Damage Detection Based on Transmissibility of a Vehicle and Bridge Coupled System

Research on damage detection is commonly carried out in the frequency domain using methods based on the modal data, transfer function or frequency response function, and transmissibility. The motivation for using transmissibility to detect damages relies on the fact that transmissibility measurements are local quantities that suggest high sensitivity. Nevertheless, the measurement of transmissibility requires external harmonic forces as the known input and sensors directly attached to the structure. There is a need for methods that can perform damage detection without attaching sensors and knowing the details of the input excitations. In a vehicle-bridge coupled (VBC) system the vehicle can serve as a force exciter and its dynamic response can be easily measured. In this paper, to make use of the vehicle response the transmissibility of a VBC system was first theoretically derived in a system consisting of a simply supported beam and a single-degree-of-freedom vehicle. Then, a numerical study was conducted to investigate the feasibility of detecting bridge damages using the vehicle transmissibility of the VBC system. In the numerical analysis, several damage indicators were constructed based on the transmissibility from both the bridge and vehicle responses, and compared to each other to select a proper one for damage detection. A parametric study was conducted on factors such as the measurement numbers, road roughness, and vehicle speeds and numbers to analyze the effectiveness of using such indicators for damage detection. To make use of the mobility of moving vehicles, two methods were finally proposed to measure the transmissibility of vehicle responses by using two vehicles. Method I used one reference vehicle and one moving vehicle; Method II used two moving vehicles at a constant distance in the longitudinal direction along the bridge. Using these two methods, the modal information such as natural frequencies and modal shape squares was successfully extracted for damage detections.