88 E. Cheynet et al. 12.2 Instrumentation and Methods 12.2.1 The Bridge Site The Lysefjord Bridge, positioned at the narrow inlet of a fjord in the South-West part of the Norwegian coast, is used as a case study. Its main span is 446 m, with the central part 55 m above the sea level. The bridge is entrenched between two steep hills with slopes ranging from 30 to 45° and a maximum altitude of 350 m to the North and 600 m to the South. The bridge is exposed to winds that may descent from mountains nearby or follow the fjord over a longer path. The West side of the bridge is exposed to a more open and levelled area. Since 2013, the bridge has been instrumented with seven sonic anemometers installed on the West side of the bridge deck and four pairs of accelerometers located inside the deck (Fig. 12.1). GPS timing is used to synchronize the windand acceleration data, acquired locally with separate data logging units. Although the highest sampling frequency of the accelerometers and anemometers is 200 Hz and 32 Hz respectively, the data are acquired with a sampling frequency of 50 Hz and decimated to 20 Hz to facilitate data handling. Finally, a 3G router enables wireless data access and transfer via a mobile network. A more detailed description of the bridge instrumentation is given in e.g. [4]. Temperature measurements are provided by a Vaisala weather transmitter WXT520 located on hanger 10, denoted H-10, whereas the accelerometers are located near hangers 9, 18, 24 and 30, where hanger 18 corresponds to the mid-span position. 12.2.2 Automated SSI-COV Procedure The mode shapes and eigen-frequencies of the Lysefjord Bridge were successfully identified by the automated SSI-COV method used in the present study [4]. The modal parameters are hereby denoted using the code XYZ, where XDfH; V; Tg represents the lateral (H), vertical (V) and torsional (T) bridge motion. YD fS; Ag is the symmetric (S) or asymmetric (A) mode shape, and Z the mode number. For example HS1 refers to the first symmetric horizontal mode shape, and TA2 refers to North tower South tower 127m 72 m 24 m 24 m 48 m N S E W H-08 H-10 H-18 H-16 Accelerometer Measuring GNSS Reference GNSS Sonic anemometer H-09 H-20 H-24 H-30 x y z Fig. 12.1 Wind and structural health monitoring system installed on the Lysefjord Bridge Table 12.1 Parameters used in the SSI-COV method applied on Lysefjord bridge acceleration data max (s) Nmin Nmax fn MAC d 15 3 30 5e 3 3e 2 5e 3 2e 2 The automated SSI-COV method developed by Magalhães et al. [11] is applied on 6 months of acceleration records obtained from July 2015 to December 2015. For the sake of brevity, the automated SSI-COV method used will not be described explicitly in this paper but the details can be found in e.g. [10, 11]. The parameters used to calibrate this automated SSICOV method are summarized in Table 12.1. The minimal and maximal order of the system for the calculation of the stabilization diagram are denoted Nmin and Nmax respectively, whereas max is the maximal time lag used to compute the cross-covariance matrix. The three accuracy thresholds for the identified eigen-frequencies, modal damping ratios and modal assurance criterion [1] are denoted fn, and MAC respectively. Finally, the threshold accuracy for the cluster analysis is d.
RkJQdWJsaXNoZXIy MTMzNzEzMQ==