To overcome the limitations that the traditional bearing life prediction method relies on a database of failure samples and it cannot effectively utilize truncated samples, an intelligent method utilizing both failure and truncated samples was proposed for bearing life prediction. Firstly, the trend model for features characterizing bearing degradation was constructed based on the function principal component analysis (FPCA), and each feature was decomposed into a mean value, an eigenvector and a score vector of function principal components (FPC-scores). Secondly, the optimal life value of each truncated sample was estimated by minimizing the similarity index between its score vector and those of failure ones. Thirdly, all features in the whole life duration of each sample were estimated and reconstructed based on the feature trend model to generate training data. Finally, the prediction model was constructed based on a least square support vector machine for bearing life prediction. The test results of rolling bearings’ life prediction showed that the proposed method can improve the bearing life prediction accuracy with truncated samples, and it is robust to a certain level data missing.

 

The existing simulation results for 2-DOF vortex-induced vibration are not satisfied, especially, the maximum amplitude estimation of its responses. In tests of Jauvtis & Williamson, the maximum amplitude of a cylinder with a low mass ratio could reach 1.5D, this phenomenon was named “super upper branch”. But, few simulation results of current studies can reach 1.3D even unable to capture super upper branch. The reasons were likely to be defects of the turbulence model itself and the unreasonable setting of numerical simulation parameters. Aiming at the problems mentioned above, an improved k-ε turbulence model was used to study the effect of acceleration on responses of vortex-induced vibration here based on the software OpenFOAM and optimize the numerical simulation method. By analyzing responses of amplitude, phase, trajectory and tail vortex, it was shown that 2-DOF vortex-induced vibration can be numerically simulated more accurately with the improved k-ε turbulence model under appropriate accelerations.

 

The dynamic simulation including modeling and automatic switching for flexible multibody systems was investigated using Lagrange multiplier method and the high order rigid-flexible coupled theory. Based on variable topology idea, dynamic equations and the corresponding constraint conditions were established, respectively for detachment, impact, viscous contact and slip contact states. The impulse-momentum method was used to solve the impact initial conditions. The concept of tangential sliding friction force potential energy was introduced to describe the corresponding generalized impact forces with Lagrange equations. The switching criteria among contact, detachment, stick and slip were built to realize the global dynamic automatic switching of the system. Examples were numerically simulated to analyze the complex non-smooth phenomena including forward/backward slip and slip/stick( micro slip) to verify the effectiveness of the proposed model and algorithms.

 

Most parts of wind loads on a large span cable-stayed bridge are caused by cables. It is very important to determine wind forces on cables exactly for bridge design and analysis. Through wind tunnel tests, aerodynamic forces of 8 types of cables with different surface roughness were measured, the influences of surface roughness on Reynolds number effect and aerodynamic force characteristics were studied, the maximum wind load on a cable of a practical bridge was calculated. Results showed that the cable surface roughness has obvious effects on aerodynamic force; with increase in the cable surface roughness, Reynolds number effect decreases gradually; for different surface roughness of a cable, wind velocities corresponding to the maximum wind loads are different; for real bridge design, the maximum wind loads of cables should be determined according to their surface roughness. 


 

Dynamic characteristics of an incompressible spatial-axisymmetric fluid-saturated porous thermo-elastic cylinder subjected to a surface temperature loading were studied in case of local thermal equilibrium. Firstly, the mathematical model of the problem was established based on de Boer porous media theory. Then, the differential quadrature method, the second-order backward difference scheme and Newton-Raphson iterative method were synthetically used to solve the mathematical model and obtain the numerical results of the unknown quantities at every discretized point, and the dynamic characteristics of the cylinder were further studied. In order to verify the validity of the proposed method, the dynamic consolidation problem of an incompressible fluid-saturated porous elastic cylinder was computed with this method. The obtained numerical results agreed well with the analytical ones published by de Boer et.al, it was shown that the proposed method has two advantages: smaller amount of computation and higher accuracy. Finally, the dynamic characteristics of a spatial-axisymmetric fluid-saturated porous thermo-elastic cylinder subjected to mechanical load or both mechanical load and temperature load were studied and compared, the effects of material’s some parameters on the dynamic characteristics of the cylinder were investigated.

