Aiming at the problem that the existing analysis methods for the floating offshore wind turbine （FOWT） cannot efficiently predict the random dynamic response when considering the nonlinear coupling model，a fast calculation method of the random response of the FOWT nonlinear system under coupled wave excitation based on the statistical linearization algorithm is proposed，and its application in the vibration control of FOWT is studied. Taking the Spar FOWT as the object，a nonlinear model under the wave coupling excitation of 4-DOF is established based on the Lagrange equation，and the accuracy of the model is verified. On the basis of the established nonlinear model，a random vibration analysis method for this Spar FOWT based on statistical linearization algorithm is proposed，and the method is verified in many aspects. The results show that the method can improve the calculation efficiency by 4~5 orders of magnitude and has sufficient accuracy. The method is applied to the vibration control of the FOWT. The optimization of the control parameters and performance analysis of the FOWT under the control of TMD are efficiently realized，and the optimal control parameters of TMD are obtained. It is found that the vibration control effect of TMD on the FOWT tends to decrease gradually with an increase of the sea state level. In general，the proposed method has high accuracy，high efficiency and generality in different sea conditions，and also provides an efficient analysis method for the design optimization，fatigue analysis，reliability analysis and other research based on statistical characteristics of FOWT.

Vibration energy harvesting technology is expected to solve the problem of self-powered wireless sensor nodes. By introducing nonlinear magnetic force，a magnetic coupling array piezoelectric energy harvester （MA-PEH） is designed in this paper. The nonlinear magnetic force model is established based on magnetic dipole method. The restoring force model of composite beam is obtained by finite element method. According to Newton’s second law and Kirchhoff’s law，the dynamic model of the system is established. The influence of excitation amplitude and excitation frequency on dynamic response is analyzed by simulation and verified by experiment. The results show that under the action of nonlinear magnetic force，the system appears chaos and periodic motion between wells near the resonant frequency，which can help to broaden the working frequency band of the energy harvester. As the excitation frequency changes，each composite beam always maintains the same motion state. When the system is in a periodic motion state between wells，only one type of beam is in a high-energy output state. When the excitation frequency is near the resonant frequency of the two beams，increasing the excitation amplitude will lead to the boosting of the output of one type of beam，while the output of the other type of beam will be suppressed. The research provides theoretical guidance for the design of array piezoelectric energy harvesters and new research ideas for improving the output performance of piezoelectric energy harvesters.

The vibration of high-rise tower structure requires active control. However，the problem of direct displacement observation and high-efficient low-dimension control strategy under single point needs to be solved. In this paper，a set of real-time continuous observation method of displacement from the inside of the tower is established by using modern video metrics technology，which provides the most direct displacement observer for the active control of the high tower. It avoids the tedious calculation process of building an acceleration observer，and then combining with the assumed external load to calculate the displacement after filtering. By using the equilibrium system space transformation and the equilibrium truncation method，a low-dimension controller is established with AMD at the top of the tower，which can effectively preserve the main dynamic characteristics of the structure. Taking a 700-meter-high tower under construction as an example，the active control simulation analysis under wind-induced vibration and earthquake is carried out. The results show that the control effect of the low-dimension controller with a few displacement states as feedback is basically the same as that of the full-dimension controller based on full state feedback，which can be used as an active control strategy for high tower structures.

A theoretical model for statics and dynamics of bistable asymmetric cross-ply composite laminated square plates is established. Three equilibrium states are determined by curing analysis in statics. Meanwhile the super-critical pitchfork bifurcation with temperature difference as the control parameter is explicated in the process of curing. Two stable states and one unstable state are demonstrated by stability analysis. The potential energy curve with two potential wells is depicted，which can contribute to studying dynamic snap-through. Moreover，the dynamic bifurcation for the equilibrium points is induced by introducing damping in dynamics. The influence of the base excitation frequency on the dynamics is discussed by numerical simulation. When the excitation frequency is located in a certain range，the large-amplitude dynamic snap-through and nonlinear vibrations with two potential wells can occur. The dynamics behaviors of the bistable system are overwhelmingly dominated by periodic vibration，quasi-periodic vibration and chaotic vibration. The certain frequency range，where the large-amplitude dynamic snap-through and nonlinear vibrations with two potential wells can occur，can be defined as a certain frequency broadband which proofs bistable asymmetric cross-ply composite laminated square plates to be applicable to bistable energy harvesters.

