Dynamic Strain Response Research Articles

Experimental Simulation of Fluidity State Evolution from Slow Earthquake to Strong Earthquake

Slow earthquakes have been frequently detected in the neighboring regions of seismogenic zones. However, the physics underlying the connection of slow earthquakes to huge earthquakes remains one of the least understood aspects in seismic research today. By observing the strain time evolution of granular materials in laboratory experiments, we report that the fluidity parameter, which represents a time dependent rate for strain evolution, can be rejuvenated under progressively increasing dynamic loading and this physical quantity finally reaches a critical threshold state, and beyond that catastrophic and uncontained dynamic strain response begins. Moreover, memory effect and anti-fluidization effect in granular materials are also observed, and these effects underlay the long-term slip events preceded some strong earthquakes, including 2011 Mw 9.1 Tohoku earthquake and 2014 Mw 8.1 Iquique earthquake. The observations indicate the key role of development of fluidity state on connecting slow earthquakes and to full-scale earthquakes.

Graphene/SiCN Thin-Film Strain Gauges Fabricated by Direct Writing

The in situ strain/stress detection of hot components in harsh environments remains a challenging task. In this study, graphene/SiCN (G-SiCN) thin-film strain gauges (TFSGs) were fabricated on alumina substrates by direct ink writing (DIW). The percolation model and the piezoresistive effect of G-SiCN composites were systematically studied. On this basis, a TFSG with high conductivity (0.1 S/cm) and high sensitivity (gauge factor (GF) 9.9) of ceramic matrix conductive composites was fabricated. The graphene/SiCN TFSG has excellent static and dynamic strain response at room temperature. Subsequently, the strain dynamic test was conducted at 400 °C, and there was no attenuation of the GF, so as to verify the high-temperature performance of the G-SiCN TFSG. Therefore, G-SiCN TFSGs provide an effective approach for the measurement of the in situ static and dynamic strain of hot components in high-temperature environments.

Data-driven identification of structural damage under unknown seismic excitations using the energy integrals of strain signals transformed from transmissibility functions

In recent years, the methodology based on the energy variations of structural dynamic strain responses has been developed. Although it is an efficient data-driven structural damage identification approach, it can only be used for structures under impulse or known excitations. To overcome this limitation, an improved method is proposed in this paper for the rapid damage identification of beam-like structures using structural dynamic strain responses under unknown seismic excitations. The influence of different excitations to an intact and damaged structure, in which the intact structure is under ambient excitations and structural damage is caused by unknown seismic excitations, is eliminated via the transmissibility functions (TFs) of measured structural strain responses. Contrary to the previous TFs of structural acceleration or displacement responses, TFs of structural strains is more sensitive to local damage in beam-like structures. Moreover, power spectral density transmissibility functions (PSDTFs) are used instead of the traditional Fourier transform TFs. Then, through the inverse Fourier transformation in the frequency domain, the PSDTFs are returned to the time-domain signals to avoid the problem of selecting the proper frequency band in previous studies on identifying structural damage based on TFs in the frequency domain. Finally, structural damage can be identified based on the energy variations of the time-domain strain signals from the intact and damaged structures. Numerical examples of structural damage identification of a seven-storey planar frame and a cable-stayed bridge girder under unknown seismic excitations are provided to demonstrate the effectiveness and robustness of the improved method. Moreover, structural damage identification of a laboratory cable-stayed bridge tower model subjected to shaking table tests is performed to validate the good performance of the proposed method.

