Large modern hot strip mill is a complex electromechanical system.Vibration of rolling mill is a serious problem which effect the stability of production. By using torque telemetry system to test the torsional vibration of mechanical drive system，it is found that two typical characteristics of the strong or weak vibration exist in the waveform and spectrum of torsional vibration,meanwhile, there are harmonics in motor current and rolling force during rolling mill production, and when harmonic frequency is close to the natural frequencies of main drive system, rolling mill present a heavier torsional vibration on mill main drive system. For the study of its mechanism ,using computer simulation , the correlation of main drive torsional vibration and harmonic excitation by motor current and rolling force is solved. Research shows that with the harmonics excitation change, torsional vibration presents polymorphic variation strong or weakly on main drive system,which lay a theoretical foundation for vibration suppression.

Application and research status of cast steel is introduced in steel structure, especially in steel frames. A cast modular joint and its design approaches are proposed, appropriate for rectangle or square tubular column to H-beam connections. The cast modular joint is conceived base on the concept of utilizing stable plastic dissipation of panel zone. Nonlinear finite element analysis is conducted to simulate monotonic loading and cyclic loading of the cast modular joint and traditional welded connections. Rupture Index is introduced to evaluate the propensity for ductile fracture initiation of different joints under large plastic deformation. Energy dissipation features are analyzed and demonstrated from multiple aspects, including stiffness, load carrying capacity, ductility, energy dissipation capacity, energy dissipation mechanism and plastic deformation modes. Research shows that the proposed cast modular joint exhibits better seismic performance than traditional welded connections, with sufficient developed dissipation capacity of seismic structure system on the premise of no ductility loss.

In order to identify the damping characteristics of transmission towers, Empirical Mode Decomposition (EMD) and Ensemble Empirical Mode Decomposition (EEMD) were employed to obtain Intrinsic Mode Functions (IMFs) of time histories of accelerations obtained from aero-elastic wind tunnel tests of transmission tower. The frequency and damping of the tower under various wind velocities and wind azimuths were identified using EEMD, improved Random Decrement Technique (RDT) and Hilbert transformer. The distribution characteristics of damping along with wind velocities in the along-line direction and in the cross-line direction were also investigated. Results show that the vibrating modes of IMFs obtained from EEMD are much clearer than those obtained from EMD. Furthermore, the mode mixing problem is still existed to some extend in the IMFs obtained from EMD. Totally speaking, the natural vibrating frequency of the tower in the along-line direction decreases with the increase of the wind velocity, whereas the natural vibrating frequency in the cross-line direction increases with the increase of the velocity. However, the trend of damping in the cross-line direction with respect to wind velocity is not obvious.

Extracting the impact feature from rotating machinery fault vibration signal, is significant to some faults dignosis. Based on the fact that S tansform is suitable for processing the signal with impact feature, an effective impact feature extracting method with the use of threshold denoising for S transform spectrum is proposed. Firstly, the time-frequency spectrum of a time-domain signal is obtained by S transform. As the spectrum coefficients are complex numbers, the spectrum threshold denoising method is carried out according to the modulus magnitude of coefficients. During denoising both the hard and the soft threshold functions are used, respectively. Most importantly, the optimal threshold estimation is available between zero to the maximum modulus of the coefficients by step iterative algorithm, evaluated with the introduced modified risk function. Finally, inverse S transform is used on the denoised time-frequency spectrum to reconstruct the time-domain impact feature. The processing results of simulated signal and rolling bearing fault vibration signal have shown that using the acquired optimal threshold, S transform spectrum threshold denoising method can successfully extract the impact feature from noise mixed signal, so as to realize related fault dignosis.

To study the effect of the number of blades and guide vanes to the reactor coolant pump radial force under variable flow transition conditions, applying computational fluid dynamics software CFX to the numerical simulation and test of the blades and guide vanes radial force variation, when changing the number of blades or guide vanes. Compared the numerical simulation results with the experimental results, the experimental data error within the control of the numerical simulation data. The results showed that: with the number of blades varying and the flow rate increasing, the radial force of the reactor coolant pump impeller increasing with the number of blades, and reaching the maximum in the number of 7, the periodicity of the impeller to withstand the radial force tends regularity, radial force balance tends to the optimum working condition. With the flow rate decreasing, the position of the impeller radial force offsets and rotates regularity in the spin cycle, and the variation gradient of the offset and rotation is different. The variation gradient of the impeller radial force offset and rotation is significantly greater than the variation gradient corresponding to the large flow increasing. Seen from the comprehensive comparison: when the number of blades is five and the guide vanes is 11,the radial force on the impeller is the minimum.

