Search

The extraction of damage-sensitive features based on deep learning for the vibration response caused by vehicle-bridge interaction (VBI) has become a mainstream method in damage identification. However, most studies remain at the stage of qualitative damage assessment, not achieving precise numerical quantification of damage. They also typically involve multiple frequent runs using only a single dedicated vehicle. In this study, utilizing a deep neural network designed for multi-objective regression tasks, a novel method for bridge damage localization and quantification is presented with a multi-objective regressor designed to create a direct mapping between the curve of bridge deflection in time domain and the damage information matrix. The final quantitative damage values are obtained through statistical analysis. The method is evaluated on a simulated data set generated from VBI simulation using various 3D heavy vehicle models. The results demonstrate that this method can accurately locate and quantify damage even with unseen vehicle load conditions, damage scenarios, and measurement noise. The structure and depth affect the effectiveness of extracting damage-sensitive features from time-domain deflection during training. The study shows potential for real-time damage identification under normal operating conditions of real bridges in the rapid development of intelligent transportation systems. • Supervised learning for damage identification is advanced with a multi-objective regression task that quantifies and localizes damage intuitively. • By learning the differencing deflection, real-time bridge damage assessment under various vehicle scenarios was achieved. • Comparing some mainstream networks, the differences in time and space feature extraction on dynamic deflection are explored. • It explores how network depth impacts feature extraction in classic deep residual neural networks for damage detection. • Deep networks' generalization is utilized for damage identification under unseen vehicle and bridge conditions. [ABSTRACT FROM AUTHOR]

The drive-by bridge dynamic measurement using an instrumented vehicle during its passage over the bridge can be an indirect bridge structural health monitoring system. The identification of the contact-point (CP) responses between vehicle wheels and bridge structure from vehicle responses offers an efficient and economical way for the modal identification and condition assessment of short- to mid-span road bridges. This paper proposes a novel method for the contact-point displacement response reconstruction of the vehicle-bridge interaction (VBI) system in a joint input-state estimation manner based on Bayesian Expectation-Maximization (BEM). An analytical model of VBI including the road approach in front of a simply-supported bridge is presented firstly. Then the augmented state-space model of the vehicle with the contact-point responses included as variables along with vehicle states is established. A new method by integrating the augmented Kalman filter (AKF) with a BEM strategy is proposed to solve the state estimation problem without knowing the vehicle axle responses. The vehicle states and CP displacement responses are identified simultaneously. The effects of the measurement noise, road surface roughness and vehicle speed on the identified results are investigated using numerical simulations. The experimental tests are used to further verify the feasibility and accuracy of the proposed method. The results show that the proposed method has potential for drive-by bridge condition assessment using on-board vehicle response measurements for a large population of short- to mid-span bridges. • A method integrating AKF and Bayesian Expectation-Maximization (EM) strategy is proposed to identify the CP responses. • The CP responses and its derivatives are included to the vehicle states to develop an augmented state-space model. • The process and measurement noises can be tracked by the EM strategy making the identification robust to the noises. • Numerical and experimental studies are conducted to verify the proposed drive-by bridge structural monitoring method. [ABSTRACT FROM AUTHOR]

This paper describes a novel method for detecting bridge damage that uses a partially instrumented vehicle fleet, each vehicle being instrumented with just one accelerometer. Importantly, no prior knowledge of the bridge or individual vehicle properties is required. Firstly, a model simplification concept is proposed to calculate the displacement under the vehicle wheel from single acceleration measurements on a half-car model. Then, optimisation is used to find individual vehicle properties using simulated noisy measurements from a fleet of vehicles. Finally, accelerations from the fleet are used to find the 'apparent profile difference', which contains bridge deflection data. The difference is found to be independent of surface profile but dependent on vehicle axle weight differences and a moving reference influence function (MRIF). When the axle spacings and the relative axle weights are reasonably consistent in a subset of the fleet, the MRIF can be simplified into a compound MRIF. Even when the MRIF and individual axle weights are not available on-site, the shape of the compound MRIF can be determined through an iterative process and used to monitor bridge health condition. Bearing damage is represented in this paper as an increase of rotational resistance of the bearings. The numerical results show that the damage severity and location can be identified. • A new drive-by fleet monitoring concept is introduced using a partially instrumented vehicle fleet without knowing individual vehicle properties. • A new concept, vehicle model simplification, is introduced to enable monitoring bridges using partially instrumented vehicles. • Vehicle wheel displacement is successfully obtained from vehicle accelerations and used in structural health monitoring. • The relation between bridge deflection under the wheel and vehicle load is modelled as a moving reference influence function. This influence function is simplified to monitor bridges. [ABSTRACT FROM AUTHOR]