Vibration analysis of thick-section sandwich structures in thermal environments

An ultra-light sandwich structure with a thin ceramic-based composite face sheet bonded to a thick thermal insulator is developed as an integrated thermal protection system (ITPS) structure to protect the substructure of a hypersonic flight vehicle. For vibration fatigue assessment and failure prediction, the global dynamic response of the three-dimensional ITPS structure when subjected to a combination of thermal and harmonic excitation is desired. In this study, a novel unified model for vibration analysis of a thick-section sandwich structure is proposed based on the variational asymptotic method. The original 3D sandwich structure problem is reduced to one-dimensional through-the-thickness analysis and two-dimensional (2D) reference plane analysis. The stiffness matrices, including an equivalent transverse shear matrix, are obtained; hence, a shear correction factor is not required. The mechanical response of the 2D reduced-order plate under general boundary conditions in thermal environments is investigated. The static thermal deformation and forced vibration of the sandwich structure with the temperature gradient in the thickness direction are determined using a set of algebraic orthogonal polynomials through the Gram–Schmidt procedure. The thermal deflection is first introduced into the kinematic relationship to modify the total strain energy for the sandwich plate with temperature-independent material properties. To demonstrate the validity and applicability of the model, some numerical results relating to displacements and fundamental frequencies of simply supported sandwich plates are presented and compared to those obtained by other models. The effects of temperature gradients in the thickness direction, boundary conditions, core thickness, and temperature-dependent material properties on the dynamic characteristics and forced vibration of the sandwich structure are discussed by a detailed parametric study. The global dynamic strain response is first obtained using the plate model, which is desirable to determine the 3D local stress and strain field and predict vibration fatigue failure for the ITPS structure.

Application of Edge Computing in Structural Health Monitoring of Simply Supported PCI Girder Bridges.

This study proposes an innovative method for structural health monitoring of simply supported PCI girder bridges based on dynamic strain and edge computing. Field static and dynamic load tests were conducted on a bridge consisting of a span with newly replaced PCI girders and numerous spans with old PCI girders. Both the static and dynamic test results showed that the flexural rigidity of the old PCI girders decreased significantly due to deterioration. To improve the efficiency of on-site monitoring data transmission and data analysis, this study developed a smart dynamic strain gauge node with the function of edge computing. Continuous data with a sampling frequency of 100 Hz were computed at the sensor node. Among the computed results, only the maximum dynamic strain data caused by the passage of the heaviest vehicle within 1 min were transmitted. The on-site monitoring results indicated that under routine traffic conditions, the dynamic strain response of the new PCI girder was smaller than that of the deteriorated PCI girder. When the monitored dynamic strain response has a tendency to magnify, attention should be paid to the potential prestress loss or other deterioration behaviors of the bridge.

Additive-Manufactured Platinum Thin-Film Strain Gauges for Structural Microstrain Testing at Elevated Temperatures

This paper investigates the feasibility and performance of the fabrication of platinum high-temperature thin-film strain sensors on nickel-based alloy substrates by additive manufacturing. The insulating layer was made of a dielectric paste by screen printing process. A 1.8-micron-thick platinum film was deposited directly on the insulating layer. The four-wire resistance measurement method was used to eliminate the contact resistance of the solder joints. Comprehensive morphological and electrical characterization of the platinum thin-film strain gauge was carried out, and good static and dynamic strain responses were obtained, which confirmed that the strain gauge was suitable for in situ strain monitoring of high-temperature complex components.

Damage detection of flange bolts in wind turbine towers using dynamic strain responses

Damage detection of flange bolts in wind turbine towers using dynamic strain responses

Dynamic Strain Response of Hot-Recycled Asphalt Pavement under Dual-Axle Accelerated Loading Conditions

Accelerated pavement testing (APT) is an effective method to study the long-term performance of pavement. Therefore, the dynamic strain behavior analysis of asphalt pavement has important guiding significance in the study of pavement failure modes. To explore the dynamic response of a high-content plant-mixed hot-reclaimed asphalt mixture under a dynamic load of vehicles, a full-scale test road was paved, and ALT biaxial accelerated loading test equipment was used to simulate the dynamic loads of vehicles. Based on parameters such as axle load, temperature, speed, and loading times, the development law for the bottom strain of the three pavement structures was analyzed. The test results show that the most unfavorable position of the asphalt pavement load is located just below the centerline of the wheel track on one side, and the damage effect of a single double-axle wheel load is far greater than that of two single-axle wheel loads. Then, the longitudinal tensile strain of the pavement bottom always maintains the alternating state of compression-tension and compression. The longitudinal tensile strain of the pavement bottom is larger than the transverse tensile strain, and transverse fatigue cracks appear first. Under normal temperature conditions, the bottom tensile strains of the three composite pavement structures under different axial loads are close, and the pavement performance of the hot-recycled asphalt pavement of structure A and structure B can meet the specification requirements. The relationship between the bottom strain and axle load is nonlinear and is directly related to the tire ground pressure, and the difference in the tensile and compressive strain values of the bottom of the three composite pavement structures is small. Under high temperature conditions, the bottom layer temperature of structure A and structure B is lower than that of structure C, and the thermal heat transfer efficiency of hot-recycled asphalt pavement is lower than that of ordinary asphalt pavement. Additionally, the longitudinal tensile strain is about 1–1.5 times that of the transverse tensile strain. Based on the Boltzmann function, the accumulative tensile strain prediction model was established to reflect the relationship between the cumulative strain at the bottom and the number of loads.

