Please use this identifier to cite or link to this item:
|Title:||Buffeting response of long span cable-supported bridges under skew winds : field measurement and analysis|
|Keywords:||Bridges -- Aerodynamics|
Bridges, Cable stayed -- Mathematical models
Suspension bridges -- Mathematical models
Hong Kong Polytechnic University -- Dissertations
|Publisher:||The Hong Kong Polytechnic University|
|Abstract:||This thesis mainly focuses on the development and verification of a modified method in frequency-domain for the analysis of buffeting response and the understanding of buffeting behavior of long span cable-supported bridges under skew winds, supported by a series of wind tunnel tests and field measurements.|
First, the Tsing Ma suspension Bridge in Hong Kong and the Wind And Structural Health Monitoring System (WASHMS) installed by the Hong Kong Highways Department in the Bridge are introduced. The field measurement data of the Tsing Ma Bridge during Typhoon Victor recorded by the WASHMS are analyzed for both wind structure and bridge response. It is demonstrated that for a long span cable-supported bridge located in a complicated topography, typhoon-induced winds seldom attack the bridge at a right angle. Important wind characteristics such as mean wind speed and turbulence intensity vary along the bridge longitudinal axis. The measured response of main cables by the WASHMS also cannot be utilized to facilitate the comparison between field measurement and analysis if the conventional frequency-domain buffeting analysis method is used. A modified frequency-domain method for long span bridges under skew winds is thus required for a reasonable comparison between field measurement and analysis and a better understanding of bridge behavior under skew winds.
The modified buffeting analysis method of long span bridges is then developed in this study based on the quasi-steady linear theory and the oblique strip theory in conjunction with the pseudo excitation method and the finite element method. It is capable of performing fully coupled 3D buffeting analysis of long span bridges under skew winds and naturally including the effects of inter-modes, multi-modes, and spatial modes as well as the interaction between bridge deck, towers and main cables. In the modified method, a concept of oblique strip along the mean wind direction is introduced and a set of universal expressions of six component buffeting forces associated with the oblique strip is derived. The general expression of cross-spectral matrix of buffeting forces is then presented after considering turbulence spatial correlation and aerodynamic admittance. Self-excited aeroelastic forces are also taken into consideration in the modified method by including 18 flutter derivatives of oblique strip under skew winds. A comprehensive computer program is then developed in accordance with the proposed modified method.
To implement the modified method for buffeting analysis of a long span bridge, it is essential to obtain six component aerodynamic coefficients for oblique strips of a bridge deck. To this end, a wind tunnel test technique for measuring such aerodynamic coefficients is developed in this study, which includes the design of sectional model, the development of test rig and measurement system, and the analysis of test results. A typical oblique sectional model of the Tsing Ma Bridge deck is manufactured and tested in a wind tunnel under a series of combination of mean wind inclination and yaw angle. The measured data are fitted to obtain the six aerodynamic coefficient curves as functions of wind inclination and yaw angle for the late comparison between field measurement and analysis. The obtained six aerodynamic coefficient curves are also used to examine the cosine rule currently used in the buffeting analysis of long span bridges under yaw winds. It is found that the traditional cosine rule may significantly underestimate buffeting forces, in particular at large yaw angles.
Self-excited aeroelastic forces, often coming on the scene in the form of aeroelastic damping and stiffness, have been recognized as not only the key factors for bridge stability but also one of the important parameters for bridge buffeting response. To include this effect in the buffeting analysis of long span bridges under skew winds, a wind tunnel test technique is developed, which includes the design of oblique sectional model, the development of a dynamic test rig, and the application of unifying least square (ULS) method for identification. The flutter derivatives of a typical oblique strip of the Tsing Ma Bridge deck under skew winds are measured. The measured data are fitted and analyzed to find the relationship between flutter derivatives, reduced frequency, wind inclination, and wind yaw angle. The results are also used to examine the approximate formulae derived from the skew wind theory proposed by Scanlan. In general, significant discrepancies exist except for lower reduced velocities, where the approximate formulae may give an acceptable estimation of flutter derivatives of a bridge deck under skew winds. Furthermore, the effects of wind yaw angle on the critical wind speed of the Tsing Ma Bridge are also investigated based on the SDOF torsional flutter theory. It is found that critical wind speed may not increase with increasing wind yaw angle.
To perform a more reasonable comparison of buffeting response of the Tsing Ma Bridge between field measurement and analysis, the aerodynamic coefficients of the bridge towers should be measured under yaw winds consistent with the measurements of aerodynamic coefficients and flutter derivatives of the bridge deck. For this purpose, a 3D stiff model comprised of a whole tower and a segment of bridge deck is designed and tested. The averaged aerodynamic coefficients of the upper and lower segments of both upwind and leeward tower legs and those of tower transverse beams under yaw winds are obtained. With additional comparative tests, the interference effects between the tower legs, the tower transverse beams, and the bridge deck are also discussed.
The comparison of buffeting response of the Tsing Ma Bridge during Typhoon Sam is then conducted. The field measurement data of wind structures surrounding the Bridge during Typhoon Sam are first analyzed and used as input data to the computer program. These data include the mean wind speed, wind inclination, wind yaw angle, auto and cross spectra of wind turbulence, fiction velocity, and others. The aerodynamic coefficients and flutter derivatives of the bridge deck and the aerodynamic coefficients of the bridge tower under the identified skew wind are then determined from the measured curves which are inputted into the computer program. Meanwhile, some empirical formulae are adopted for determining the drag and lift coefficients of the main cables for the identified skew wind. Finally, the computed buffeting acceleration responses of both bridge deck and cables are compared with those measured by the WASHMS, and the comparison is found to be satisfactory in general. The discrepancies of RMS response are, respectively, less than 24%, 17% and 17% in the lateral, vertical and torsional acceleration responses of the main span of the bridge deck. The differences of RMS acceleration response of the main span cables are not more than 20%, 5% and 6% in the lateral, vertical and longitudinal directions, respectively.
Finally, this thesis takes the Tsing Ma Bridge as an example to carry out a parametric study. The parameter study focuses on the effects of wind inclination and yaw angle on the buffeting response of the Bridge. The results reveal that the variation of buffeting response of the Bridge is not monotonous with wind yaw angle and inclination. The maximum bridge response may not occur at zero wind yaw angle, as indicated by the traditional cosine rule. For the Tsing Ma Bridge within a range of wind yaw angles (+-30o) and wind inclinations (+-5o), the maximum displacement response of bridge deck occurs at about +-12o (yaw angle) and 4o (inclination) in the vertical direction, about 5o (yaw angle) and -2.5o (inclination) in the lateral direction, and about 0o (yaw angle) and -2o (inclination) in the torsional direction. In particular, the lateral RMS displacement response of the bridge deck is not sensitive to wind yaw angle within a range of +-15o for all wind inclination concerned. Furthermore, the influences of aerodynamic crosswind force, yawing and rolling moments on bridge responses are rather small whereas the effects of flutter derivatives may be significant on bridge responses.
|Description:||1 v. (various pagings) : ill. ; 30 cm.|
PolyU Library Call No.: [THS] LG51 .H577P CSE 2002 Zhu
|Rights:||All rights reserved.|
|Appears in Collections:||Thesis|
Show full item record
Files in This Item:
|b16378222_link.htm||For PolyU Users||166 B||HTML||View/Open|
|b16378222_ir.pdf||For All Users (Non-printable)||11.36 MB||Adobe PDF||View/Open|
Checked on Jan 15, 2017
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.