[Display omitted] The dynamics of vehicle-bridge interaction (VBI) involve the adoption of two-dimensional (2D) and three-dimensional (3D) vehicle models in simulations. It is worth noting that the 2D vehicle model lacks several crucial degrees of freedom and a complete road roughness compared to the 3D vehicle model. Despite the common usage of the 2D vehicle model in VBI simulations, the difference between the two models has not received sufficient attention in previous studies. Therefore, this study aims to quantitatively assess the differences in bridge response between the 2D and 3D vehicle models. To achieve this goal, the study employs a VBI modeling method based entirely on general finite element software first. Then, a series of comparisons between the 2D and 3D vehicle models are implemented through a steel composite girder bridge and a cable-stayed bridge. The results indicate that road roughness significantly contributes to the observed differences. Specifically, the 2D vehicle model generates larger bridge responses than the 3D vehicle model, primarily due to the absence of rolling and vibration cancellation effects. Furthermore, the differences between the two models increase as the road surface roughness worsens, with a maximum difference of 25.07 % in the cable-stayed bridge's mid-span acceleration. Notably, the discrepancies in bridge acceleration are more significant than those in displacement or cable stress. Adopting the 2D vehicle model in VBI simulations may lead to more conservative engineering recommendations, resulting in higher economic costs and greater consumption of building materials. Thus, this study emphasizes the importance of considering the 3D vehicle model in VBI simulations, particularly for bridges with poor road conditions. [ABSTRACT FROM AUTHOR]

The consideration of interaction effects between vehicle and structure is an important aspect in the dynamic analysis of railway bridges. They usually lead to a reduction of the structural responses compared to computations without consideration of the vehicle-bridge interaction. The interaction effects can either be accounted for by Finite Element analysis including multi-body models of the vehicles or in a simplified way – as adopted by the actual regulations – by artificially increasing the bridge damping. In this way, a simplified dynamic analysis under consideration of the interaction effects is possible using the moving load model. However, several studies have shown that the current approach of the additional damping included in Eurocode 1 is non-conservative in many cases, possibly leading to an underestimation of the structural responses. The approach in Eurocode 1 as well as the more recently proposed approaches were based on a limited number of combinations of trains and bridges. In order to verify these approaches in a more thorough way, the present paper investigates the vehicle-bridge interaction effects based on 25 real high-speed passenger trains (including conventional and articulated trains) and over 100 single-span bridges with spans ranging from 5 m to 55 m. A typical criterion of analyzing the vehicle-bridge interaction phenomenon is the reduction of the structural acceleration computed using a 2D multi-body representation of the vehicle compared to the simplified moving load approach. Therefore, the reduction of the acceleration due to the interaction is discussed in detail and put in relation to the results of previous research. The results of over 800 000 dynamic analyses performed in the present study confirm that the additional damping approach depending on the bridge span as included in Eurocode 1 overestimates the actual vehicle-bridge interaction effects. Furthermore, the results show that recent additional damping approaches depending on bridge and train characteristics lead to more reliable additional damping values. Their application requires public availability of train characteristics and an extension of the approaches for articulated as well as heterogeneous trains. • Comprehensive collection of multi-body vehicle model characteristics. • Computational series of over 800 000 train crossing calculations with and without interaction. • Systematic analysis and critical discussion of different additional damping approaches. [ABSTRACT FROM AUTHOR]

