Modelling and identification of the dynamic response of a supported bridge

This paper describes the dynamic tests performed on a simply supported brid ge in Northern Italy under traffic excitation. The acceleration data have b een used for the identification of the natural frequencies, viscous damping ratios and mode shapes of the bridge. Modal parameters have been extracted using the wavelet estimation technique, previously implemented by the auth ors of this paper. This work represents the first attempt in using the wave let estimation technique directly on transient data and not on the impulse response estimates obtained via the random decrement technique. The capabil ity of the wavelet estimation technique for extracting modal parameters fro m transient time responses has first been inspected by analysing a simulate d set of data. The data have been obtained from the analytical continuous m odel of a three-span supported bridge. The bridge, excited by moving vehicl es, has been modelled as a supported orthotropic plate and its response has been evaluated using the convolution technique. The vehicles have been mod elled as multi-body systems, with linear suspensions and tyres flexibility, having globally seven degrees of freedom. An iterative procedure to includ e the dynamic interaction between the bridge and the vehicles has been impl emented. The real bridge, 20 m long approximately, has been monitored using six capacitive accelerometers, measuring the accelerations in seven points of its north edge in two points of its south edge. In particular, the acce lerometers on the south edge have been kept in fixed positions, acting as r eference points. On the north edge, one accelerometer has been kept fixed a t the mid-span location, while the remaining three have been positioned in two different set-ups. Each test has been repeated four times. The estimati on of the modal parameters has been performed three times, using as referen ce point each of the fixed accelerometers. The results obtained from each e stimation have been evaluated by means of a modal estimation 'quality index ' introduced within the wavelet estimation technique. (C) 2000 Academic Pre ss.