Based on the classical laminated plate theory and von Kármán-type equations, the nonlinear governing equations of motion for the composite laminated plates under axial loading, in which, the boundary damping is taken into consideration, are derived. Galerkin approach is employed to transform the governing partial differential equations to ordinary differential equations. Using the method of multiple scales, the approximate solution of the plates by deterministic parametric resonance excitation are obtained and analyzed. Results show that the unstable interval bandwidth for the trivial solution is only related with linear damping, whereas the boundary nonlinear damping can weaken the resonance amplitude of the non-trivial solution and reduce the corresponding resonance range effectively. On the basis of nonlinear dynamic model, the resulted Foker-Planck-Kolmogorov equation is analyzed by using the narrow-band random excitation combined with finite difference method. The stochastic bifurcation and jump phenomenon of the response for the amplitude stationary joint probability density caused by the boundary nonlinear damping are sequentially investigated. Results show that a very small increment of the damping will induce a sharp jumping of the stationary motion from the non-trivial solution to the trivial one.

responses, and then the Structured Low Rank Approximation (SLRA) method wasimplemented to achieve the filtered data. Finally the modal parameters are estimated by using complex exponentialmethod (Prony’s method) from the noise reduction signals. The effectiveness of the proposed method was verified byusing numerical and experimental studies.

Three used most commonly specimens：Normal Brazilian disc，flattened Brazilian disc and holed Brazilian disc were simulated by the finite element analysis software LS-DYNA, based on which we adopted flattened Brazilian disc to carry on dynamic splitting tests with granite specimens. By comparing the simulation and test results, we gain the relation of strain rates and tensile strengths of rocks and discuss the optimal specimen size of rock dynamic splitting tests. The results show: Fattened Brazilian disc fits the condition of broking from center well so that the test results will be more credible; For the rock mode used in the paper, we suggest the central angle of flattened Brazil disk to be 20°~24°; If holed Brazilian disc is used in the dynamic splitting tests, the suggested aperture ratio is about 0.4~0.5; Otherwise, the dynamic splitting strengths of rocks are increasing with the increase of strain rate, of which the increase rate decreases while this growing trend is affected by the specimen size observably.

Combining the features of Frequency Slice Wavelet Transform (FSWT) and denoising source separation (DSS), we propose an underdetermined DSS method based on FSWT. This method is used to deal with the blind source separation (BSS) problem of rotating machinery in the case of the number of observed mixtures being less than that of contributing sources. New mixed signals can be reconstructed through the FSWT method, which is effectively way to solve the problem of the dimension insufficient in underdetermined blind source separation. As the same time, the method can solve the problem of time-frequency analysis in FSWT. Applying FSWT-DSS method to the rotor fault detection, we have diagnosed the sudden unbalance phenomenon through the measured fault signals of the rotor. The simulation and experimental results, and their analysis show preliminarily that the FSWT-DSS method is indeed efficient in analyzing the fault diagnosis and it has an important engineering significance for fault detection of rotating machines.

A wind tunnel test is conducted to investigate the aerodynamic behaviour of transmission tower subjected to ambient excitation and the eigensystem realization algorithm is employed in the procedure of modal parameter identification. The concept of modal participation factor (MPF) is introduced, which represents the impulse response amplitude of each mode therefore is an indicator to discriminate system modes from noise. Then the modal energy contribution (MEC) of each mode is evaluated on the basis of MPF as well as corresponding eigenvalue. With the aid of MPF and MEC an auto identification scheme is applied to minimize the subjective influence in modal parameter identification. The experimental results are examined in a variety of wind speeds and wind attack angles and theoretical values derived from quasi-steady assumption are compared as well. The study reveals that the aerodynamic damping ratios of transmission tower are generally positive. The longitudinal aerodynamic damping tends to intensify as the wind speed increases, while in lateral direction no clear trend is observed. There is a nonlinear dependency of aerodynamic damping ratio on mean wind speed therefore the quasi-steady assumption is adjusted by introducing higher-order terms of wind speed.

