The safety of cross-active fault tunnels poses a significant challenge in current railway engineering construction. Research on such tunnels primarily relies on numerical simulation. However, the refinement level of the finite element structural models used to analyze cross-fault tunnels is often inadequate, and it is difficult to accurately reflect the actual stress and deformation characteristics of the tunnel structure under fault displacement. The influence of invert-filling, railway track foundation, and steel rails are considered in this paper. Three tunnel structure models with different levels of refinement are established by using the finite element software ABAQUS, and the simplified and commonly used ring structure models are compared. The stress and deformation characteristics of the cross fault tunnel structure and its associated track system under the action of active reverse faults are analyzed in detail. The results indicate that the refinement level of the model significantly affects the stress levels and deformation calculation results of the tunnel structures. The longitudinal stiffness and integrity of the tunnel structure are strengthened when considering bottom filling layers and internal auxiliary structures, and the ability of the structure to resist fault is improved. Meanwhile, in light of the severe damage to tunnel tracks and ancillary facilities during the M6.9 earthquake in Menyuan, Qinghai, in 2022, a dedicated study is undertaken to further examine the mechanical behavior and deformation characteristics of tracks under strike-slip fault displacement. This analysis is based on a refined tunnel model, aiming to gain a deeper understanding of how tracks respond and deform under such geological forces.

Aiming to accurately differentiate between natural and non-natural earthquakes, a neural network model based on one-dimensional convolution and residual structures, named ResNet-1D, was constructed. This model automatically extracts features from three-component seismic records using convolutional layers with convolutional kernels of different lengths, pooling layers composed of max-pooling, and residual structures. The adaptive moment estimation method (Adams) is used to optimize parameters, and a linear discriminant function (Linear) is applied to distinguish between natural and non-natural earthquakes. Using 40000 velocity records of natural and non-natural earthquakes, compiled by the China Earthquake Networks Center from 2008 to 2020, the data was randomly divided into training, validation, and test datasets in a 6∶2∶2 ratio. The test results show that the classification accuracy for natural and non-natural earthquakes is 92.65% and 94.30%, respectively. Compared with traditional machine learning methods, the ResNet-1D model significantly improves the test results in terms of accuracy, precision, recall, and F1 score, effectively enhancing the accuracy of identifying natural and non-natural earthquakes. Moreover, variations in magnitude and epicentral distance also affect the classification accuracy of the model, with higher magnitudes and greater distances resulting in lower accuracy. The model proposed in this paper offers higher accuracy and provides technical support for accurately distinguishing between natural and non-natural earthquakes in seismic monitoring.

A rapid prediction method of seismic-induced damage in high-speed railway ballastless track simply-supported bridge system is proposed based on convolutional neural networks. To obtain more information of seismic motion, one-dimensional seismic motion data is transformed into three-dimensional image through continuous wavelet transform as the input of convolutional neural network. The reliability of the proposed method is validated by comparing with results in damage samples database. The influence of different hyperparameters of convolutional neural networks on prediction results and training duration are analyzed, and a combination of hyperparameters of convolutional neural networks optimized by Bayesian optimization is obtained. The time required for seismic analysis of high-speed railway ballastless track simply-supported bridge system using different seismic analysis methods is compared. The optimized convolutional neural network is utilized to predict seismic-induced damage of different key components in high-speed railway ballastless track simply-supported bridge system. The research indicates that the initial learning rate is the most significant factor affecting the accuracy of network prediction, while the learning rate decay factor, batch size, and number of training epochs have certain effects on the network prediction results. The training duration of convolutional neural network is mainly determined by the number of training epochs and batch size. The proposed method demonstrates high prediction accuracy for seismic-induced damage in various components of high-speed railway ballastless track simply-supported bridge system, and the network structure exhibits high applicability. The optimized convolutional neural network has shorter training time and more accurate prediction for seismic-induced damage in high-speed railway ballastless track simply-supported bridge system. The research findings can provide reference for rapid repair of seismic-induced damage in high-speed railway systems after earthquakes.

