Damages of bridge bearings detected from asymmetry of time-varying frequencies measured by a test vehicle

A novel method is firstly proposed for detecting damages in bridge support bearings from the asymmetry of time-varying frequencies extracted from vehicle responses. Initially, closed-form solutions are derived for the system’s time-varying frequencies, i.e., instantaneous frequencies (IFs), of a vehicle traversing a bridge supported by elastic springs. It is revealed that the bridge’s IFs are strongly correlated with its mode shapes. When the end supports degrade symmetrically, the mode shape remains symmetric; conversely, significant stiffness loss on one side results in pronounced asymmetry. Subsequently, a technique is proposed for damage detection in bridge end supports based on the vehicle response, which comprises three steps: (1) vehicle acceleration responses are collected using a single onboard accelerometer, and the bridge’s IFs are extracted via the Synchroextracting Transform (SET); (2) overall stiffness degradation is evaluated using the Average Offset Ratio (AOR) of the extracted IFs; and (3) asymmetric damage is identified by quantifying the spatial asymmetry of IF distributions through the Difference in Shannon Entropy (DSE). Numerical simulations demonstrate that the method effectively detects different support damages and is robust against road roughness and environmental noise. Additionally, model-scale experiments are conducted, and the engineering feasibility of the proposed technique is validated. Owing to its simplicity in equipment requirements, the method enables rapid and cost-effective screening of bridge-end support conditions and offers a new perspective for substructure health monitoring.