Damage Detection of Steel-Truss Railway Bridges Using Operational Vibration Data

Abstract

In this paper, a damage identification framework for steel-truss railroad bridges, based on acceleration responses to operational train loading, is presented. The method is based on vertical and longitudinal sensor clustering–based time-series analysis of the operational acceleration response of bridges to the passage of trains. The results are presented in terms of damage features extracted from each sensor, which were obtained by comparing actual acceleration responses from the sensors to the predicted responses from the time-series model. Bridge damage was detected by observing changes in the damage features of the bridges as structural changes occurred in the bridges. The relative severity of damage was quantitatively assessed by observing the magnitude of the changes in the damage features. A finite-element model of a steel-truss railroad bridge was utilized to verify the method. Continuous condition assessment of railway bridges in this manner is deemed very valuable for the early detection of damage and, therefore, for increasing the safety and operational reliability of railway networks.