TiB₂/SiCN Thin-Film Strain Gauges Fabricated by Direct Writing for High-Temperature Application

The in situ strain/stress monitoring of hot components in harsh environments remains a challenging task. In this study, TiB <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> /SiCN thin-film strain gauges were fabricated on nickel base alloy substrates via direct writing. The static and dynamic strain responses were investigated from 25 °C up to 800 °C. The results show that this thin-film strain gauge fabricated by ceramic-based materials exhibits excellent thermal stability and strain response. Without any anti-oxidative protective layer deposited, its operating temperature is as high as <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$700^{\circ }\text{C}$ </tex-math></inline-formula> , which is 200 °C higher than that of the high-temperature piezoresistive thin-film strain gauges that have been developed and comparable to thin-film strain gauges with protective layers. The gauge factor of theTiB <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> /SiCN high-temperature thin-film strain gauge is 7.12, which is higher than that of most high-temperature thin-film strain gauges. In addition, the strain gauge exhibits excellent resistance stability with a mechanical hysteresis of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$3~\mu \varepsilon $ </tex-math></inline-formula> and a resistance drift of 0.0008/h at room temperature. Therefore, TiB <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> /SiCN thin-film strain gauges provide an effective approach for the measurement of in-situ static and dynamic strain of hot components in harsh environments.

Seismic behaviour of granular slope under railway embankment in large-scale shaking table test

This study investigates shear zone development in a coarse-grained soil slope beneath a railway embankment under seismic action. Clarifying the dynamic stress and residual strain responses under different seismic excitations reveals the slope instability mechanism. A typical coarse-grained soil slope under construction of the Sichuan-Tibet Railway Railway was selected for large-scale shaking table model testing. The evolution of slope response parameters (dynamic displacement, acceleration, dynamic strain and residual strain, and dynamic earth pressure) was monitored. The seismic damage modes and resulting deformations of the embankment and slope were analysed. The evolution characteristics of a shear zone were described; the relationship among acceleration, dynamic stress, and dynamic strain was clarified; and the critical state of earthquake-induced slope instability was defined. The main conclusions are as follows: under seismic excitation, the slope slip surface forms mostly near the interface between the sand and breccia layers, and a weakness zone exists between the subgrade and slope surface. Under high-energy seismic excitation (i.e., PGA ≥ 0.4 g), positive peak acceleration is more suitable for assessing the elevation amplification effect. Considering the cumulative strain distribution contour maps, the growth characteristics in the dynamic stress peak area can describe the development of a shear slip zone. Acceleration amplification factors can reflect slope responsiveness, but the energy contained in the seismic waves determines deformation. When the ratio of dynamic to shear stress exceeds 2, the dynamic strain in the slope slip zone increases considerably with shear zone extension and slope instability. With increasing dynamic stress, the growth characteristics of residual strain exhibit stability, transition and failure states.

Virtual sensing in an onshore wind turbine tower using a Gaussian process latent force model

AbstractWind turbine towers are subjected to highly varying internal loads, characterized by large uncertainty. The uncertainty stems from many factors, including what the actual wind fields experienced over time will be, modeling uncertainties given the various operational states of the turbine with and without controller interaction, the influence of aerodynamic damping, and so forth. To monitor the true experienced loading and assess the fatigue, strain sensors can be installed at fatigue-critical locations on the turbine structure. A more cost-effective and practical solution is to predict the strain response of the structure based only on a number of acceleration measurements. In this contribution, an approach is followed where the dynamic strains in an existing onshore wind turbine tower are predicted using a Gaussian process latent force model. By employing this model, both the applied dynamic loading and strain response are estimated based on the acceleration data. The predicted dynamic strains are validated using strain gauges installed near the bottom of the tower. Fatigue is subsequently assessed by comparing the damage equivalent loads calculated with the predicted as opposed to the measured strains. The results confirm the usefulness of the method for continuous tracking of fatigue life consumption in onshore wind turbine towers.

