2025 Volume 81 Issue 15 Article ID: 24-15022
Evaluating the residual capacity of seismically isolated bridges using vibration measurements in the operational condition is important from a life cycle management perspective. For this purpose, it is necessary to quantify the uncertainty in the hysteresis characteristics of rubber bearings, taking into account not only the ageing deterioration but also the spatial variability. In this study, a five-span seismically isolated bridge is modelled as a 2-dimensional finite element model and Bayesian model updating is performed using its seismic response data. The posterior distribution of a total of 20 model parameters is inferred using the transitional Markov chain Monte Carlo sampler. The results show that, by observing multiple points on either the girder or pier tops, the posterior distribution can accurately infer the varying hysteresis characteristics of the rubber bearings on different piers.