Several breakthroughs have appeared in different applications due to deep learning which can be applied in Bridge Weigh-In-Motion (BWIM) systems as well. Therefore, deep learning applications should be examined in BWIM systems. However, some deep learning-based solutions have already been proposed in the international literature, they cannot be compared to static methods because available datasets are not appropriate to train and test artificial neural networks. In addition, numerous other aspects have been already considered during tests. In the current paper, a numerically simulated and validated comprehensive dataset is provided. It is designed to meet requirements of incremental development. This dataset makes it possible to benchmark static algorithms and deep learning-based methods on the same dataset. In the second part of the paper, a deep learning-based solution is proposed. The developed solution shows promising results on the provided dataset and also demonstrates the applicability of the dataset. • Deep learning-based algorithm is developed for Bridge Weigh-In-Motion system. • Validated FEA-based influence surfaces and strain gauge data are used as input. • BME-Simulated 1 (BME-S1) dataset with more than 100000 vehicle models is provided. • A BWIM system is developed with class B+ in accordance with COST 323 standard. [ABSTRACT FROM AUTHOR]

• A KF-based method is adopted to extract 191 vehicle-induced modal loads from a real bridge during its operation. • The extracted loads are used to propose a response spectrum model for the vibration design under vehicle dynamic excitation. • Factors affecting the shape and value of the response spectrum model are analyzed and considered. • The proposed model is validated in a field test. The load exerted from passing vehicles is a combination of the static moving load and dynamic load owing to the uneven bridge deck, which leads to bridge vibration. Vehicle-induced bridge vibration has a negative influence on many aspects, including driving safety and pedestrian comfort. Therefore, an efficient and accurate approach to predict bridge vibration due to the passage of vehicles is needed in the design stage. In this study, a response spectrum model is proposed to predict the bridge root-mean-square (RMS) acceleration response caused by a single-vehicle passage. The vehicle dynamic loads were extracted from a field test by employing a Kalman filter-based method. The extracted vehicle dynamic loads were then used to derive a mathematical model of the vehicle-induced loads. The proposed response spectrum model consisted of four parts: 0–2 Hz, 2–5 Hz, 5–7 Hz, and 7 Hz. A mathematical representation of each part was provided for different confidence levels. Additionally, the pavement roughness, mode shapes, and driving speeds affecting the shape of the design spectrum are discussed, and modification factors are suggested. Moreover, the procedure for applying the proposed response model for bridge vibration prediction under a vehicle passage is presented. The validation from field tests shows that the proposed method has decent applicability for the prediction of bridge vibration levels induced by passing vehicles. [ABSTRACT FROM AUTHOR]

• Framework for detecting damages in the constituents of rail-bridge systems from the vehicle-rail contact response. • New closed-form solutions for the dual-beam system responses by a frequency domain method. • IAS computed from the driving component of the contact response with high sensitivity for detecting damages. • Comprehensive parametric study for relevant factors. A theoretical framework is presented for detecting the damages in the constituents of track-bridge systems from the vehicle-rail contact response. The rail and bridge are modeled as dual beams interwoven by a viscoelastic layer to account for track fasteners, sleepers and ballast. The single-axle test vehicle is modeled as a single degree-of-freedom (DOF) system. Firstly, closed-form solutions are newly derived for the vehicle, dual beams and vehicle-rail contact response by a frequency domain method. Next, the instantaneous amplitude squared (IAS) derived from the driving component of the contact acceleration response is adopted as the damage index. The reliability of the proposed technique to detect single or multi damages of various severities in different locations of the constituents of the track-bridge system are numerically validated. Through the parametric analysis, it is confirmed that: (1) single or multi damages in the rail, interlayer and bridge can be well identified, especially for the rail; (2) the IAS computed from the driving component of the contact acceleration is baseline-free and sensitive to damages; (3) a linear relationship exists between the IAS and damage severity for the rail; (4) rail irregularity is harmful to damage detection; but tolerable for low disturbance; (5) higher vehicle speeds can be used to detect rail damages, but not for interlayer and bridge; (6) the simple supports and finite length adopted for the rail in the dual beam model are justified by comparison with the results for infinite rails; and (7) larger vehicle mass is beneficial to damage detection. [ABSTRACT FROM AUTHOR]