 

In the traditional bearingless permanent magnet slice motor (BPMSM), radial displacement’s closed loop control was used to realize indirectly the stable control of suspension force. If the motor’s rotor is subjected to radial interferences, the accuracy of suspension force control and its dynamic performance are restricted. Besides, the phase information needed for suspension force control relies on accurate measurement of rotor position to make the control system be more complex. In order to solve the problems mentioned above, here the relationship between suspension force changes and flux linkage changes of suspension force winding in a BPMSM was deduced. Then the double-closed loop compensation control strategy for radial suspension force and radial displacement was proposed. The flux linkage of torque winding was identified based on the voltage-current model to improve greatly the motor control flexibility. The simulation and test results showed that the proposed suspension force control method can improve the control accuracy and dynamic performance of suspension force; the control system has a strong anti-disturbance ability, good static and dynamic performances.

 

In multi-crack detection of ultrasonic guided wave, defects are not easy to recognize because the amplitudes of defect echoes are small and the waveforms are complex. Here, an approach for detection of steel pipes was proposed based on phase trajectories of a Duffing chaotic system. Firstly, the principle of using a chaotic system’s phase trajectories to recognize ultrasonic guided wave signals was introduced, the parameters of the chaotic system were determined based on phase trajectories recognition combined with ultrasonic guided wave detection. Then, guided waves propagating in pipes were simulated via ANSYS, and the white Gaussian noises were added into the simulated signals to simulate the environmental noise. The test signals were obtained from a six-meter long steel pipe with different sizes of two-crack in it. The simulated and test guided wave signals were detected, respectively with the chosen Duffing chaotic system, they were compared. Finally, the defects’ position was determined with the dichotomy method. The detected phase trajectories results showed that the proposed approach can be applied to effectively recognize defects of two-crack in pipes, and it improves the sensitivity of ultrasonic guided wave detection.

 

Here, a reduced order model (ROM) was developed based on Galerkin method and the proper orthogonal decomposition (POD) with higher effectiveness and wholeness for two-dimensional panel flutter. Firstly, the classic Galerkin method, the traditional POD mode extraction method and the ROM construction method for two-dimensional panel flutter were introduced briefly. Then, in order to simplify the process and improve the efficiency, a new method was proposed to extract POD modes and construct ROM in the modal space spanned with Galerkin basis functions, the equivalence of this method to the traditional POD-ROM method was proved. Furthermore, through the POD modal analysis of typical responses of a panel, it was clarified that the POD modes can effectively reflect the most intrinsic characteristics of the system. Finally, a global ROM for the panel flutter was established based on POD modes in the case of a chaotic response of a panel, and it was employed to study the bifurcation behavior and boundaries of the stable region of the system in detail. The calculation results showed that the accuracy of this POD-ROM is very close to that of Galerkin method, its efficiency, however, is highly improved; the proposed method can be extended to construct ROMs for other complicated dynamic systems.

 

A vibration isolation structure is composed of a superstructure and a vibration isolation layer including vibration isolation bearings and dampers with different damping features from those of superstructure, these dampers possess typically non-proportional damping features. So, the damping matrix of this system can’t be decomposed via the system’s undamped modal shapes and the traditional modal shapes superposition response spectrum method is not applicable to this system. Here, based on the random vibration theory and considering features of vibration isolation structures, a multi-dimensional earthquake complex modal shapes superposition response spectrum method was proposed, it could consider non-proportional damping features. The error of the forced decoupling method, an approximate approach commonly used, was studied. It was found that the energy transfer between the superstructure and the vibration isolation layer is prevented with this method to cause smaller seismic responses of the superstructure. The vibration isolation benchmark model was taken as an example to implement the time-history method, the forced decoupling method and the proposed complex modal shapes superposition response spectrum method, respectively. Three results were compared, it was shown that the proposed method has a better accuracy and can fully reflect the non-proportional damping characteristics in vibration isolation systems; when the damping of the vibration isolation layer is larger, the forced decoupling approach has a worse accuracy, it can’t reflect the amplification effects of the superstructure seismic responses due to damping of the vibration isolation layer.

