Influence of vehicle traffic on modal-based bridge monitoring

Abstract

In recent years, there has been a surge of interest in modal-based monitoring systems, especially for bridge structures, to detect damage at an early stage and to extend the life of existing bridges as well as to minimize maintenance costs. The modal parameters, natural frequencies and mode shapes, depend on stiffness and mass of a structure. For an economic application of a modal-based monitoring system, the estimation of the modal parameters based on the Operational Modal Analysis needs to be estimated. The Operational Modal Analysis uses the naturally existing vibration sources, such as wind and traffic, to estimate the modal parameters. A significant advantage is therefore that the bridges do not have to be closed to traffic. In this context, the modal parameters are influenced by the vehicle traffic due to the associated temporally and spatially variable mass, stiffness and damping properties of the overall system. In this paper, the influence of those operational loads and road roughness on the modal parameter estimation is numerically investigated. The vehicles are modeled as half-vehicle models. In addition, the influence of vehicle traffic is shown using in situ measurements on an existing bridge. A high sensitivity of the natural frequencies to operational loads can be shown, so that they are not reliable damage identification parameters. In contrast, the mode shapes exhibit small amplitude scatter due to the operating loads if the excitation frequency is not in the range of the natural frequency of the bridge. This is very unlikely, especially for bridges under vehicle traffic, due to the variety of vehicle designs and thus the broadband excitation.