Infilled reinforced concrete (RC) frame structures are one of the most common structures. It is found that infilled walls have a significant impact on seismic performance of RC frames in past earthquake damage investigations and experimental tests. To accurately and rapidly assess seismic damage states of infilled RC frames after an earthquake, 660 infilled RC frames were firstly designed based on different building structure information (i.e. the seismic design intensity, constructed period, number of stories, story height, number of bays and the filling rate), then the non-linear time history analysis was performed for the 660 infilled RC frames with 10 ground motions in OpenSees. 6 600 data points were gained from the analysis, resulting in a dataset which was used to develop seismic damage state assessment models of infilled RC frames. Based on the dataset, nine machine learning models predicting seismic damage states of infilled RC frames were developed using naive Bayes (NB), K-nearest neighbors (KNN), decision tree (DT), artificial neural network (ANN), random forest (RF), adaptive boosting (AdaBoost), extreme gradient boosting (XGBoost), light gradient boosting machine ( LightGBM), category boosting (CatBoost) algorithms. The results indicated that CatBoost and RF models had the highest prediction accuracy for the seismic damage state which was 0.93 in testing dataset, followed by LightGBM and XGBoost models with an accuracy of exceeding 0.90. Compared with actual damage investigated in the past earthquakes indicating that RF and CatBoost models achieved an identical accuracy of 47%. However, the difference in the remain damage states within one damage state level occupied 76% for CatBoost model, which was higher than that of RF model. Based on the CatBoost, importance analysis was performed for different input variables. It is found that three input variables had the greatest impact on infilled RC frame, including seismic design intensity (SDI), peak ground velocity (PGV) and the spectral acceleration at S_a(0.4 s). Furthermore, the importance of the number of stories on the seismic damage state for infilled RC frames increased as the increase of the number of stories.

This paper investigates the effects of spatial effects of seismic ground motions on the seismic responses of long-span cable-stayed bridges, considering their large range and notable variations in local site conditions. Shaking table tests were conducted on a cable-stayed bridge, and the dynamic responses (acceleration, displacement, and strain) of critical sections of the cable-stayed bridge were compared and analyzed. The results demonstrate the significant influence of spatial effects of ground motion on the dynamic responses of the cable-stayed bridge. Specifically, the analysis reveals that the wave passage effect has the least impact on the dynamic responses of the cable-stayed bridge, followed by the combined effect of the wave passage and coherence, while the combined effects of the wave passage, coherence, and local site conditions exert the largest influence. Taking the main tower as an example, the maximum acceleration, displacement, and strain responses increased by 55.69%, 62.37%, and 67.37%, respectively, when spatial seismic motions incorporating the wave passage effect, coherence effect, and local site effects were considered, as compared to uniform excitation. Consequently, the seismic responses of cable-stayed bridges may be underestimated if only uniform excitation or the wave passage effect is considered. It is therefore imperative to comprehensively account for the effects of wave passage, coherence, and local site conditions of the spatial ground motion in the dynamic response analysis of long-span cable-stayed bridges.

In order to study the seismic performance of the new adobe wall structural system. A one-story foot-scale new adobe wall building test model with plan size of 3.6 m×3.0 m and storey height of 2.45 m was designed and fabricated for the shaking table test, and El Centro waves and Yunnan Ludian (LD) waves were selected for the test to carry out unidirectional loading of seismic waves in the X and Y directions, respectively, and the test model was measured to be under the peak acceleration of 0.10、0.22、0.40、0.62、0.90 g, the spectral characteristics, acceleration response, displacement response, torsion response, and strain response of the dark column reinforcement were measured. The results show that: the model is only slightly damaged in the exceeding 9 degree rarefied intensity, and there is no obvious cracking and damage phenomenon under other seismic excitations. With the increase of seismic excitation, there is no obvious change in the self-resonance frequency of the model, and the acceleration amplification coefficient gradually decreases, with the maximum amplification coefficient of 3.666 at the roof. The angle of the interstorey displacement reaches the maximum of 1/399 when the peak acceleration is 0.62 g. The overall structural torsion is 1.599 at the peak acceleration of 0.62 g, and the displacement angle of 0.62 g reaches the maximum of 1/399. Because of the basic homogeneous symmetry, the structural torsion is 0.666 at the roof. Because the model is basically homogeneous and symmetric, the overall torsional effect of the structure is not obvious. The strain value of the steel reinforcement of the dark columns increases gradually with the increase of the seismic excitation, but the strain value of the steel reinforcement is small and the performance of the steel reinforcement fails to give full play. In summary, the new adobe wall model has good seismic performance, and the study provides a test basis for expanding the application range of adobe buildings.

