Computational modeling of vehicle-bridge interaction with experimental validation under moving loads for structural health monitoring.

This study investigates vehicle–bridge interaction under moving loads, with a focus on reduced-order modeling strategies applicable to structural health monitoring (SHM). Two computational approaches with increasing kinematic fidelity are developed and analyzed: a classical Euler–Bernoulli beam model and a slab-based Kirchhoff–Love plate formulation. Both models are coupled with multibody vehicle dynamics to predict bridge response under realistic traffic conditions.The beam model captures the dominant longitudinal bending response, whereas the slab formulation additionally resolves transverse deformation and bending–torsion coupling induced by lane-dependent vehicle excitation. Numerical predictions are validated against in situ measurements obtained from a reinforced concrete slab bridge instrumented with displacement and acceleration sensors, providing spatially resolved response data under operational moving load.Results show that the beam model accurately reproduces the global quasi-static response, while the slab model captures additional transverse and torsional effects that are not accessible to planar formulations. After model updating, the slab-based representation achieves close agreement with measured responses across a range of vehicle speeds and transverse load positions. The study demonstrates that slab-based reduced-order models provide a computationally efficient and mechanically consistent intermediate alternative between beam formulations and full three-dimensional finite element models. The proposed approach is particularly suitable for SHM applications, including model updating, inverse load identification and condition assessment under moving loads.