To explore the dynamic behavior mechanism between vibrating wheels and compacted pavement during subgrade vibration compaction. To provide theoretical support for continuous compaction monitoring technology and intelligent compaction of pavement. Based on the consideration of the mass of the vibrating soil，a 3-degree-of-freedom vibratory roller-soil coupling dynamic model is established. The nonlinear dynamic response characteristics of the vibratory roller vibration wheel in the compaction process are studied by analyzing the time-frequency domain plot of the vibrating wheel，Poincaré Map，Largest Lyapunov Exponent，and the dynamic contact force of the wheel-soil. The numerical simulation results show that with an increase of subgrade compaction，the motion of the vibrating wheel evolves from a single cycle to a multi-cycle motion，and finally enters a chaotic state. In the process of evolution，the vibration wheel acceleration frequency domain characteristics from the initial single fundamental frequency to the fundamental frequency accompanied by an integral frequency harmonic transition. In the end，the fundamental frequency and 1/2 times the subharmonic are the main parts. After entering the chaotic state，reducing the excitation force and increasing the excitation frequency can make the chaotic state degenerate to approximately a single cycle motion. Among them，the response characteristics of reducing the excitation force are weak 1/3×，2/3× and high harmonics in the frequency domain response. Among the response characteristics of increasing the excitation frequency，the single cycle of the compaction movement of the vibrating wheel is more obvious.

To solve the problem of friction pendulum system without pull-out resistance under external load excitation，based on electromagnetic principles and semi-active control ideas，an electromagnetic chuck friction pendulum composite isolation system （ECFPS） based on electromagnetic force combined with traditional friction pendulum is proposed. The structural characteristics and energy dissipation mechanism of the composite isolation system are introduced，and theoretical formulas for electromagnetic suction force，equivalent stiffness，period，and equivalent damping ratio of ECFPS are derived based on electromagnetic principles. The ECFPS model is established. A 1∶3 scaled ECFPS specimen is designed and fabricated to investigate the hysteresis performance under different input currents，vertical loads，and displacement amplitudes. The effects of different currents on the anti uplift performance are investigated. The experimental results show that the theoretical values are in good agreement with the experimental values，which verifies the correctness of the theoretical formula derivation. The changes in equivalent stiffness and unit cycle energy consumption are significant，with the maximum variation amplitude of 19.81% and 28.16%，respectively. As the current of the electromagnetic suction cup increases，the anti pull performance of the ECFPS system improves，achieving the vertical resisting pull function of the system.