• Most common monorail bridges do not have an isolation system due to their low superstructure weight. • Comparing novel isolated monorail bridges, traditional monorail bridges, and conventional railway bridges. • Evaluating the seismic bridge's responses in the presence or absence of trains using 3D continuum and substructuring models. • Different stop locations of trains on the bridges have a minor influence on the bridge's seismic performance. • In contrast to railway bridges, for monorail bridges, the train presence creates critical conditions under combined vertical-horizontal excitations. Most of the common monorail bridges in the world are non-isolated ones due to their low superstructure weight. Recently, a limited study on the innovative isolated monorail bridges demonstrated that seismic isolation systems can be beneficial to monorail bridges. In this study, a parametric analysis of soil-pile-bridge-train (SPBT) interaction during frequent earthquakes was performed to better understand the seismic behavior of the novel isolated monorail bridges compared to the traditional non-isolated monorail bridges and the conventional railway bridges. To evaluate the influence of train-bridge interaction (TBI) on the seismic response of bridges under the combined vertical and horizontal ground motions, Taiwan Railway Bridge (TRB), and Qom Monorail Bridge (QMB) with and without isolators were selected. A total of 2268 time history analyses, including different isolator shear modulus, train locations in two vehicle modeling approaches, pier heights, and ground motions in different directions, were performed using the substructuring method. The results were verified by comparing the 3D continuum finite elements. The observations indicated that when lateral and vertical excitations were applied simultaneously, for monorail bridges where the trains' weight usually reached up to 50% of the weight of a couple of the guideway beams, the presence of the train led to creating the critical conditions. However, for the conventional railway bridges with relatively low train-to-deck weight ratios (less than 30%) ignoring the train could cause the most critical responses in the SPBT system, regardless of the vehicle modeling approach. It was also found that the efficiency of seismic isolation systems for conventional railway bridges is higher than that for monorail bridges. [ABSTRACT FROM AUTHOR]

• Overall damage assessment in masonry buildings under vehicular impact is studied. • FE model with material and contact nonlinearities predicts failure under Impact. • FE model captures vibration- induced damage propagation to wall edges. • Such damage in unreinforced masonry structures is more severe at low velocities. • Validated vehicle model with deformable characteristics is used to predict HIC. • HIC of 23.5 at impact velocity of 100 km/hr is less than tolerance limit of 1000. • This is due to vehicle penetration through the masonry wall at high impact velocity. Vehicular crashes into buildings seem to be an on-going problem with severe consequences. When the building is of masonry, the damage at the impact zone is more severe with possible intrusion of the vehicle into the building, depending on the velocity of impact. In all these cases, vibration is propagated from the impact zone to the wall edges and then to adjoining walls in the building and can result in their damage. While the damage at the impact zone has been studied, the vibration propagation to the edges of the impacted wall and the adjoining walls and their potential for damage have not been treated. Dynamic response and damage of masonry walls due to vibration caused by vehicular impacts is important for the global safety assessment of the masonry structures. With this in mind, this paper presents a numerical study on the vibration-induced damage characteristics of the masonry structures subjected to vehicular impacts. A homogenised masonry material model incorporating strain rate effects suitable for impact applications using layered shell elements is adopted in this research with improved computational efficiency. The vibration induced damage at the edges of masonry walls is studied through numerical models of various types of wall structures. A validated vehicle model with deformable characteristics is employed to predict the HIC (Head Injury Criteria) to evaluate the head injury risk of the occupants of the vehicle when it crashes into the masonry structure. The outcomes of this study demonstrated that the vibration-induced damage in unreinforced masonry structures due to vehicular collision is more severe at low velocities compared that at high-velocity impacts. Moreover, the HIC value calculated for impacting vehicle velocity of 100 km/hr is 23.5, which is considerably lower than the HIC tolerance limit of 1000. This is a desirable safety feature for the occupants of the impacting vehicle, and it is due to the significant amount of energy absorbed at the impact zone and through the vibration transmission across the masonry wall. [ABSTRACT FROM AUTHOR]