In order to investigate the seismic performance of the suspended structure with viscous damper during earthquakes, the shaking table tests were carried out on two 1/20 scaled models of suspended structures, which were equipped with rigid rods and viscous dampers respectively. The dynamic characteristics, damping ratio, structural response and damping effect of model structures were researched. The test results show that compared with the common suspended model structure, the natural frequency of the suspension damping model structure is reduced, while the damping ratio is improved, especially the first frequency and the corresponding damping ratio. The peak accelerations of the top of the main structure and the fifth suspended-floor of the suspension damping model structure are less than those of the common suspended model structure, but the damping amplitude of the peak acceleration of the fifth suspended-floor is even up to 94.34%. The maximum displacement of the top of the main structure is also distinctly smaller than that of the common suspended model structure. Different seismic wave input has different damping effect, in which the effect of Taft wave is the best, the El Centro wave is the second, and the last is artificial wave. However, the maximum relative displacement between the main structure and the suspended-floor is greater than that of the common suspended model structure, which shows that the stronger the connection between the main and secondary structures, the smaller the relative displacement. It also indicates that the suspension damping model structure takes advantage of the swing of the suspended-floors and the viscous dampers to consume energy, so that the suspended structure with viscous dampers has better effect of energy dissipation and vibration reduction.

To investigate the impact of socket depth on the seismic performance of socket structures, a group of a cast-in-place pier and precast piers considering the variation of socket depth were designed and fabricated, of which the socket depths of the socket piers were 1.0d (where d represents the side length of the square pier), 0.8d, and 0.6d, respectively. Pseudo-static tests were conducted to analyze the damage patterns, hysteresis performance, and energy dissipation capacity of the piers, allowing for an examination of the influence of socket depth on their behavior. The test results demonstrate that the bearing capacity of the four groups of specimens is similar. The socketed piers with a socket depth of 1.0d exhibit damage patterns and hysteresis curves that are comparable to those of cast-in-place piers, indicating equivalent seismic performance. However, the ductility of the socket piers with embedment depths of 0.8d and 0.6d is weaker than that of the cast-in-place piers due to damage to the footing. By comparing the damage phenomenon, the relative relationship between the shear bearing capacity of the footing sidewall and the ultimate bearing capacity of the column is analyzed under the existing reinforcement ratio. As the socket depth of the pier decreases, the shear capacity of the groove sidewall decreases. This increases the probability of damage to the groove sidewall due to the reduced shear bearing capacity. Therefore, it is crucial to consider the relative relationship of the bearing capacity between the footing and the column when selecting the socket depth in the design to prevent damage in the footing position.

In order to analyze the trend and significance of the influence of different factors on the seismic performance of flexible connection autoclaved areated concrete(AAC) masonry filled wall steel frame structures, the feasibility of the modeling method and parameters selection was verified based on the results of pseudo-static tests, nine orthogonal finite element models were established by ABAQUS, and variable parameter analysis was carried out on axial compression ratio, reserved seam width, spring stiffness and height span ratio of models. The results show that the effects of reserved joint width, stiffness of energy-consuming materials, height span ratio and axial compression ratio on seismic performance of the structure decrease successively. On the whole, it is beneficial to increase the reserved joint width, reduce the stiffness of energy-consuming materials, high span ratio and axial compression ratio to improve the seismic performance of the structure. In actual application, the recommended value for the joint width between flexible connection AAC wall and frames is around 40 mm, and the stiffness of the filler between reserved joints is between 100~500 N/mm.

Traditional partially encased composite(PEC) beams usually adopt straight web plates, however, such web plates exhibit lower shear carrying capacity and lack mechanical interlock with concrete, which hinders their collaborative performance. To address the adverse effects of straight web plates, this study proposes the utilization of corrugated web plates as a cross-sectional replacement, leading to a novel corrugated web PEC beam. Through low-cycle repeated loading tests, the seismic ductility performance of the corrugated web plate PEC beam under earthquake loads was analyzed. The influence of flange thickness and shear-span ratio on the load-bearing capacity, failure mode, deformation capacity, hysteresis energy dissipation capacity, and stiffness degradation of the sinusoidal corrugated web PEC beam were investigated. The research demonstrated that a properly designed corrugated web plate PEC beam exhibited favorable load-bearing capacity and seismic ductility performance. The shear-span ratio significantly affects the failure mode and ductility of the beam, while increase the flange thickness effectively and enhances the load-bearing capacity of section. However, excessively thick flanges can impact the collaborative performance between the steel section and concrete. In this experiment, specimens with smaller flange thickness and larger shear-span ratios demonstrated high compatibility between the steel section and concrete, resulting in full utilization of their ductility and energy dissipation capacity.

