Characterization of the road profile and the rotational stiffness of supports in a bridge based on axle accelerations of a crossing vehicle

Abstract

AbstractIn an effort to find more cost‐effective and proactive ways to keep bridges in good condition, the use of instrumented vehicles has gained great interest in the last decade. Two bridge components that can wear rapidly are the bearings and the road surface. However, past research on drive‐by monitoring has placed focus mostly on detecting losses of bending stiffness in the bridge deck, while assuming ideal support conditions that may differ from real cases significantly, and ignoring the characterization of the road profile. Even further, the need for specialized vehicles equipped with high‐tech instrumentation, low speeds, or very good road profiles has been a major obstacle preventing its practical implementation. This paper investigates the use of axle accelerations from an ordinary two‐axle vehicle crossing the bridge to quantify the rotational stiffness of the supports and the height of the road irregularities while overcoming the limitations exposed above. In contrast to previous research where the response of the contact point has been derived from other vehicular locations based on complex differential equations of motion, transfer functions are employed here. The key advantage of transfer functions is their simple algebraic form that can be easily calibrated on the field. The road profile is then obtained by subtracting the displacement of the bridge under each axle from the displacement of the contact point. There is one prediction of the road profile per axle but only a unique value of rotational stiffness at each support that will yield the same prediction by both axles. The algorithm is successfully tested with a half‐car traveling at 5, 10, 15, and 20 m/s, over a 15‐m bridge beam model with ISO road classes “A,” “B,” and “C,” for boundary conditions ranging from simply supported to fixed. The solution's robustness to modeling inaccuracies and noisy data is also investigated.