Investigation on the dynamic impact factor of a concrete filled steel tube butterfly arch bridge

This study introduces the dynamic impact factor (IF) assessment of a concrete filled steel tube (CFST) butterfly arch bridge with both experimental and numerical results. The modal parameters are identified from the measured accelerations and the IFs are computed from the dynamic strain responses. Four well-known bridge design codes are applied to assess the experimental IFs. A vehicle-bridge coupled vibration system is established to numerically investigate certain effects of vehicle speed, road roughness and structural damping on the IF. Both experimental and numerical results indicate that the IFs of distinct structural components vary significantly, suggesting that a uniform formula is insufficient to describe the impact effect of the whole bridge, which has been generally adopted in many bridge design codes. For the irregular shaped CFST arch bridge, the experimental method is currently the most accurate way to determine dynamic IF. New methodologies must be developed to support the engineers to design safer complex bridges and new formulations must be developed to be part of the bridge codes.

Characteristics and analysis of dynamic strain response on typical asphalt pavement using Fiber Bragg Grating sensing technology

The purpose of this research was to reveal the dynamic strain response characteristics and variation law of asphalt pavement structure with different base types under traffic load, and provide reliable measured data for the development of a pavement structure design method combined with China's national conditions and highway construction characteristics. Four types of base pavement were designed, cement treatment base, cement treatment base + graded gravel base, asphalt treatment base + cement treatment base, and asphalt treatment base + graded gravel base asphalt pavement. The dynamic responses of four kinds of asphalt pavements at different transverse positions, depths, and speeds under 100 kN loading conditions were tested using fiber Bragg grating sensors. The results show that the longitudinal strain at the bottom of the surface courses under dynamic load reveals the compressive-tensile-compressive alternating law. The longitudinal strain at the bottom of the base courses is mainly tensile strain. The maximum peak value of the surface courses appears at the corresponding single wheel load center. The maximum peak value of the base courses appears at the corresponding dual-wheel gap center. The faster the traveling speed is, the shorter the loading time of each structural layer of asphalt pavement, and the higher the loading frequency is. The loading position has a significant influence on the peak value of dynamic strain response. Reducing the cement-treated base course's strength and stiffness appropriately and ensuring the continuity of each layer's modulus to the greatest extent will benefit each structural layer's dynamic strain response distribution and improve the bearing capacity and durability of each layer of asphalt pavement against a heavy load.

Two-Step Approach to Processing Raw Strain Monitoring Data for Damage Detection of Structures under Operational Conditions.

Strain data of structural health monitoring is a prospective to be made full use of, because it reflects the stress peak and fatigue, especially sensitive to local stress redistribution, which is the probably damage in the vicinity of the sensor. For decoupling structural damage and masking effects caused by operational conditions to eliminate the adverse impacts on strain-based damage detection, small time-scale structural events, i.e., the short-term dynamic strain responses, are analyzed in this paper by employing unsupervised modeling. A two-step approach to successively processing the raw strain monitoring data in the sliding time window is presented, consisting of the wavelet-based initial feature extraction step and the decoupling step to draw damage indicators. The principal component analysis and a low-rank property-based subspace projection method are adopted as two alternative decoupling methodologies. The approach’s feasibility and robustness are substantiated by analyzing the strain monitoring data from a customized truss experiment to successfully remove the masking effects of operating loads and identify local damages even concerning accommodating situations of missing data and limited measuring points. This work also sheds light on the merit of a low-rank property to separate structural damages from masking effects by comparing the performances of the two optional decoupling methods of the distinct rationales.

Data‐driven virtual sensing and dynamic strain estimation for fatigue analysis of offshore wind turbine using principal component analysis

AbstractVirtual sensing enables estimation of stress in unmeasured locations of a system using a system model, physical sensors and a process model. The system model holds the relationship between the physical sensors and the desired stress response. A process model processes both the physical sensors and the system model to synthesise virtual sensors that ‘measure’ the desired stress response. Thus, virtual sensing enables mapping between the physical sensors (input) and the desired stress response (output). The system model is a mathematical model of the system based on knowledge or data of the system. Here, the data‐driven system model is constructed directly on data analyses for the specific system. In this paper, supervised learning and data‐driven system models are applied to strain estimation of an offshore wind turbine in the dynamic range through a novel use of principal component analysis (PCA); 40 min of training data is used to establish the data‐driven system model that can estimate the dynamic strain response with high precision for 2 months, while the estimated fatigue damage averaged out to −1.76% of the measured strain response.