In this study, based on the extended Kalman filter (EKF), an approach is proposed for identifying unknown ground motion time history and structural parameters by using the absolute acceleration measurements. First, the interfacial force of the lower boundary of the 1st floor is considered as the unknown excitation. The EKF-based approach is employed for the identification of this unknown excitation and the remaining structural stiffness parameters. Then, by using Fourier transform, the structural natural frequencies can be obtained from the acceleration measurements and used for identifying structural stiffness of the 1st floor. Finally, based on these identified structural parameters, the unknown interfacial force mentioned above is updated and employed for identifying ground acceleration with the aid of Newmark-β method. The effectiveness of the proposed approach is validated via a numerical frame structure and a five-story experimental model under shaking table tests.

There is a significant difference between the spatial distribution of ground motion along cross-fault regional areas (Ground motion in the area of the extreme proximity to the fault where the cross-fault engineering structures are located, referred to as cross-fault ground motion) and near-site ground motion. The lack of records on cross-fault ground motion poses challenges to studying the seismic resistance of cross-fault structures. This paper aims to outline the basic theory of the broadband hybrid method for simulating ground motion, and examines the distribution pattern of cross-fault ground motion using the Zemu River fault as a case study. The results indicate that the simulated cross-fault ground motion aligns with the fault sliding mode, displaying significant directional, up-disk effect, and slip-impact effect. Generally, the intensity of the simulated cross-fault ground motion follows a certain attenuation law. However, it is influenced by fault rupture, which can result in irregularities or even a counter-law phenomenon. Furthermore, the actual location of surface rupture and the zone of large slip along the fault have a substantial impact on the distribution pattern of cross-fault ground motion. By employing the broadband hybrid method, artificial cross-fault ground vibration time series can be generated to address the lack of recorded cross-fault ground motion. This methodology provides substantial support for research on the seismic resistance of cross-fault structures.

In order to reasonably represent the fully non-stationarity and randomness of the land acquisition seismic process, a random seismic model based on a combination of phase difference spectrum and power spectrum has been developed. Firstly, the relationship between phase difference and non-stationarity of seismic ground motion was elucidated, and the identification and statistical analysis of phase difference spectrum model parameters were carried out using strong motion records. Secondly, power spectrum models with deterministic and stochastic parameters were used to obtain the values of deterministic parameters and the normalized optimal probability distribution of stochastic parameters, respectively. Finally, by selecting representative point sets of basic random parameters, the corresponding representative time history set of seismic acceleration was obtained. The calculation example shows that the method in this paper only requires 2 or 4 basic random variables to simulate the seismic acceleration process with natural variability and rich probability information, and the simulated acceleration response spectrum fits well with the strong motion records. The study lays the foundation for applying probability density evolution theory to the random seismic response and seismic reliability refinement analysis of complex engineering structures.

As more and more water conservancy projects are constructed in the seismogenic fault zone in the western region, the probability of water conservancy projects encountering near-site vibration is also increasing. Due to the shallow buried depth of near-fault ground motions, the assumption of vertical incidence of seismic waves is no longer applicable. At present, there is little research on the seismic response of the intake tower under the oblique incidence of near-fault ground motions. In this paper, taking an engineering intake tower as an example, a three-dimensional plastic damage finite element analysis model of the intake tower is established, and the nonlinear response analysis of near-fault pulse ground motion SV wave under multi-angle oblique incidence is carried out. The results show that the displacement response of tower top under the oblique incidence of near-fault ground motion SV wave increases significantly, and the damage area and damage degree of tower body increase with the increase of angle.

In order to solve the problems of low out-of-plane stiffness of the energy dissipation web of traditional shear plate damper and easy to occur out-of-plane buckling under large deformation, a sinusoidal waveform steel plate damper was proposed. Considering the placement direction of the waveform web and the yield strength, four specimens were designed and tested under low cycle repetitive load to analyze the hysteretic performance, bearing capacity and stiffness degradation. The results show that strong energy dissipation capacity, good ductility, and stable hysteresis performance under large deformation is the characteristics of sinusoidal wave plate damper, and the transverse wave steel plate damper with low peak load is the best. The energy dissipation capacity and ductility are better than those of the vertical waveform steel plate damper, and its stiffness degradation rate is higher than that of the transverse waveform steel plate damper. The hysteretic curve of webs made of low yield point steel is full.

