Bayesian-based bridge influence line identification and uncertainty estimation

Abstract

Being a unique characteristic of a structure, the identification of an accurate bridge influence line from real measurement of bridge response under moving loads is on high demand for new structural design, model calibration, structural condition assessment and bridge weigh-in-motion. Despite its uniqueness, the identified influence lines from different measured signals tend to differ due to perturbations owing to the dynamic effect, road surface roughness, difference in calibration trucks and so on. This paper proposes a robust method using Bayesian framework for the identification of a single influence line by considering multiple measured signals. The effectiveness of the proposed method is demonstrated through a field experiment on a simply-supported steel girder bridge using three calibration trucks with different axle numbers and axle configurations. Using Bayesian framework, the uncertainties associated with the identified influence lines are also estimated which are useful for knowing the accuracy of predicted signals and estimated axle weights using a particular influence line. © 2023 The Author(s).