Long-term monitoring data of wind velocities and accelerations on the Hardanger Bridge are used to investigate the relationship between the wind-loading and response processes. The extensive measurement system consisting of 20 accelerometers and 9 anemometers is described as well as the local topography of the site. The wind and response characteristics are presented using scatter plots and wind rose diagrams. The considerable variability observed in the bridge dynamic response is investigated by utilizing response surface methodology. Simple parameters of the wind field are selected as the predictor variables in the analyses. The variability in response is attributed to the variable wind field, and the effects of the significant parameters on the response are presented in a statistical framework. The agreement of the findings with previous considerations and the implications on the design of long-span suspension bridges are discussed.

As Storm Tor struck the western coast of Norway, wind speeds and bridge deck accelerations along the Hardanger Bridge girder were recorded by the monitoring system installed on the bridge. Using 13.5 hours of data, mean wind speed, turbulence intensities, gust factor, turbulence length scales, angle-ofattack, and one-point and two-point turbulence spectra are studied using 10-minute stationary averaging intervals. Using the measured turbulence statistics as inputs, the buffeting response of the bridge deck is calculated in the frequency domain. The calculated response is compared with the measured response in terms of the root-mean-square (RMS) of acceleration and displacement components and the power spectral density of the acceleration response. Significant discrepancies are found in the case of the vertical response. Predicting the spectral response is found to be more difficult than predicting the RMS response, in particular for high-frequency responses. Considering the spanwise non-uniformity of turbulence statistics did not affect the predictions significantly.

10Wind-induced vibrations of the Hardanger Bridge deck were studied with reference to turbulence 11 characteristics at the bridge site to evaluate the performance of the state-of-the-art methods for buffeting 12 response analysis. Long-term monitoring data from an extensive monitoring system were used to obtain 13 the bridge vibrations and wind characteristics. The acceleration response of the bridge was calculated 14 in the frequency domain using multimode buffeting theory. Design regulations were used directly and 15 also modified using measurement data to deduce the wind turbulence spectra. The aerodynamic 16properties of the bridge section obtained from previous wind tunnel tests were used in the analyses. The 17 predicted root mean square (RMS) acceleration response was compared to the measured response. The 18 analysis using the design methodology gave underestimations of the measured responses. The use of 19 average values of wind statistics obtained from the monitoring data only slightly improved the results. 20When the variability of the wind field was reflected into the design method by using the probability 21 distributions of the wind field parameters, more satisfactory design curves were obtained. 22

The quality of the information about the wind field characteristics is crucial for accurately predicting the structural response of a long-span bridge subjected to dynamic wind loads. In this paper, in situ mast measurements and terrain model wind tunnel tests are compared with full-scale measurements of the wind field along the Hardanger Bridge girder. The aim is to investigate the performance of mast measurements and wind tunnel terrain model tests in predicting the wind field characteristics for longspan bridges in complex terrains. Wind field spatial variations and statistical distributions for the mean wind velocity and turbulence intensity are investigated. Extreme value statistics have been applied to compare the mean wind velocity recordings from two different measurement periods. Results showing terrain-induced effects on the wind directions, turbulence intensities and mean wind velocities are presented. Simultaneous spanwise wind profiles for the mean wind velocity and along-wind turbulence intensity are compared between the terrain model wind tunnel tests and the full-scale measurements, and large nonuniformities are identified. The extreme profiles of the turbulence intensities vary as much as 100% along the span, and the mean wind velocity profiles vary up to 50% along the span.

No abstract