For super high-rise shear wall structures with large stiffness and small inter-story displacements, the viscous dampers are arranged in the refuge floors using toggle-brace-dampers , and there are still deficiencies in the value of the amplification coefficient and the arrangement. The toggle-brace-damper is a displacement amplification device, which improves the energy dissipation capacity of the damper by amplifying the axial travel of the damper. From the perspective of geometric analysis, the structural of the reverse toggle-brace-damper vibration mitigation efficiency and arrangement optimization are explored: The analytical equation for the maximum displacement of the damper axis of the reversed device is derived, and the dynamic response of the structure under the device with different arrangement heights and different amplification factors is compared. When the arranged frame span is too large, the dynamic response of the structure before and after the setup is compared by setting up the device of the overhanging truss. The results show that the efficiency of the toggle-brace-dampers system is related to the height and span of the arranged frame, and the theoretical amplification coefficient determined according to the angle is not proportional to the vibration mitigation efficiency of the structure. The structural vibration mitigation efficiency is higher after the optimized arrangement than the original arrangement. The optimized scheme has been verified in actual projects, i.e., a reasonable selection of the amplification factor and arrangement can improve the additional damping ratio of the structure to a greater extent.

Aim at the complexity of response analysis methods for energy dissipation systems composed of series-parallel layout II inerter dampers (SPID-IIs) in high-rise structures under random excitation, closed-form solutions for structural displacements and damping force response of SPID-IIs are proposed. Based on the proposed solutions, the influence of real mode number on analysis accuracy was studied, and the feasibility of the layout strategy of SPID-IIs installed on floors with interlayer displacement exceeding the limit of the main structure was explored. Firstly, based on the mechanical construction diagram of SPID-II and its setting method between adjacent floors in structures, a differential constitutive relationship between the damper damping force of SPID-II and the horizontal displacement of structural nodes was established, and the coupled seismic motion equation of the high-rise building and SPID-II was reconstructed. Secondly, the real mode decoupling method is used to obtain the concise equivalent dynamic parameters of high-rise structures, and the power spectrum quadratic decomposition method is applied to energy dissipation systems to derive the accurate quadratic solutions of the power spectrum of a series of responses such as the absolute displacement of high-rise structure nodes relative to the ground, interlayer displacements of vertical components of the high-rise structure and damping force of SPID-IIs. Then, a concise closed-form solutions of the 0-2nd order spectral moments of those design parameters of the energy dissipation system subjected to random seismic excitation modelled by double filtered white noise were derived. Finally, the correctness of the method proposed in this paper was verified through numerical examples, and the influence of the number of real mode shapes on the accuracy of energy dissipation system design parameter analysis was studied, as well as the influence of the position of inertial dampers on the seismic reduction effect. Results shows that for the response analysis of multi-degree of freedom energy dissipation structures, it is recommended to use the number of vibration modes corresponding to the cumulative mass participation coefficient reaching 100% in the original structural free vibration analysis, which can achieve stable accuracy and improve the analysis efficiency of energy dissipation systems with SPID-IIs installed in large and complex high-rise structures. The proposed strategy for setting SPID-IIs with appropriate mechanical parameters can effectively reduce the seismic response of high-rise structures by placing a SPID-II in the floor where interlayer displacement of the original structure exceeds the limit or in the middle floor between three adjacent floors where interlayer displacements of the original structure exceed the limit. The proposed closed-form solutions method and the layout strategy of SPID-IIs in high-rise building can provide useful reference value for the practical engineering application of SPID-IIs.

