Response Of Bridge Research Articles

In Situ Testing and Finite Element Analysis of a Discontinuous Mortise and Tenon Stone Bridge Under Natural Excitation

To study the dynamic response of multi-span mortise and tenon stone bridges under natural excitation, a bluestone multi-span stone bridge with a main span of 2.56 m in southern China was taken as the research object. Based on the collected pulsating signals of bridge piers and slabs, the natural frequencies and damping ratios of the main span bridge slab and pier were analyzed using the half-power broadband method (HPBM) and random decrement technique (RDT). Modal analysis was conducted using ANSYS, and the results were compared with those obtained from on-site experiments for further performance analysis. The research results of this article indicate that the natural frequency range of the 2.56-m bridge slab identified by measured signals is 48–49 Hz, and the damping ratio range is 33.33–36.61%. The natural frequency of the central pier is 75–76 Hz, and the damping ratio range is 26.39–27.83%. Through finite element modal analysis, the natural frequency of the bridge slab is 54.401 Hz, with an error of 10.5%. The natural frequency of the overall stone bridge is about 82.2 Hz, with an error of about 8.2%. The validated finite element model was subjected to normal water flow impact and erosion simulation. The results indicate that under erosion with fewer particles and lower flow rates, the upstream pier bottom at the center receives the highest relative erosion mass and displacement per unit area. The bridge deck near the main span also experienced relative displacement. Therefore, in the subsequent protection work, special attention should be paid to these components.

Dynamic response analysis of a floating bridge structure under combined actions from seismic loading and waves at different incident angles

Abstract This paper investigates the complex coupled response of a cross-sea floating bridge under the combined actions of seismic loading and wave at different incident angles. The bridge structure is modeled using the finite element method, considering the interaction between the girder, pier, pontoon, and mooring chains. Wave forces are calculated based on potential theory, and an added mass approach is applied to simulate the hydrodynamic forces induced by the earthquake. Simulations were conducted using a one-year random wave at three different incident angles of 0o, 45o, and 90o, combined with three-dimensional seismic loading of varying magnitudes: 0.3 g, 0.5 g, and 1 g. The simulation results indicate that the direction of the incident wave significantly influences the bridge's response. The coupling effect between wave and seismic forces on the floating bridge's response is not simply additive. Generally, when the earthquake magnitude is low, the wave loading significantly impacts the bridge's dynamics. However, as the earthquake magnitude increases, the bridge's response becomes dominated by the seismic forces, rendering the influence of wave forces negligible.

Computer-Vision-Aided Deflection Influences Line Identification of Concrete Bridge Enhanced by Edge Detection and Time-Domain Forward Inference

To enhance the accuracy and efficiency of the deflection response measurement of concrete bridges with a non-contact scheme and address the ill-conditioned nature of the inverse problem in influence line (IL) identification, this study introduces a computer-vision-aided deflection IL identification method that integrates edge detection and time-domain forward inference (TDFI). The methodology proposed in this research leverages computer vision technology with edge detection to surpass traditional contact-based measurement methods, greatly enhancing the operational efficiency and applicability of IL identification and, in particular, addressing the challenge of accurately measuring small deflections in concrete bridges. To mitigate the limitations of the Lucas–Kanade (LK) optical flow method, such as unclear feature points within the camera’s field of view and occasional point loss in certain video frames, an edge detection technique is employed to identify maximum values in the first-order derivatives of the image, creating virtual tracking points at the bridge edges through image processing. By precisely defining the bridge boundaries, only the essential structural attributes are preserved to enhance the reliability of minimal deflection deformations under vehicular loads. To tackle the ill-posed nature of the inverse problem, a TDFI model is introduced to identify IL, recursively capturing the static bridge response generated by the bridge under the influence of successive axles of a multi-axle vehicle. The IL is then computed by dividing the response by the weight of the preceding axle. Furthermore, an axle weight ratio reduction coefficient is proposed to mitigate noise amplification issues, ensuring that the weight of the preceding axle surpasses that of any other axle. To validate the accuracy and robustness of the proposed method, it is applied to numerical examples of a simply supported concrete beam, indoor experiments on a similar beam, and field tests on a three-span continuous concrete beam bridge.

