alone, the full bridge model test uses a three-dimensional aeroelastic model which is (a) Bridge in complete state similar to the real bridge in its entire shape... Table (b) Bridge during erection Comparison of natural frequencies between real bridge and full bridge model (at completion) Analysis value of Required value of Measured value of full bridge model*2... : Analysis value of real bridge : Measured value of full bridge model − − − − − Distance from center (m) Fig Comparison of vibration modes between real bridge and full bridge model (at completion)
- Xem thêm -

Xem thêm: WTT_Binh bridge, WTT_Binh bridge

Gợi ý tài liệu liên quan cho bạn

Nhận lời giải ngay chưa đến 10 phút Đăng bài tập ngay