The braking behavior of trains will notably affect the longitudinal movement of suspension bridges. Therefore, it becomes imperative to delve into the longitudinal movement of kilometer-level railway suspension bridges under the influence of train braking. This study takes a long-span railway suspension bridge with a main span of 1060 meters as its research object, and explores the response characteristics of its longitudinal movement at the girder end when subjected to train braking forces and the controlling effect of a fluid viscous dampers (FVDs) by numerical simulation. Firstly, the engineering background of the kilometer-level railway suspension bridge and the finite element model established using ANSYS software are introduced. Then, the loading and solution methods for the longitudinal movement of the long-span railway suspension bridge, the finite element simulation methods for bearing friction and FVDs, and the simulation method for braking force are described. Subsequently, the effects of different braking positions and consideration of bearing friction on the longitudinal movement response of the suspension bridge are investigated. Finally, the control effect of FVDs on longitudinal movement is studied, and parameter analysis is conducted. The results show that as the braking position approaches the point where the train exits the bridge, the shape of the longitudinal displacement curve becomes more similar to a sine function. Bearing friction has a certain control effect on the displacement response at the girder end under train braking, but its control effect on the velocity response is not ideal. The utilization of FVDs effectively controls both displacement and velocity responses at the girder end of the suspension bridge under train braking. The optimal control effect is achieved when using a FVD with a damping coefficient of 2500 kN•(m/s)^-α and a velocity exponent of 0.1.

By combining the partial self-centering shape memory alloy(SMA) braces (SCB) with the shear slotted bolted connection-very short shear link (SSBC-VSSL), an innovative self-centering shear link (SC-SL) used in the eccentrically braced frame (EBFs) that has an advantage of good energy dissipation capacity and self-centering capacity, low damage and excellent seismic resilience capacity is developed. Cyclic tests were carried on SC-SL specimens designed by the proposed method, so the deformation modes and hysteresis curves can be obtained. Based on the test results, ten SC-SL models with the effect of SMA areas, high-strength bolt pretensions and shim friction coefficients were designed and analyzed by the validated finite analysis (FE) method. The results show that the VSSL is in the relative static during the slip stage, then the development process of very short shear link (VSSL) in SSCB-VSSL including yield, bearing, energy dissipation and inelastic deformation are appeared sequentially during the non-slip stage, while the SMAs in SCB are always subjected to elongation, thus increasing the energy dissipating capacity and self-centering capacity. Finally, the simplified mechanical models of SC-SL in the slip stage and non-slip stage are proposed by using the FE results, which can provide a new way to give important seismic design reference for EBFs.

In order to study the mechanism of pile-soil-structure dynamic interaction and analyze its influencing factors, the horizontal dynamic response laws of pile-soil-structure system under different superstructure mass, different input wave frequency and acceleration peak input are analyzed and discussed by using the method of shaking table test and numerical simulation analysis. The soil model of the test foundation is medium hard soil, and the shear speed is about 213 m/s. The pile group foundation is composed of five foundation piles with a length of 1.35 m and a diameter of 0.1 m arranged in cross shape. The superstructure model is simulated by mass block. The experimental results show that the bending moment and shear force of pile body are the largest at the pile-bearing joint, and decrease with increasing depth. With the increase of the superstructure mass, the acceleration reaction between soil and pile foundation increases significantly, and the bending moment and shear force of the pile body also show an increasing trend. With the increase of the input sine wave amplitude and frequency, the motion interaction becomes larger, and the bending moment and shear force in the pile body become larger. The change of superstructure quality has the greatest impact on the dynamic interaction of the pile-soil-structure system, followed by amplitude and frequency.

Underground rail transit is the lifeline of urban transportation. Due to the frequent occurrence of earthquakes and the increasing frequency of train entry and exit in China, the structure of subway stations is susceptible to both seismic and train loads. However, the current research on the seismic performance of subway station structures has not considered the practical problem of seismic and train load coupling. In order to ensure the safety of subway station structure when the seismic struck as the train was entering and leaving the station occur, a three-dimensional refined calculation model of soil, subway station and track interaction is established. The load form of train braking entering the station is selected to apply on the track surface, and the three-dimensional seismic wave is input at the bottom of the model. The variation law of stress, displacement and acceleration of subway station structure under the two working conditions of seismic action and seismic train coupling action is compared and analyzed. The results show that under the coupling effect of seismic and train, the peak stress at the bottom of the lower column is the highest, which is the weakest position of the station structure. After deformation occurs, the maximum yield stress value is first reached and plastic failure occurs. In the early stage of train arrival (0~5 s), the acceleration and displacement amplitudes of the board under seismic train coupling are greater than those under seismic action. In the later stage of train arrival, the acceleration and displacement of the board under both working conditions are basically the same. Under the coupling effect of seismic and train, when the seismic acceleration is small and the train speed is high, resonance occurs when the vibration frequencies of the two are close, leading to the amplification of the vibration of the plate.