Plane irregular buildings are prone to torsion and cause serious damage to the structure. The analysis and control of structural torsion effect is the key to this kind of structural design. In this regard，the simulation calculation of the planar irregular multi-story frame model in the 8-degree fortification zone was carried out，and the planar L-type four-story frame isolation model of five typical class B buildings was established. The ratios of the length B and width W of the protruding limbs of the structure were set to be 1∶1，1.5∶1，2∶1，2.5∶1 and 3∶1，respectively. The dynamic characteristics and seismic response of different structures were analyzed. The influence of the LRB stiffness ratio and the rigid center position of the isolation layer on the structural torsion was discussed，and the effective control method of structural torsion was proposed. The results show that with the increase of the length-width ratio B/W of the protruding limb，the X-direction torsion of the structure increases gradually，and the Y-direction torsion decreases gradually. Increasing the proportion of LRB stiffness can reduce the torsional effect of the structure. By comprehensively controlling the structural torsion and structural damping coefficient，a reasonable proportion of LRB stiffness can be obtained. The proportion of LRB stiffness of the isolation layer under the condition of 8 degrees earthquake is proposed. At the same time，the rigid center of the isolation layer and the rigid center of the superstructure are arranged on both sides of the mass center of the superstructure to suppress the torsion of the superstructure of the plane irregular structure.

The complex response of the ocean thermal energy conversion platforms in the marine environment makes the structural safety design of cold seawater intake pipes suspended beneath the platform a challenge. Recent research has shown that the in-plane motion of the platform （heave oscillation） triggers out-of-plane vortex-induced vibration （VIV） in the riser connected to the platform and the complex vortex vibration response can cause rapid accumulation of fatigue damage to the riser，resulting in structural damage. In this paper，we focus on the VIV response of a free-hanging riser under multi-degree-of-freedom motion，which has rarely been reported in previous studies. A pool model test to measure the VIV strain information of the riser using fiber-optic grating strain gauges is carried out. After analysis of the experimental results，it can be found that： the maximum oscillation velocity at a large KC number is the main parameter affecting the dominant frequency of the out-of-plane vortex vibration response； the dominant frequency of vibration at a small KC number is twice frequency of the motion of the top platform. By comparing the experimental and numerical results of the free-hanging riser under the three degrees of freedom motion of the platform，it is found that the influence of the VIV on the overall dynamic response of the free-hanging riser is not negligible in this case. These results can provide a reference for further research on the VIV of the free-hanging riser taking into account the influence of the platform motion.

The full-field vibration displacement acquired via digital image correlation method is widely applied in aerospace structural testing and monitoring because of its advantages of high environmental adaptability and non-contact full-field. However，structural damage identification based on full-field vibration displacements faces two critical problems： Current modal analysis methods possess low computational efficiency when dealing with high spatial resolution displacement fields； The baseline-free damage identification method based on modal shape is difficult to extract damage features and has poor anti-noise performance. In order to solve those problems，a frequency domain modal analysis method based on kernel decomposition and joint principal component analysis，and a damage localization method based on pseudo-excitation are proposed. Singular value decomposition is employed to process the full-field displacement fields for obtaining the kernel functions and their coefficients，which contain the local damage characteristics. On this basis，a frequency domain modal analysis enhanced by joint principal component analysis is adopted to evaluate the noise-robust and high spatial resolution modal shapes. The disturbance of local dynamic equilibrium equation caused by structural damage is equivalent to a pseudo excitation force for damage detection. In addition，the local proximity of damage features and sparse spatial distribution of measurement noise are harnessed to optimize the damage localization accuracy via a hierarchical clustering method. Multi-modal information fusion damage index is proposed to improve the accuracy of damage localization. Numerical and experimental results demonstrate the effectiveness of the proposed method.

Aiming at the problem that the recognition accuracy of the model is not high due to the complex and variable engineering environment，a rolling bearing fault diagnosis model integrating Markov transition field and graph attention networks （MTF-GAT） is proposed in this paper. Using the advantage of MTF to retain the time correlation of the signal is applied to transform one-dimensional signals into two-dimensional feature maps，and the nodes and edges of the graph are defined. The graph attention layer can adaptively assign different weights to adjacent nodes to improve the ability of the model to capture useful fault features，and the abstract information of the graph is further extracted through the deep convolution module. By simulating the actual engineering environment，the various fault signals are input into the trained MTF-GAT model for fault diagnosis，and the model is verified by experiments on two data sets. The results show that the proposed model in this paper can accurately complete the task of fault classification in a variety of environments. Compared with other deep learning models，the MTF-GAT model has better recognition accuracy and generalization performance.

A double Gamma distribution model to determine the probability density function （PDF） of the time domain rainflow-range corresponding to the broadband random stress power spectral density （PSD） is proposed，and a neural network method is used to implement the parameter prediction of the model. A series of stress PSDs are given，and the corresponding stress time histories are generated using the time-domain randomization method. The number of rainflow-range is counted for the stress time histories using the rainflow counting method，and the stress rainflow-range probability density values are calculated. Based on the calculation results of each stress PSD mentioned above，the proposed stress rainflow-range probability density double Gamma distribution model is parametrically fitted to obtain a set of corresponding model parameters. The results of the double Gamma distribution model are compared with the Dirlik method and fatigue life prediction is carried out，and the results show that the proposed double Gamma distribution model is more accurate for determining the broadband random stress rainflow-range PDF.