Combining long short-term memory and shape superposition for estimating moving load-induced global responses of bridges

Recently, a method of estimating the global responses of bridges was introduced, by combining a deep neural network (DNN) and shape superposition method (SSM). This DNN-SSM model demonstrated superior performance to the traditional estimation method based on the least-squares method. However, the application and validation of this approach have been restricted to static problems. For bridges, the dynamic effect induced by the moving load requires consideration. Therefore, this study proposes a long short-term memory (LSTM)-SSM model to improve the estimation performance for dynamic global responses. The LSTM-SSM model consists of LSTM layers, followed by fully connected DNN layers and a post-process part based on SSM. Limited displacement, slope, and strain data were used as the inputs, and the global displacement, slope, and strain were estimated. To validate the effectiveness and improvement of the LSTM-SSM model, a numerical model of a cable-stayed bridge was used to compare the proposed model with the previously developed DNN-SSM model. The validation results indicated that the LSTM-SSM model can provide more accurate estimates of the dynamic global response than DNN-SSM models. Finally, a technique for structural safety monitoring based on the proposed method was introduced.

A refined output-only modal identification technique for structural health monitoring of civil infrastructures

The increasing installation of structural health monitoring systems has raised the need for efficient and rapid data interpretation algorithms. Operational modal analysis is currently one of the most used techniques for extracting modal properties, and some attempts have been made to automate this procedure. However, automated techniques often require manual calibration of hyperparameters and show inconsistencies across different case studies. This paper proposes a refined version of the automated frequency domain decomposition (AFDD) using the modal assurance criterion (MAC) to obtain natural frequencies and mode shapes. Sensitivity analyses were conducted on the ambient vibration response of the Yonghe cable-stayed bridge in China, accounting for influential factors including noise levels, acceleration record length, sensor layouts. A novel approach is then introduced to interpret the results of the sensitivity analyses. The approach consists in plotting the result of each analysis in a stabilization diagram and then using a Gaussian mixture model that clusters the poles into core and outliers. This allows to identify regions where the MAC thresholds are optimal. By comparing the results of all the sensitivity analyses it was possible to define a single optimal MAC threshold, avoiding the need for fixing a value based on the user’s experience. Three substantially different case studies were analyzed to extensively test the methodology: the Yonghe cable-stayed bridge, the PolyU footbridge in Hong Kong, and Moletta Tower in the Circus Maximus archeological site in Italy. The analysis compared the proposed AFDD algorithm to the traditional frequency domain decomposition and covariance-driven stochastic subspace identification. Specifically, the efficiency in identifying close frequencies, weakly excited modes, spurious peaks, and complex modes was evaluated for each method, which highlighted the robustness of the proposed optimized AFDD. The analysis showcases peculiar characteristics and drawbacks of each method when trying to identify complex vibrational modes of the specific case studies. It was found that the proposed AFDD procedure performs better than traditional methods despite it may misidentify complex modes due to the constraints of a narrower modal domain and similar geometries.

Seismic response of rocking bridge systems under three-dimensional ground motions

Compared to theoretical models, finite element methods provide greater versatility for dynamic response analyses of rocking bridges, especially when considering the influences of various nonlinear boundary conditions. A finite element modeling method based on fiber section elements for three-dimensional seismic analyses of rocking bridges is developed in this paper. This method solves the problem of energy dissipation and stiffness determination of the rocking interface. Based on the proposed modeling method, considering nonlinear boundary conditions (such as abutment-soil interaction, impact effect, and pile-soil interaction), the seismic responses of rocking bridges, including free rocking, prestressed rocking and prestressed rocking with dampers, under the excitation of three-dimensional ground motions are analyzed. The results show that: after reasonable design, the prestressed rocking bridge with dampers exhibits similar seismic responses to the cast-in-place bridge, and maintains a good self-centering ability. The responses of all the rocking bridges under bi-directional loading are greater than those under unidirectional loading. In contrast, vertical ground motion's influence on the response of the rocking bridges is minimal. Ignoring the pile-soil interaction significantly underestimates the seismic demand of the evaluated bridges, especially the free rocking bridge and rocking bridge with prestressed tendons only.