 

As composite material has a higher damping capacity than metallic materials do, a supercritical rotating composite shaft under the action of material’s internal damping is easier to have an unstable self-excited vibration. Here, based on the basic equations for constitutive relations and strain-displacement relations of composite material, the kinetic energy, the potential energy, and the internal damping dissipative energy of the rotor system including the rotating composite shaft were derived with Bernoulli-Euler beam theory and considering dissipative characteristics of viscoelastic damping. The rotor system’s equations of motion were deduced using Hamilton principle. Galerkin method was used to solve the rotor system’s equations of motion in complex coordinates to derive the characteristic equations of the rotor system. The rotor system’s natural frequency versus rotating speed curve and damping versus rotating speed curve were obtained through numerical analysis. From these curves, the critical rotating speed and instability threshold of the system were gained. The effects of ply angle, stacking sequences, and ratio of length to outer radius on the system’s critical rotating speed and instability threshold were analyzed .The correctness of the dynamic model built here for the rotor system was verified by comparing the calculated results of critical speed and damping of the rotor system with those available in literatures.

 

Fast Kurtogram is one of the most useful methods in fault diagnosis of rolling element bearings. However, in some cases of complex interferences, it cannot exactly recognize the optimal resonance frequency band for envelope demodulation due to that the kurtosis index is too sensitive to impulsive noise. In fact, the envelope spectrum of demodulated signals in frequency domain has a certain immunity ability to noise, the bearing fault characteristic frequency and its harmonics often appear clearly with typical periodic impulse features in the envelope spectrum. Here, the frequency domain correlated kurtosis was proposed to quantitatively describe envelope spectrum amplitudes of narrow-band signals and generate a kurtogram. Simultaneously, the proposed method was applied in the compound fault detection based on the directivity of correlated kurtosis. In addition, two cases of real bearing fault signals were employed to verify the effectiveness and robustness of the proposed method in bearing weak fault diagnosis and compound fault diagnosis.

Assuming the profile cross section of grain grinding grooves is a series of triangles whose vertex angle is equal to the cone angle of grinding particle and the grooves’ depth obeys Rayleigh distribution, considering the effects of grinding particles’ trajectories overlapping, the surface roughness model of hard and brittle materials in axial ultrasonic vibration grinding was established. K9 optical glass was taken as the test material, the fitting accuracies of the model without the overlapping effect of grinding particle trajectories and the one with this effect were analyzed contrastively. The results showed that the latter to characterize the surface roughness of ultrasonic grinding is more close to the test results; the effect trend of grinding parameters and vibration parameters on surface roughness reflected by the latter are consistent with the test results; so, the correctness and effectiveness of the proposed model is verified.

Periodic lattice structures applied in spacecrafts as a new type structure require considering both structural lightweight and excellent sound insulation performance. In order to optimize the traditional pyramidal lattice structure lightweight design and improve its sound insulation effect, the space composition and sound insulation characteristics of pyramidal lattice structures with solid trusses and hollow trusses having the same geometrical parameters were studied here, it was found that the lightweight performance and sound insulation properties of pyramidal lattice structures with hollow trusses in most parts of frequency domain are better than those of pyramidal lattice structures with solid trusses. Based on the direct acoustic-vibration coupling theory, the sound insulation characteristics of pyramidal lattice structures with hollow trusses were studied systematically with numerical methods. The influences of several key structural parameters of hollow truss, such as, elevation angle, wall thickness, length, and cross section shape as well as crystal lattice constant and in-plane size of lattice structure on sound insulation characteristics of the system were analyzed. The results provided a reference for the acoustic optimization of lattice structures.

A non-iterative algorithm for simulating non-Gaussian random process was proposed to avoid the divergence problem in iteration. Firstly, based on the nonlinear translation process, the conversion relationship between latent Gaussian stochastic process and non-Gaussian one was analyzed in detail. Then, it was proved with the reduction to absurdity that the target power spectral density (PSD) and the marginal probability distribution (MPD) function for an arbitrary non-Gaussian random process being compatible is necessary. Thereby, the criterion to judge the compatibleness between the target PSD and the MPD was established (i.e., the target PSD function of the latent Gaussian stochastic process must be a nonnegative function). Furthermore, the modification programs were developed for the case of the target PSD and the MPD being incompatible, a non-iterative algorithm was proposed for simulating a non-Gaussian random process with a single variable. Finally, a non-Gaussian fluctuating wind pressure with different skewness was simulated with the proposed non-iterative algorithm. Its feasibility and validity were verified through comparing correlation functions, PSD, and MPD with their corresponding targets.