• Bridge frequency identification from moving vehicle at a high speed. • Combine dynamic responses of multiple vehicles moving via a high speed. • Track irregularity can be effectively eliminated using vehicle residual response. • The proposed method is verified through the MDOF vehicle model. The identification of bridge frequencies using dynamic responses of moving vehicles has been studied by many researchers, however, most of them assume that the vehicle moves at a relatively low speed. In the application of high-speed railway bridges, the bridge span is typically short and the train always moves at a very high speed, which implies that the duration of the vehicle traveling on the bridge is very short and the signal recorded from the vehicle is not long enough for signal processing to extract the bridge frequency. The present study proposes a new method to identify the bridge frequency from vehicles moving at a high speed by combining the responses of multiple vehicles. The theoretical analysis is conducted using the simple vehicle-bridge interaction (VBI) model. By subtracting the acceleration responses of adjacent vehicles at the same position on the bridge, the residual vehicle responses were obtained and combined, which eliminated the additional driving frequencies and avoided its interference on the bridge frequency identification. Numerical simulations are then conducted to verify the proposed method, which shows that by combining the acceleration of several vehicles to extend the overall duration of the signal, the frequency resolution can be improved and the bridge frequency can be successfully identified using vehicles travelling at a high speed. The influence of track irregularity, vehicle speed, bridge stiffness, and bridge damping on the identification results has also been studied. In addition, a multi-degree-of-freedom vehicle model is used to further verify the proposed method. [ABSTRACT FROM AUTHOR]

• The feasibility assessment of drive-by bridge inspection is investigated. • A novel index to quantify the transmissibility performance is proposed. • An induced change is identified through drive-by bridge inspection. • Successful identification of the first three modes of the bridge is verified. This paper presents results of numerical and experimental investigations for the feasibility assessment of drive-by bridge inspection, where the health status of a bridge is solely identified from the dynamic response of an inspection vehicle, passing over a target bridge. To this aim, a novel metric based on the transmission performance between the vehicle and the bridge is proposed, and it is demonstrated that the vehicle having a higher sensitivity index has superior performance, in moving state, to identify bridge frequencies, and consequently separate different states of a bridge from one another. In addition to numerical validation, experimental validation from a laboratory-based experiment is presented, where a two-axle vehicle model is adopted as an inspection vehicle and a single-span steel beam is employed as a target bridge. Vibration data are collected using three highly sensitive accelerometer sensors installed on the vehicle, one at the middle of each axle, i.e., the front axle, and the rear axle, and the third one at the center of gravity of the vehicle, to measure the pitching and bouncing modes of the vehicle. The contributions and the key findings of the work are as follows: First, this paper presents one of the early attempts of experimentally validating the feasibility of drive-by bridge inspection to identify an induced change in the structure of a bridge which has caused less than 2% variation in the fundamental frequency. Second, a novel index for quantifying the performance of transmission between the bridge and the vehicle is proposed. It is verified that a vehicle with a higher sensitivity index captures more bridge-related information, hence making the bridge modal frequencies more identifiable. Third, through experiments, successful identification of the first three vibration modes of the bridge, obtained from the moving vehicle tests, is verified under the condition of constant and slow speed, when there is ongoing excitation on the bridge. Fourth, the indices and criteria presented herein would be helpful in designing a real drive-by inspection system which can be potentially used for drive-by bridge inspection of a network of bridges with similar structural design in practice. [ABSTRACT FROM AUTHOR]

• The common Additional Damping Method can lead to non-conservative results. • Shortcomings of the common Additional Damping Method are identified and eliminated. • A parametric study shows vehicle-bridge interaction effects for four train types. • A new approach to additional damping delivers conservative lower bounds. In the dynamic analysis of simply supported railway bridges, the choice of vehicle models directly affects the quality of computed results. Simple moving load models usually overestimate the bridge response in comparison with models taking into account vehicle-bridge interaction effects. In order to compensate for this discrepancy, the Eurocode allows the increasing of structural damping for analyses performed with a moving load model. The specified additional damping is based on investigations carried out by the European Rail Research Institute in 1999 and may yield non-conservative results. This paper describes an extensive parametric study inspired by the common Additional Damping Method, aiming to eliminate its identified shortcomings. The applied input parameters for the study are based on existing European simply supported bridge structures with a ballast layer. In order to match the vertical accelerations computed with the Detailed Interaction Model, the additional damping for the Moving Load Model is determined for four European high-speed trains. The results show that the additional damping as integrated in the Eurocode is non-conservative. A redesigned approach enables the definition of conservative additional damping functions depending on the natural bridge frequency, the bridge mass and the train type. [ABSTRACT FROM AUTHOR]