Flexural behavior of reinforced concrete arch ribs

For many years, arch bridges have been celebrated for their striking appearance and structural effectiveness. However, the response of arch bridges to dynamic loadings, particularly during earthquakes, remains poorly understood, and reliable seismic design guidelines are lacking. Previous research has suggested that reinforced concrete (RC) arch ribs may experience flexural yielding at their springings due to the interaction of axial force and bending moment under earthquake loading. To assess the ductile behavior and ensure the adequate performance of RC arch ribs, a comprehensive investigation combining experimental and analytical approaches was conducted in this study, taking into account the influence of high axial load and the hollow section configuration present in RC arch ribs. The test results revealed that all RC arch ribs exhibited a typical flexural failure mode. As the axial load increased, the lateral load bearing capacity also increased, but the displacement ductility decreased significantly. Comparing different section configurations, the RC arch rib specimen with a double-cell boxed section demonstrated higher lateral load bearing capacity but lower displacement ductility compared to the single-cell boxed section, while the energy dissipation capacity was approximately similar for both section configurations.

Performance Evaluation of Inerter and Negative Stiffness-Based Dampers for Seismic-Induced Vibration Response Control of a Cable-Stayed Bridge: A Comparative Study

The excessive main girder displacement responses of the large-span cable-stayed bridges during seismic events may precipitate collisions between the main girder and the approach bridge. The viscous dampers (VDs) have been extensively employed in the seismic response control of long-span bridges, yet their effectiveness is limited. To augment the seismic-induced vibration control performance of the VD, inerter element, negative stiffness (NS) element, and spring element have been incorporated into the VD to achieve energy dissipation capacity enhancement. The equations of motion for the simplified cable-stayed bridge-damper systems are established under stochastic seismic excitation, and the design parameters of inerter and NS-based damper are optimized by using the genetic algorithm. The seismic control performance of the VD and five types of dampers are systematically compared, demonstrating that these dampers can substantially enhance the main girder displacement mitigation performance in cable-stayed bridges. Owing to the NS characteristics attributes of the inerter element and NS element and the tuning effect of the spring element, the tuned viscous mass damper (TVMD) with NS achieves a 24.56% higher efficiency in control performance compared to the VD.

Study on the Application of Determinant-MTMD(TMDD) Vibration Reduction in Cable-Supported Pedestrian Suspension Bridge

In this study, multiple tuned mass dampers (MTMDs) were studied to understand their impact on the human-induced vibration response and comfort level of a pedestrian cable-supported suspension bridge. A spatial finite element model based on a specific engineering case was established. The dynamic characteristics of the bridge under human-induced loads were investigated, and its comfort level under human-induced vibrations was analyzed using the time-history method. Then, this study adjusted the design parameters of the dampers based on various optimal damper parameter expressions. Furthermore, the damping effectiveness of MTMD under different mass ratios (μ) was evaluated, and it was found that increasing the mass ratio significantly impacts damping performance. Finally, determinant-TMD (TMDD) was introduced, and a comparison between the damping effect, robustness, and performance of TMDD and MTMD was conducted. The results indicate that while increasing the mass ratio does not linearly affect maximum vibration acceleration, the damping effect increases initially and then stabilizes, with a damping rate converging at approximately 55%. However, with the TMDD approach, the maximum damping rate can reach approximately 70%, enhancing comfort levels from the “minimum CL3” achieved with MTMD to the “medium CL2” level. Additionally, while TMDD’s robustness is slightly inferior to MTMD at lower mass ratios, it demonstrates superior robustness at higher mass ratios.

Nonlinear seismic response analysis of long-span railway cable-stayed bridges crossing strike-slip faults

Active faults along railways in the mountainous regions of western China pose significant challenges to bridge safety. To ensure the safe operation of long-span railway bridges under complex geological conditions, this study investigates the synthesis of artificial ground motions for bridges crossing strike-slip faults and analyzes their nonlinear seismic response. First, we develop a theoretical method for simulating high- and low-frequency seismic motions using a finite fault and an equivalent velocity pulse model. Next, using a specific long-span railway cable-stayed bridge as a case study, we construct a nonlinear finite element model with OpenSees software. Finally, we assess the seismic response of key bridge components considering various crossing angles, seismic amplitudes, fault rupture directivity, and fling-step effects. The results show that the crossing angle significantly influences the seismic response, with longitudinal and transverse responses exhibiting opposite patterns. Additionally, the scaling factor of seismic motion significantly affects bridge response. For bridges crossing strike-slip faults, the longitudinal response exhibits a sudden increase in displacement due to instantaneous velocity pulses, while the transverse response shows notable residual displacement influenced by the fling-step effect. However, the critical section curvatures of bridge towers and piers remain within the elastic range across all crossing angles, indicating that controlling large displacement deformations is crucial for the seismic design of bridges crossing strike-slip faults.