The normal contact stiffness model between bearing hole and shaft journal in a radial sliding bearing was built through combining the fractal theory and the joint interface virtual material. Through revising the condition of Weierstrass function’s nondifferentiability at any point, it was proved rigorously that the limited range of fractal dimension is 1≤D<2. Numerical simulation showed that the side face contact coefficient in bearing contact is equal to or less than 1; the side face contact coefficient of inner contact is larger than that of outer contact; when the shaft journal radius and normal contact load increase and joint interface virtual material thickness decreases, the side face contact coefficient in bearing contact increases; the real contact area of inner contact is bigger than that of outer contact; when the shaft journal radius increases and fractal roughness, plane Brinell hardness of bearing hole and joint interface virtual material thickness decrease, the actual contact area increases; when the fractal dimension increases from 1.4 to 1.5, the real contact area increases; when the fractal dimension increases from 1.5 to 1.9, the real contact area decreases; Hertz stress decreases with increase in shaft journal radius; Hertz stress of inner contact is less than that of outer contact; the normal contact stiffness of bearing inner contact is larger than that of outer contact; when the normal contact load, fractal dimension and shaft journal radius increase and fractal roughness, elastic modulus of shaft journal, bearing length and joint interface virtual material thickness decrease, the normal contact stiffness increases.

 

Bearings are important parts in machinery products. Their performance and life are closely related to operating lives of mechanical systems. It is necessary to consider the effects of different failure modes on rolling bearings’ reliability. Here, aiming at the full failure data in bearings’ reliability tests, the prior distributions of bearing life distribution parameters were established with Bootstrap method. Then, the corresponding posterior distributions were estimated using Bayes method. The bearing life distribution parameters were obtained through the posterior expectation reduction. The life distribution model of bearing local failure was gained through further analyzing bearings’ vibration performance degradation data. Copula function was used to analyze comprehensively the life distribution model of bearing full failure and that of bearing local failure. The relative parameters of Copula function were estimated with the experience Kendall relative ranks of test data. Finally, the reliability assessment results of bearings under competing failure were achieved. The results were helpful to find defects in bearing design and improve the bearing reliability.

Collapse accidents of transmission tower-line systems under typhoon occur frequently. Here, the extreme wind speed and wind profile index were simulated using Monte-Carlo method and YanMeng typhoon field, the fluctuating wind of typhoon was calculated using Shiyuan typhoon spectrum. A model consisting of one tower and two-span conductors was established with the FE software ANSYS, and the wind load effect of the model was calculated under typhoon wind load. The analysis indicated that the wind load of conductors and ground wires make a significant contribution to the axial force of principal members, the contribution rate reaches 58.9% during extreme value status of typhoon; the damage due to typhoon can be reduced through appropriately decreasing span distance or dropping lines to protect towers in typhoon region; the dynamic responses of the model during extreme value status of typhoon are much larger than those of the model during the equivalent static wind status due to high strength and high turbulence’s characteristics of typhoon; the adjustment coefficient of wind load in the existing design code seems to be smaller, its dynamic amplification effect must be fully considered under typhoon condition.

 

 

In order to overcome the limitation of the traditional jitter measurement analysis method for TMT (thirty-meter-telescope) tertiary mirror bearing, a mathematical method for the analysis of jitter properties called Gabor transformation was proposed. The location distribution and connection type of muti-accelerometers were introduced, and the measurement process of jitter signals at pitching axis and azimuth axis with accelerometers was fully illustrated. In jitter signal processing, the disadvantages of the time series analysis method and those of the periodogram method were compared. The new method called Gabor transformation was introduced, Gabor transformation could intercept the property of frequency spectrum information at any time interval, the characteristics of a signal was analyzed in time-frequency domain, it made the jitter analysis more real and reliable. At last, the jitter measurement test was performed on a 2m-long telescope, the frequency properties curve within 2s showed that the excitations with lower frequencies close to 0Hz are more sensitive, the noise excitation with 50 Hz always exists, and the high frequency signal with 470 Hz is on and off; when the bearing rotates clockwise, there is an obvious noise signal within 1s; when the bearing rotates counterclockwise, noise signals is significant within 1.5s-2s; lower frequency noise signals still exist after denoising, and higher frequency ones are excited, the realness of the signal is guaranteed. Gabor transformation provided a reference for the future bearing jitter detection of large photo-electric telescopes.