ABSTRACT
Bridge frequency identification using passing vehicles has been demonstrated as a promising approach in the last two decades. Compared to the traditional method that requires sensor systems installed on the bridge, the indirect method only needs a few sensors mounted on passing vehicles, making it potential to monitor a large number of bridges rapidly and economically. However, currently, most studies just checked the existence of the bridge’s frequencies in the vehicle’s vibrations and explored to identify its frequencies accurately. Few studies investigated the damage detection of bridges using indirectly extracted bridge frequencies. This paper presents an indirect bridge damage detection method using bridge frequencies identified from a passing two-axle vehicle’s vibration data. Initially, the bridge’s finite element model is carefully built, updated, and divided into several substructures. Damage factors of different substructures are utilized to represent their damage degrees. Then, the bridge’s frequencies will be identified from the passing vehicle’s vibrations. At this stage, contact-point responses of the vehicle are back-calculated from its accelerations, and the residual contact-point responses of the two axles are employed to eliminate the vehicle’s frequencies and inverse effects of road roughness, making the bridge’s frequencies highlighted in the frequency domain. Thirdly, an objective function based on numerical experiments is proposed to locate and quantify the damage. The proposed method is verified numerically by a half-car model and a simply supported bridge in this paper. The results indicate that both the locations and degrees of damage can be identified under different influence factors and show great potential in practical applications.