Compressed sensing with smooth L0 constraints for moving force identification from bridge response measurements

Abstract Compressed sensing (CS), as an emerging information sampling technique, has been successfully applied in the field of moving force identification (MFI). However, existing MFI CS models often fail to obtain the optimal sparse solutions and frequently underestimate the amplitude of local impact forces. To effectively address this issue, a new CS method is proposed for MFI based on smooth L0 norm constraints and bridge response measurements. Firstly, a smooth function is used to approximate the L0 norm, establishing a noise CS reconstruction model for MFI. The introduction of the smoothing function can locally convexify the original MFI problem and enhance the smoothness and differentiability of the objective function, making the optimization problem easier to solve. Subsequently, the Polak–Ribiere–Polyak formula is adopted to point the descent direction of the new objective function, and the sparse solution is iteratively advanced through the conjugate gradient algorithm. Finally, the applicability and feasibility of the proposed method is confirmed by numerical simulations and vehicle–bridge interaction tests, respectively. The results show that the proposed method can accurately identify moving forces from limited measurements of bridge responses. Compared with existing methods, it can provide more precise sparse solutions with higher robustness to measurement noises, and address the issue of underestimating on the amplitude of local impact forces, which is expected to enhance the performance and in-situ applicability of MFI.

Bridge Monitoring Using Smartphones Installed on Driving Vehicles: Oriented to Crowdsourcing Model

The acquisition of bridge frequency by using the ubiquitous ordinary passenger vehicles and smartphones in cities has attracted an increasing amount of attention. This study proposes a bridge monitoring method and develops a vehicle–bridge crowdsourcing monitoring (VBCM) platform for public participation. It takes smartphones carried by vehicles as monitoring terminals for a long-term bridge monitoring task. To validate the feasibility of available smartphone brands as a signal collector, a small shake table test is carried out and the processing scheme for the smartphone sampling data is investigated. A smartphone sensor frequency gain technique is proposed to satisfy the fusion of multisensor data in different smartphones. Furthermore, a magnetic target identification technique is proposed to pick up and extract the bridge response signal segment corresponding to the vehicle’s entering and leaving, as well as to eliminate redundant data. To this end, the network of multiple smartphones is synchronized. For the sake of applicability and feasibility, a series of scaled experiments are conducted in the laboratory considering an adapted toy car driving over a simply supported steel bridge. The results demonstrated the practicality of the proposed methodology. The combination of easily accessible smartphones oriented to a crowdsourcing model and a driving vehicle lowers the threshold for the bridge monitoring process, which also pinpoints a potential for future structural health monitoring development.

Influence of velocity pulse directivity on seismic response of cross-fault bridges

Pulse-type ground motions have special destructive effects on structures, and directivity is one of the fundamental characteristics of velocity pulses. In this paper, the impact of the velocity pulse directivity of seismic motions on both sides of the fault on the seismic response of fault-crossing bridges was quantified using the original records from the Chi-Chi earthquake in Taiwan. First, an automatic baseline correction method was applied to restore the permanent ground displacement, and the direction corresponding to the strongest pulse energy on both sides of the fault was identified based on continuous wavelet transformation. On this basis, the variability of the peak intensity of ground motions on the hanging wall and footwall in different directions, as well as the differences in pulse characteristics between the strongest pulse component and the two initial horizontal components, were analysed. Finally, a typical 5 × 30 m continuous girder bridge was considered as the research object to reveal the influence of the velocity pulse directivity on the seismic response of key parts on both sides of the fault. The results showed that the variability in the peak ground-motion intensity on the hanging wall of the fault was higher than that on the footwall. The displacement variability was up to 863 %, velocity was 262 %, and acceleration was 124 %. The pulse characteristics of the strongest pulse component were stronger than those of the parallel and vertical fault components, and its acceleration, velocity, and displacement response spectra were larger. The amplification effects of the velocity pulse directivity on the transverse bending moment, longitudinal bending moment, and torque of the pier bottom reached 1.3, 1.1, and 1.6, respectively. However, the amplification effect of the velocity pulse directivity on the relative displacement of the pier top, displacement of the main beam, and displacement of the bearing was more than 1.6 times, which was more significant than the internal force of the pier bottom.