• Establishes a framework to identify bridge frequencies from a moving test vehicle. • Gives theoretical support to eliminate the negative effect of road roughness. • Adopts a dimensionless description to cope with the time-varying nature of VBI. • The method is verified via numerical experiments with both SDOF and MDOF vehicles. • Allows identification considering practical vehicle speeds and poor road surface. This study establishes a short-time stochastic subspace identification (ST-SSI) framework to estimate bridge frequencies by processing the dynamic response of a traversing vehicle. The formulation uses a dimensionless description of the response that simplifies the vehicle-bridge interaction (VBI) problem and brings forward the minimum number of parameters required for the identification. With the aid of the dimensionless parameters the analysis manages to successfully apply ST-SSI despite the time-varying nature of the VBI system. Further, the proposed approach eliminates the adverse effect of the road surface roughness using a transformed residual vehicle response obtained from two traverses of a vehicle at different speeds over the bridge. The study verifies the proposed ST-SSI approach numerically: it first performs the dynamic VBI simulations to obtain the response of the vehicle, and then applies the proposed ST-SSI method, assuming the dynamic characteristics of the vehicle are available. The numerical experiments concern both a sprung mass model and a more realistic multi-degree-of-freedom (MDOF) vehicle model traversing a simply supported bridge. The results show that the proposed approach succeeds in identifying the first two bridge frequencies for test-vehicle speeds much higher (e.g., 10 m/s = 36 km/h and 20 m/s = 72 km/h) than previously considered, even in the presence of high levels of road surface roughness. [ABSTRACT FROM AUTHOR]

• During an anti-tank mine explosion, torsion bars are a threat to the crew. • Torsional Springs radically increase workspace inside the tracked vehicle. • Torsional Springs decrease mass of the tracked vehicle. • Hyperbolic torsion springs have the stiffness characteristics as the torsion bars. • This kind of torsional springs may radically change tracked vehicles construction. In the article, the authors present an advanced model of hyperbolic torsion spring based on the Finite Element Method (FEM). This model takes into account the contact forces between the spring arms, which allows you to simulate its operation even in the case of a large twisting angle. On the basis of this FEM model, the spring stiffness characteristics were constructed. Based on this characteristics, the existing 2S1 track platform model has been rebuilt. In this model, a much higher mass of the body is allowed than in the original model. On the basis of the elastic spring model, parameters such as the static deflection of the suspension have also been determined, as well as its range of mobility compared to the original suspension based on torsion bars. In addition, spring deflection phases have been presented in terms of internal stresses, which are important for the strength analysis of the spring. The originality of the described solution lies in the unique design of the so-called a hyperbolic spring. This design has not been previously described in the literature. [ABSTRACT FROM AUTHOR]

A rigid-mass vehicle model with wheels modelled as unsprung mass is newly proposed considering the pitching effect, which enables us to obtain the first three bridge frequencies from the vehicle's spectrum even in the presence of surface irregularity and damping. As the vehicle moves along the bridge, it generally confronts a bumpy ride due to the presence of surface irregularity. To investigate the up and down of the vehicle body on the identification of frequencies, the use of rigid-mass vehicle model was introduced in the literature. Nevertheless, such a simple model cannot accurately capture the dynamic responses of vehicle and bridge. To this end, this work comes out a three-mass vehicle model by including a rigid-mass and two unsprung masses. Important factors such as unsprung mass, vehicle damping, bridge damping, and surface irregularity are discussed. The pitching effect is unveiled by considering three different locations for installing sensors in the vehicle body. The numerical simulation shows that the proposed vehicle-bridge interaction system is able to identify vehicle's vertical and rotational frequencies if the sensor is installed in the suitable locations of vehicle body. [ABSTRACT FROM AUTHOR]