A theoretical approach towards the modeling of vibrothermography using finite element methods

Vibrothermography offers the capability to detect surface cracks mainly through capturing vibration-induced frictional heat generation of contacting crack surfaces in metallic alloys. The multidisciplinary nature of this inspection technology makes it rather difficult to theoretically capture the system response without the proper application of finite element methods (FEM). It requires relating the material mechanical and thermal properties in a coupled finite element model to properly address wave propagation, contact mechanics, fracture mechanics, friction heat generation and heat diffusion in simulated vibrothermography inspections. In this study, a theoretical model is developed to estimate the dynamic strain response at any location in commonly used cantilever beams oscillating at a given frequency. This model is utilized to assess the convergence of FEM dynamic system response while determining the element size requirements to realistically model elastic wave propagation throughout the sample. Three different meshing criteria are assessed to properly capture the singularity at the crack tip. It is demonstrated that the use of solid element circular meshing criteria around the crack tip converges to singular quarter-node solid element solution commonly used in fracture mechanics and allows coupling it with heat generation and diffusion around the vicinity of a crack to deliver a practical approach for modelling vibrothermography. This effort brings FEM a step closer towards realistic simulation of vibrothermography in the future.

Multi-attribute seismic data spectrum analysis of tunnel orthogonal underpass landslide shaking table test

To explore the spatial deformation characteristics and dynamic response law of the tunnel's orthogonal underpass landslide under potential seismic action, we carried out the shaking table test for the first time. We measured and obtained the seismic data with various attributes, such as the characteristic image of different probability horizontal seismic action, acceleration, dynamic soil pressure and dynamic strain. Based on the time-domain analysis of multi-attribute data such as deformation characteristics and acceleration, we revealed the spatial dynamic response characteristics of the lining structure at different positions in the annular particle. By Wavelet Transform, we obtained the time-frequency variation relationship between the multi-attribute data. In addition, we discussed in detail the relationship between the frequency and the specific period of acceleration, dynamic soil pressure and dynamic strain at different probabilities of horizontal vibration stages. We then proposed the damage level correlation quantitatively characterized by the spectrum characteristics of multi-attribute data through Fast Fourier Transform and statistical probability scatter-matrix operations. The sensitivity difference of multi-attribute seismic data was significant. The time-history effect of dynamic soil pressure was the most sensitive and the time-history effect of dynamic strain had obvious hysteresis compared with acceleration. In the frequency domain of multi-attribute seismic data signals, the predominant frequencies were mainly concentrated in 1–10 Hz, and their dominant frequencies were significantly different. The singular boundary between low-frequency and high-frequency was 10 Hz, and the structures with frequencies between 15 and 20 Hz were relatively safe. Multi-attribute data response variables were positively correlated, acceleration was highly correlated with dynamic soil pressure response, while acceleration was weakly correlated with dynamic strain response, showing a significant correlation. Due to the spatial location of the tunnel's orthogonal underpass landslide, and the effect of seismic wave amplification and bias effect of mountain tunnel, the lining was damaged in a regional difference, and the inverted arch uplift cracked into a weak area. The research results can provide a theoretical reference for the failure model prediction and engineering safety evaluation of the tunnel's orthogonal underpass landslide in high intensity earthquake area.