Correlation analysis of long-span cable-stayed bridge strain response with environmental and operational variations based on feature selection

The structural response of large-span bridges is significantly influenced by environmental and operational conditions. This study employs a data-driven approach, which utilizes various techniques such as multiple linear regression, best subset regression, stepwise regression, Lasso regression, Ridge regression, and ElasticNet regression to analyze in detail the effects of environmental variables (temperature variables including temperature principal components, lateral temperature gradients, and vertical temperature gradients), and traffic loading on bridge strain response. Based on the analysis of 21 months of health monitoring data from a newly constructed cable-stayed bridge, Lasso regression is shown to perform best in predicting strains on the bridge deck and bottom slab. Furthermore, sensitivity analysis identifies several key influencing factors and quantifies their relative contributions to the strain. Although the 10-min average of the strain due to traffic loading has a minimal impact, the root mean square of the strain shows a significant correlation with the cumulative gross vehicle weight during busy traffic periods. The methodology proposed in this study helps to understand the strain response patterns of large-span bridges from a data perspective, and provides an effective approach to establishing a strain baseline model that eliminates environmental and operational variations.

Influence of the vertical seismic component on the response of continuous RC bridges

This paper investigates the influence of vertical seismic accelerations on the seismic response of RC bridges through numerical simulations using an enhanced non-linear numerical model. Results confirm that the incorporation of vertical accelerations, either through actual records or scaled horizontal records, can considerably modify the seismic response and the collapse mechanism. In the case of actual vertical records, the vertical component significantly contributes to premature structural deterioration, intensifying demand and accelerating failure mechanisms. On the other hand, the study underscores the inadequacy of using scaled horizontal records to represent vertical accelerations, as suggested by some seismic codes, as it not only distorts seismic response evaluation but also alters failure modes. The analysis of vertical vibration reveals higher displacements, increasing flexural demand on the deck, and leading to a progressive loss of vertical support at the central column. The research establishes the need to accurately account for vertical seismic accelerations in bridge design evaluations, as their impact on structural response and failure mechanisms cannot be underestimated. The work highlights the importance of a highly detailed 3D numerical model in assessing traditional parameters and capturing complex collapse mechanisms arising from material and geometric nonlinearities.

Study of the dynamic response of an offshore continuous beam bridge under nonlinear wave and scour

The study highlights the critical factors and findings regarding bridge damage susceptibility during typhoons and hurricanes, primarily due to extreme waves, scour, and storm surges. While existing research has extensively studied bridges' responses to either extreme waves or scour individually, their combined effects have not been sufficiently explored. Experiments were conducted on a scaled two-span bridge to examine its behavior under simultaneous wave and scour conditions. Results from these experiments indicate that as scour depth increases, there is a corresponding escalation in displacement of the bridge pier, acceleration of the bearing platform, and strain along the pile foundation. To further investigate these dynamics, fluid-structure interaction analysis was employed, revealing significant insights. Notably, the study found that wave height exerts a substantial influence on wave load. For instance, at a wave height of 6 m, the average peak horizontal wave load on bridge piers was 2.48 times higher than at 3 m wave height. Moreover, local scour was identified as a critical factor reducing the bearing capacity of pile foundations, thereby significantly impacting the bridge's dynamic response to nonlinear waves. Under identical wave conditions, varying scour depths (3 m, 6 m, 9 m, and 12 m) resulted in increases in peak lateral displacements at the pier top compared to non-scouring conditions. The study concludes by emphasizing the increasing risk posed to pile foundations with deeper scour depths, particularly under stronger wave conditions. Consequently, there is a crucial need to enhance the resilience of offshore bridges against these dual hazards through advanced design and protective measures.

Seismic Response of a Cable-Stayed Bridge with Concrete-Filled Steel Tube (CFST) Pylons Equipped with the Seesaw System

Seismic Response of a Cable-Stayed Bridge with Concrete-Filled Steel Tube (CFST) Pylons Equipped with the Seesaw System

Life-cycle seismic performance of bridges with thin-walled steel piers under bi-directional excitations