The method of Empirical Mode Decomposition(EMD) can filter and denoise the nonstationary microseismeic signal.But there exists serious end effect in its application which could affect the decomposition results. For improving the end effect, the wave extension of microseismic data endpoint based on LMS adaptive filter was proposed to the low frequency characteristics of microseismic signal.Trained by the internal data, LMS adaptive filter extend the wave from the endpoint to the left and right sides.The minimum and maximum that were in coincidence with internal signal trend were obtained in extension process and the extension length was determined adaptively by extension algorithm.The wave was decomposed by EMD method and mode components were acquired.For correcting the error brought by the interpolation algorithm, he residual signals were extended again.According to characteristics of the signal frequency, the mode components were reconstructed selectively in order to denoise.The adaptive LMS fiter extension method is verified by the simulation data and the microseismic monitoring data from coal mine, the results show that the method can obviously reduce the end effect of empirical mode decomposition, and have great effect especially to microseismic signals of low frequency.

For improving rainbow shape piezoelectric transducer generating capacity under given dimension, aiming at external force and displacement incentive environment, relations between generating capacity and structural parameters and material properties of rainbow shape piezoelectric transducer were analyzed by simulation. The results show that on external force incentive environment, the output voltages of the rainbow transducer decrease as the width and initial curvature radius increase, the output voltages of the transducer increase as the length and elastic modulus ratio increase, the output voltages of the transducer have a maximum as the thickness ratio increases; on external displacement incentive environment, the output voltages of the transducer decrease as the thickness ratio and length and elastic modulus increase, the output voltages of the transducer increase as the initial curvature radius increases. Moreover, the output voltages of the inside piezoelectric film are larger than the outside piezoelectric film.

In order to improve precision and dynamic response speed of the correlation method for phase difference measurement under the asynchronous sampling circumstance, an improved method for phase difference measurement based on correlation theory is proposed. Firstly, two reference signals which have 90 phase shifted are introduced by using two original signals with the same frequency, then, via digital filter with zero phase, correlation function of the original signals and reference signals are directly computed, some properties about correlation and sine function are adopted to obtain the computational formula for phase difference. Theory analyses and experiment results show that it is not necessary for the proposed algorithm to know the signal frequency in advance, and the measuring accuracy for this method is not affected by whether synchronous sampling or not, it can also measure the dynamic phase difference. Compared with the traditional methods, it is superior to accuracy, robustness and universality.

The transient exterior flow field of a car was computed by LES, and then sunroof buffeting noise characteristic was predicted using FW-H acoustic model based on Lighthil-Curle acoustic analogy theory. Based on the study of the vehicle’s sunroof buffeting noise character, this paper introduces modern optimization algorithm to the sunroof deflector’s design. The main objective of this study is to minimize the weight sound pressure level (SPL) value of driver’s left ear and passenger’s right ear. The deflector’s installation angle、groove width and depth were taken as design variables. Based on twenty sampling points by the method of the Design of Experiment (DOE), the Kriging model was constructed. Then, the Genetic Algorithms were used to optimize the shape of the deflector, and a set of optimal parameters was obtained. Simulation results show that the monitor’s SPL of buffeting noise after the optimization of deflector reduces by up to 21.6% compared with that of the original deflector.

Touching upon that the crack fault of the rotor may occur after the RCP has operated a long time, the fault feature can be identified effectively by the method of the wavelet analysis. Based on the simulation signal of the vibration model of the RCP cracked rotor and the method of CWT, the cross-correlation coefficients between the fault signal and the different wavelet basis which are selected from the wavelet basis library can be computed. After confirming the maximum of the cross-correlation coefficients, the optimal wavelet basis applied to the fault signal of the cracked rotor will be found. And the three-dimensional diagram which covers the power spectrum, the scale of transform and the frequency will be described, once the signal analysis of the fault is completed by use of the optimal wavelet basis. The results of simulation illustrate that the wavelet basis selected by the maximum cross-correlation coefficients can become the optimal wavelet basis, and the fault feature of the cracked rotor can be recognized effectively.

To study the ride comfort of the long span bridge, five tests of vibration for vehicle passing the bridge in different condition were conducted. The effect of the dynamic characters of the vehicle and the bridge, the roughness of the deck and the vehicle speed on the vibration inside vehicle was investigated by analyzing measured vibration data. A time frequency domain method based on wavelet transform was used to extract the energy distribution of vibration inside the vehicle The results show that the dynamic characters of the vehicle have more influence on vertical vibration inside the vehicle. However the dynamic characters of the bridge deck have more influence on lateral vibration inside the vehicle. The vibration amplitude is mainly determined by the roughness and the expansion joint and increase with the vehicle speed. The lateral vibration of the vehicle passing the Cross Sea Bridge has a rapid increase as the vehicle speed exceed 80km/h. The research results have important reference value in evaluating ride vibration comfort of long-span Bridge.