RETRACTED ARTICLE: Shaking table test on rock-socketed piles in saturated sand foundation considering the effect of karst caves

Karst landform is a major unfavorable geological condition that significantly impacts the pile foundation in bridge engineering. The karst caves may cause foundation failure and induce severe damage to pile foundation under seismic action. Currently, there is a limited number of studies on the seismic dynamic response rules and mechanical properties of rock-socketed piles in karst areas. In this paper, the shaking table test on the seismic response of penetrated rock-socketed piles in saturated sand foundation considering the effect of karst caves is carried out, and a large-sized stratified suspended shear box is adopted to simulate the soil boundary effect. The effects of ground motion intensity, karst cave height, and foundation state on the seismic dynamic strain response of rock-socketed piles are explored. Test results show that the peak dynamic strain of the penetrated rock-socketed pile in the saturated sand layer initially increases and then decreases from bottom to top of the pile, and its maximum peak strain response occurs in the middle of the pile. The increase of karst wave height causes the penetrated rock-socketed pile in the saturated sand layer to become susceptible to premature failure in the middle. With the increase of earthquake intensity, the effect of karst caves on the response becomes weaker. However, in dry sand foundation, the peak strain of the pile is small in the middle and large value occurs at pile ends, which indicates that the seismic failure modes of pile foundation in dry and saturated sand foundation are significantly different.

Research on seismic response of new lining structured of shallow double-arch tunnels under unsymmetrical pressure

A physical test model of a new lining structure for a shallow double-arch tunnels under unsymmetrical pressure with a scale of 1:20 was designed and manufactured. Kobe seismic waves and EI seismic waves were selected as the loading waves and a large-scale shaking table test was carried out. The acceleration and dynamic strain response of shallow double-arch tunnels under unsymmetrical pressure under different seismic wave types and seismic intensities are studied. The results show: Under different seismic wave excitations, only the horizontal acceleration amplification factor of the left-hole vault, right-hole invert and the top right of the mid-partition is less than 1, and the horizontal acceleration amplification factors of other measurement points are all greater than 1. The measurement points with relatively large horizontal acceleration response are the left-hole shoulder, the top left of the mid-partition, the right-hole vault and shoulder; The overall response of the right half-arch of the left-hole is greater than the left half-arch of the left-hole, and the overall response of the left half-arch of the right-hole is greater than the right half-arch of the right-hole. The measured points in the left half-arch of the left-hole and the right half-arch of the right-hole have small differences in acceleration response; The effects of Kobe wave on horizontal acceleration and vertical acceleration are greater than EI wave, and the average value of the vertical acceleration response of the lining is greater than the average value of the horizontal acceleration response. With the increase of seismic intensity, the larger the acceleration amplification factor is, the greater the increase is; Under the action of different seismic waves, the seismic wave excitation has a greater impact on the dynamic strain response of the left-hole, and less impact on the right-hole. Among them, the strain value of the left-hole shoulder, left-hole invert and the top left of the mid-partition is much larger than the other measurement points, the trend of the right-hole is relatively gentle, and the strain values of the shoulder are slightly larger. The research conclusions have certain guidance and reference value for the seismic design of shallow double-arch tunnels under unsymmetrical pressure.

BwimNet: A Novel Method for Identifying Moving Vehicles Utilizing a Modified Encoder-Decoder Architecture.

Traffic loading monitoring plays an important role in bridge structural health monitoring, which is helpful in overloading detection, transportation management, and safety evaluation of transportation infrastructures. Bridge weigh-in-motion (BWIM) is a method that treats traffic loading monitoring as an inverse problem, which identifies the traffic loads of the target bridge by analyzing its dynamic strain responses. To achieve accurate prediction of vehicle loads, the configuration of axles and vehicle velocity must be obtained in advance, which is conventionally acquired via additional axle-detecting sensors. However, problems arise from additional sensors such as fragile stability or expensive maintenance costs, which might plague the implementation of BWIM systems in practice. Although data-driven methods such as neural networks can estimate traffic loadings using only strain sensors, the weight data of vehicles crossing the bridge is difficult to obtain. In order to overcome these limitations, a modified encoder-decoder architecture grafted with signal-reconstruction layer is proposed in this paper to identify the properties of moving vehicles (i.e., velocity, wheelbase, and axle weight) using merely the bridge dynamic response. Encoder-decoder is an unsupervised method extracting higher features from original data. The numerical bridge model based on vehicle-bridge coupling vibration theory is established to illustrate the applicability of this new encoder-decoder method. The identification results demonstrate that the proposed approach can predict traffic loadings without using additional sensors and without requiring vehicle weight labels. Parametric studies also show that this new approach achieves better stability and reliability in identifying the properties of moving vehicles, even under the circumstances of large data pollution.