Xinqun Zhu

Recently, drive-by bridge structural health monitoring has gained increasing attention due to its potential to be a cost-effective way to monitor the highway infrastructure. The pre-installed sensory system on a passing vehicle is used to capture the spatiotemporal response of the bridge for structural health monitoring. The vehicle passing over the bridge is a time-varying process, and it is a big challenge to extract the time-varying characteristics of vehicle–bridge interaction systems for structural health monitoring. This paper aims to develop a drive-by time-varying feature extraction approach for bridge structural damage detection using the second-order synchrosqueezing transform. The research first examined the impact of various factors on the frequency changes in VBI systems, including the vehicle mass, stiffness, speed, road surface profiles, measurement noise, and different types of damage. When compared with traditional synchrosqueezing transform, the proposed method provides a clearer and more accurate time–frequency representation. A 6 m-long two-span bridge model is also built in the laboratory and the pre-installed wireless sensory system on a passing vehicle captures the vehicle and bridge interaction response. The time-varying features are extracted from dynamic responses of the vehicle passing over the bridge using the proposed method. Numerical and experimental results show that the proposed approach is effective and accurate enough to extract the time-varying features for bridge damage detection.

River-crossing bridges often utilise curved design to account for the highway profile change, distribute traffic loads and manage water flow. The vehicle-curved bridge interaction (VBI) is complicated, especially combining with environmental thermal effects. These thermo-mechanical interactions differ substantially from those in straight bridges and warrant detailed study. In this study, a curved bridge is modeled as a Timoshenko beam (TB) with pinned supports at both ends. Governing equations are derived using the four-node isoparametric beam element and cubic polynomial shape function. The vehicle is modelled as a half-car model to account for its bouncing and pitching motions. Thermal effects are represented by a linear temperature gradient through the beam depth, which is converted into equivalent thermal loads. An analytical framework that superimposes moving vehicle loads and thermal loads within the coupled motion equations is developed. The dynamic response of the VBI system is analyzed using a zero-order displacement and velocity overshoot (U ₀ -V ₀ ) iterative method. The parametric study reveals that: (1) the shear deformation significantly affects high vibration modes; (2) the bridge stiffness increases when the radius of its curvature reduces, which induces the reduction of bridge and vehicle displacement responses; (3) the thermal bending and expansion of the bridge have a large effect on dynamic amplification factor (DAF), while the impact of its curvature on dynamic load coefficient (DLC) becomes dominant for high vehicle speeds. These findings provide valuable insights for the design, analysis, and maintenance of curved bridges operating under combined vehicular and thermal loading conditions.

Civil infrastructure, such as bridges and buildings, is susceptible to damage from unforeseen low-speed impacts during service. Impact force identification from dynamic response measurements is essential for structural health monitoring and structural design. Force identification is an ill-posed inverse problem, and the regularization technique is widely used to solve this problem using a full transfer matrix. However, existing regularization techniques are not suitable for large-scale practical structures due to the high computational cost for the inverse calculation of a high-dimensional transfer matrix, and impact excitation locations are often unknown in practice. To address these challenges, a novel two-step truncated transfer matrix-based impact force identification method is proposed in this study. In the first step, a sparse regularization-based technique is developed to determine unknown force locations using modal superposition. In the second step, the full transfer matrix is truncated by time windows corresponding to short durations of impact excitations, and a Tikhonov regularization-based technique is adopted to reconstruct the time history of impact forces. The proposed method is verified numerically on a simply supported beam and experimentally on a 10 m steel–concrete composite bridge deck. The results show that the proposed method could determine the impact locations and reconstruct the time history of impact forces accurately. Compared with existing Tikhonov and sparse regularization methods, the proposed method demonstrates superior accuracy and computational efficiency for impact force identification. The robustness of the proposed method to noise level and the number of modes and sensors is investigated. Experimental studies for both single-force and multiple-force localization and identification are conducted. The results indicate that the proposed method is efficient and accurate in identifying impact forces.

Most existing data-driven-based methods for structural damage identification are based on black-box models without interpretability and explainability. In this study, a new parallel neural ordinary differential equation-based method is proposed for structural damage identification. The proposed method integrates the state-space equation-based physical model with a network block, which includes a group of parallel neural networks (NNs). The complex structural system is decomposed into physically interpretable subsystems, and each subsystem is represented by an independent NN. The group of parallel NNs is specifically designed to learn residual discrepancies between observed dynamic responses and predictions from the healthy structural system. The proposed approach explores the closed-form expression for a group of parallel NNs to identify the structural parameters, which increases the interpretability and enhances the reliability of the network model. A three-story building structure and a three-dimensional IASC-American Society of Civil Engineer benchmark frame are used to verify the performance of the proposed method. The results show that the proposed method is reliable and accurate to localise and quantify structural damage.

Structural damage in bridges is typically a local phenomenon. When a vehicle passes over the damage location, it induces a local response, which is highly sensitive to the damage. The interaction between the bridge and moving vehicle is a non-stationary time-varying process. The local damage can be accurately identified by analyzing the time-varying characteristics of the bridge response subjected to a moving vehicle. Synchrosqueezing transform, a reassignment method used to sharpen time–frequency representations, offers an effective tool to decompose the non-stationary signal into distinct components. This paper proposes a novel method based on multi-synchrosqueenzing transform to extract the time-varying characteristics of the vehicle–bridge interaction systems for bridge structural health monitoring. A vehicle–bridge interaction model is built to simulate the bridge under moving vehicles. Different damage scenarios of concrete bridges have been simulated. The effect of bridge damage parameters, the vehicle speed, the road surface roughness on the time-varying characteristics of the vehicle–bridge interaction system is studied. Numerical and experimental results demonstrate that the proposed method efficiently and accurately extracts the time-varying features of the vehicle–bridge interaction system, which could serve as potential indicators of structural damage in bridges.