In corrosive environments, steel bridges demonstrate time-dependent seismic performance and varied directional responses to bi-directional excitation at different life-cycle stages. However, current seismic assessment methods for bridges have not adequately addressed the time-dependent effects of bi-directional excitation directionality on seismic responses throughout their entire life cycle. Thus, this study aims to investigate the time-varying directional effects of bi-directional seismic excitations on steel bridges with multiple parameters and to propose a method for predicting the range of critical incident angles while considering the aging effects of bridges. Initially, time-dependent functions that characterize steel corrosion are integrated into multiscale finite element models of the steel bridges to simulate corrosion progression across their life-cycle stages. Subsequently, a quantitative analysis of the bi-directional seismic response of steel bridges is conducted throughout their life cycle, accounting for the directionality of seismic excitation. Finally, a method and process for predicting the range of time-varying bi-directional critical incident angles with 95 % probability are derived by fitting the statistical probability distribution boundary values of the critical incident angle deviation index. The study's findings indicate that the aging effects of steel bridges amplify the displacement response under bi-directional seismic excitation, with the maximum displacement response difference exceeds 40 %. The service time may increase the sensitivity of steel bridges to changes in incident angles, potentially amplifying the displacement response by a factor of 1.69. The study's findings highlight the importance of considering aging effects and bi-directional excitation directionality in bridge seismic assessments.

Blast resistance of CFRP composite strengthened masonry arch bridge under close-range explosion

Carbon fiber reinforced polymers (CFRP) are recognized for their exceptional strength-to-weight ratio. They offer a viable and effective solution for strengthening and retrofitting masonry bridges, helping to extend their service life, improve structural performance, and meet modern safety and load requirements. Wrapping of CFRP around masonry elements can enhance their confinement and ductility. This flexibility plays a crucial role in preventing sudden brittle failure, allowing for controlled deformation, which is essential for blast resistance. Additionally, CFRP materials possess the ability to flex and absorb energy, which proves beneficial in containing and redistributing forces generated during an explosion, consequently reducing the risk of catastrophic failure. This study employed the coupled Eulerian–Lagrangian (CEL) technique available in the finite element software Abaqus/Explicit to simulate the blast loads. Various detonation scenarios were considered, taking into account factors such as location and their impacts on bridge structures. A detailed micro-model was developed using finite element software and accurate geometric data acquired from FARO laser scanning of the case study. The properties of masonry units and backfill were characterized using the Johnson-Holmquist II damage model and Mohr–Coulomb criteria. The Jones-Wilkins-Lee equation of state (EOS) was applied to replicate the behavior of trinitrotoluene (TNT). In accordance with the JH-II model, the researchers formulated a VUMAT code. The study examined the distinct damage mechanisms and overall structural responses of bridges. By evaluating the blast resistance of individual bridge models, the most critical scenarios were pinpointed. Carbon Fiber Reinforced Polymer (CFRP) was then utilized as a method to fortify bridges against blast loads. A comparison was made between the damage propagation before and after the reinforcement.

Nonlinear unsteady aerodynamic forces prediction and aeroelastic analysis of wind-induced bridge response at multiple wind speeds: A deep learning-based reduced-order model

Machine learning-based aerodynamic reduced-order models (ROMs) combine high accuracy with extremely low computational costs, making them highly effective in predicting nonlinear and unsteady bridge aerodynamic forces. Although several machine learning-based nonlinear aerodynamic models have been developed, the majority are built on a single wind speed parameter. However, in nonlinear aerodynamic prediction and aeroelastic analysis of bridges, the variability in incoming wind speed significantly influences the computed results. A ROM relying solely on a single wind speed lacks the ability to accurately forecast the intricate dynamic behaviors arising from changes in wind speed. When the incoming wind speed changes, the model's prediction accuracy significantly decreases. Usually, it is necessary to establish a new database and train a new model, which not only increases time and cost but also greatly reduces the convenience of the ROM. Addressing this challenge, this study proposes a multiple-wind-speed (MWS) nonlinear unsteady bridge aerodynamic model based on the LSTM deep neural network. Taking the Taohuayu Yellow River Bridge in the Henan Province of China as an example, the modeling process of the proposed MWS-ROM is demonstrated, along with non-linear aerodynamic predictions of the deck under various conditions and aerodynamic-elastic analysis of the deck under different wind speeds. The research results show that the trained LSTM network can accurately predict the nonlinear aerodynamic forces of bridges under single and double degrees of freedom vibration conditions. The MWS-ROM performed well in predicting convergent vibrations at low wind speeds and limits cycle oscillations at high wind speeds, aligning closely with results from the CFD full-order model. Compared to CFD, the aerodynamic ROM based on the LSTM network significantly enhances computational efficiency, consequently boosting the convenience and efficiency of bridge flutter analysis. Additionally, the methodology proposed herein can be extended for wind-induced vibration control and response prediction in other types of deck sections.