The long-span continuous beam bridge with parallel twin steel box decks is common in engineering practice，but the complex vortex shedding and interaction of parallel twin steel box girders may cause significant vortex-induced vibration （VIV），affecting the fatigue performance of the structure，driving comfort，and possibly causing social panic. This paper takes parallel twin box girders as the research background，and a large-scale segment model vibration and pressure measurement wind tunnel test is carried out. The evolution characteristics of the distributed aerodynamic force in the entire vertical bending vortex vibration process （before the vortex vibration，the ascending zone，the amplitude extreme point，the descending zone，and the end of the vortex vibration） under different spacings are compared，and effective aerodynamic measures to control the vortex vibration of the parallel twin box girders are proposed. The study shows that the vortex-induced vibration lock-in regime of the parallel twin steel box girders is long，the amplitude is large，+3° is the most unfavorable angle of attack （AOA） and the frequency multiplication effect of the vortex excitation force is related to the amplitude and the spacing between the box girders. When the spacing makes the inter-slot vortex fully developed，it significantly increases the pulsation of the aerodynamic force distributed near the slot. In the case of small spacing and low wind speed，the distributed aerodynamic force on the lee side of the upstream and downstream girders enhances the vortex-induced vibration. At large or small spacing with high wind speed，the distributed aerodynamic force on the upper surface near the slot and the position of the inclined web of the downstream girder plays a major role in enhancing the vortex-induced force，which is the inducement of the large amplitude of vortex-induced vibration of the parallel twin box girders. A comprehensive aerodynamic control measure of setting the apron between the slots and setting the wind fairing at both ends is proposed，which can cut off the propagation path of the vortex between the slots and reduce the contribution of the distributed aerodynamic force at the bridge deck. The measure can effectively reduce the vertical bending vortex-induced vibration of the parallel twin steel box girders.

An underwater acoustic structure with metal perforated plate inserted into the traditional anechoic coating is proposed to improve the pressure resistance and sound absorption performance of the structure. The deformation of the acoustic structure under different hydrostatic pressure is studied by the static method. By establishing the acoustic finite element equation under the action of hydrostatic pressure，the sound absorption effect of the acoustic structure under different hydrostatic pressure is analyzed. Compared with the traditional anechoic coating，under the hydrostatic pressure of 0 to 6 MPa，the acoustic structure achieves better broadband sound absorption in the mid-to-high frequency range. In addition，the effects of the thickness，material and porosity of the metal perforated plate on the sound absorption performance of the acoustic structure are discussed. The research shows that the underwater anechoic coating with metal perforated plate is an effective design to improve the pressure resistance and sound absorption performance of underwater acoustic structure.

Based on the measured acceleration response under Typhoon Kompasu，the modal parameters of the 356 m Shenzhen Meteorological Gradient Tower （SMGT） are identified. The Non-dominated Sorting Genetic Algorithm Ⅱ（NSGA-Ⅱ），which is a fast and elitist genetic algorithm，is applied to update the finite element （FE） model of SMGT. The results show that the vibration modes of SMGT are very dense，and the involvement of cable vibration modes is obvious. The fundamental frequencies of SMGT in X and Y directions are 0.614 Hz and 0.603 Hz，respectively，and the damping ratio of the first 3 order bending modes are about 1%~2%. The tower density and cable elastic modulus have a significant effect on the modal frequency and mode shape of SMGT，the lineic mass of high-rise cable and tower elastic modulus also have a certain influence，while the cable tension has a relatively low influence on the modes of SMGT. The wind-induced response of the updating FE model is higher than that of initial model，and closer to the actual measurement，which verifies the accuracy of the updating FE model.

Vehicle-assisted bridge damage identification has great application potential，but it is still difficult to extract damage-sensitive features from multi-source monitoring data and accurately evaluate the bridge damage status. To solve this problem，an Attention-LSTM-based Feature Fusion Model （ALFF-Net） is proposed. The model improves the perception ability of Bi-LSTM cells for multi-scale feature information in time series data through a preset data reconstruction layer. Furthermore，by employing attention mechanism and feature fusion strategy，the model reduces the prediction difficulty of downstream branches of deep neural networks and further improves the modeling ability for the important dependency relationships in the sequence data. A monitoring dataset under different road roughness and vehicle speeds is generated through a vehicle-bridge interaction system simulation，and the bridge damage identification performance of the ALFF-Net model is comprehensively tested. The results show that the ALFF-Net model improves the damage identification accuracy by up to 19.30% compared to the classical LSTM network while significantly reducing computational costs，and the identification errors under different road roughness levels are less than 3%. Moreover，by comparing the identification accuracy of the ALFF-Net model under different data-driven schemes，the robustness of the bridge damage detection results with synergistic multi-source monitoring data is verified.