Based on the elastic foundation beam theory，the mechanical model for the seismic response of the vertical shaft is established. The shaft primary lining and secondary lining are simplified as Euler Bernoulli beams and the normal interaction between the primary lining and secondary lining is simulated by using uniformly distributed springs. The differential governing equations of the shaft under the horizontal seismic excitation are derived. The rapid solution of the horizontal seismic response of the primary lining and secondary lining is achieved through the distributed transfer function method and the correctness of the analytical solution is verified by comparison with the finite element numerical simulation method. The peak seismic response at the top of the shaft is parametrically analyzed from the perspectives of elastic foundation stiffness，secondary lining stiffness，shaft outer diameter，and stiffness of the elastic connection layer between the primary and secondary linings，respectively. The results show that the peak seismic response at the top of the secondary lining is greater than that of the primary lining. With the increase of the elastic foundation stiffness，the peak response of the top of the primary lining and the secondary lining decreases. The increase of secondary lining stiffness leads to the obvious increase of peak response at the top of secondary lining. With the increase of the shaft outer diameter，the peak response of both the primary and secondary linings increase significantly. When the elastic connection layer stiffness increases，the peak response of the primary lining increases slightly，but the peak response of the secondary lining decreases significantly.

The graph neural network models have been widely used in the field of fault diagnosis due to the advantage of abundant fault characterization capabilities. However，the existing models only utilize the local information among neighboring nodes when dealing with fault data，and fail to fully extract the global feature information. Meanwhile，in order to overcome the problems of low accuracy and insufficient generalization ability of single model. This paper proposes an ensemble method with multi-scale graph pooling feature fusion and graph convolutional network （MSGP-GCN）. The graph model is constructed from the original signal，and global information is obtained using graph pooling coarsening. Then weights are assigned at different scales based on the degree of the nodes，and the global information is used to update the node features in combination with the weights. The updated node features are input into different classifiers respectively，and the intelligent fault diagnosis result is obtained by majority voting strategy among these classification results. The proposed approach is fully verified by two fault datasets，the SEU simulation dataset and the real coal mill dataset. The experimental results show that the proposed model can effectively improve fault diagnostic accuracy and generalization ability in aforesaid two real datasets，and the average diagnostic accuracy reaches 98.31% and 97.21%，respectively.

Based on the strain gradient nonlocal Biot theory，the analytical solution of the dynamic response of the tunnel lining under the action of P-wave is obtained by using the wave function expansion method and the boundary conditions between saturated soil and lining with the deeply buried circular lining as the research object. The influence of non-local parameters and size factors on the dynamic stress concentration factor （DSCF） is investigated for different incident P-wave frequencies. The results show that when the incident wave frequency is low，the non-local parameters and size factor have almost no effect on the DSCF. As the incident wave frequency increases，the effects of the non-local parameters and size factor on the DSCF become more and more obvious. The non-local parameters are negatively correlated with the DSCF，and the size factor is positively correlated with the DSCF. The maximum dynamic stresses in the lining appear on the right side of the lining. With an increase of frequency，the DSCF in the lining shows obvious radial direction and the dynamic stresses in the lining appear in the right side of the lining. The maximum dynamic stresses in the lining all appear on the right side of the lining，With an increase of frequency，the DSCF in the lining shows obvious fluctuation along the radial direction，and the non-local parameters and size factor have little influence on the distribution pattern of the cyclic stresses in the lining.

The incipient fault characteristics of planetary gearbox are weak and difficult to effectively identify under strong background noise interference and variable working conditions. To address these issues，an improved feature mode decomposition （IFMD） algorithm is proposed to extract the weak fault characteristics of planetary gearbox under time-varying speed conditions. Firstly，for the key input parameters of the FMD algorithm，such as the number of decomposition mode n，the number of filter K，and the length of filter L，which need to be set manually and lack adaptability，an adaptive scale space spectrum segmentation method is proposed to determine the required number of decomposition modes n. On this basis，the Spectral Gini Index （SGI） is used as the objective function，and particle swarm optimization algorithm is used to automatically determine the optimal filter number K and filter length L. Subsequently，the IFMD is applied to perform optimal modal decomposition on the fault signal under the optimal parameter combination，and the decomposed component with the highest SGI value is selected as the sensitive modal component. Finally，significant fault feature orders are extracted from the envelope order spectrum of sensitive component to accurately diagnose the fault type and location of planetary gearbox. The analysis results of variable speed simulation signals and engineering experimental data indicate that compared to the PSO-VMD method，MED method，SGMD method，and fast spectral kurtosis method，the proposed method can extract weak fault information more clearly and comprehensively，thereby improving the characterization ability and diagnostic accuracy of early fault features of planetary gearbox under time-varying speed conditions.

In this paper，the strong earthquake records of 13 soil stations within 41 km from the fault of the 2008 Wenchuan 8.0 magnitude earthquake are selected. These records are combined with the strong earthquake records of 29 calculation points obtained by using the stochastic finite fault method simulation. Moreover，nonlinear least squares fitting is performed by using three near-fault ground motion attenuation models and the fifth-generation ground motion parameter zoning map attenuation model to obtain the peak ground motion acceleration of the Wenchuan earthquake near-fault. The attenuation relationships are obtained and analyzed by considering ±1 times standard deviation and correlation coefficients for reliability. The results show that there are some differences in the attenuation relations obtained by different attenuation models. The peak acceleration obtained by the Shao Guangbiao model is low. The prediction results of Wang Guoquan model has a large deviation for other magnitude earthquakes because the magnitude term is not considered. Besides，the peak acceleration of the fifth-generation ground motion parameter zoning map attenuation model is low.