Based on the approach that ‘replace curved by straight’, the steady combined force is introduced directly into the fluid structure interaction vibration differential equation of straight pipe to describe straight-curved one’s transverse motion. Taking clamped-elastically supported combined pipe as an example, the new transfer matrix based on Laplace transform is used to derive the system’s characteristic equation calculating its natural frequency, and then the vibration characteristics such as natural frequency and critical velocity are studied. During this process, influences of the steady-state combined tension, flow model modification factor, and system’s components etc. on the vibration characteristics are investigated. According to the above investigation, the ‘fake coupled-mode divergence’ is firstly put forward, it can be concluded that different steady-state combined combined tension may lead to different critical velocity, change of system’s components may lead to distinguishing judgement for stability. The vibration differential equation is also established based on the approach ‘replacing straight by curved’, results of the above two thoughts are verified to be the same. The above investigation can provide insights for studying vibration characteristics of other types of pipes and behaviors of other fluid structure interaction mechanics as well, and be of high guiding meanings for theory and values for practice.

A two-stage high-static-low-dynamic stiffness vibration isolation system composed of Euler buckling beam negative stiffness regulator and two-stage linear vibration isolation system in parallel is tested and studied. The mechanical principle of high-static-low-dynamic stiffness is described, and the compression test of the Euler buckling beam negative stiffness regulator prototype is carried out to verify its negative stiffness mechanism. According to the different parallel forms of the negative stiffness regulators, two vibration isolation systems, constrained and unconstrained, are proposed. The dynamic equations of the negative vibration isolation model of the system are solved, and vibration isolation performance of the two types of vibration isolation systems with the different upper and lower stiffness are analyzed in combination. Two vibration isolation test systems of high-static-low-dynamic stiffness vibration isolation systems are built, and their vibration isolation performance is verified by sweep frequency and fixed frequency tests, and the reasons for the deviation of the results are analyzed.

In modern daily life, people are often exposed to many types of vibrations generated by machine operations, traffic and other human activities. These vibrations can cause annoyance to residents, and even interfere the operations of precision instruments. Generally, these adverse effects of vibrations can be eliminated or prevented by installation of various types of wave barriers, such as multi-row of holes. In the paper, the investigation is focused on effects of using multi-row of holes for the reduction of nearby vibration response generated by dynamic machine foundation laid on saturated soil. Two semi-analytical BEM models are established to calculate the environmental vibrations due to the machine foundation and the vibration isolation efficiency by multi-row of holes, respectively. In order to increase calculating efficiency of the semi-analytical BEM, a simple SPMD parallel algorithm is developed using Matlab software. The SPMD parallel algorithm is optimized by using the corresponding relationship between holes on the spatial coordinates. By using the optimized SPMD parallel algorithm, the effects of the model parameters on effectiveness of vibration isolation are calculated and discussed in detail. The results show that the optimized SPMD parallel algorithm of semi-analytical BEM is much faster than the chained program dramatically. Multi-row of holes can isolate the ground vibrations successfully, and the holes layout has less effect on the screening efficiency. Increasing the radius and the number of holes in a row, decreasing the net spacing between two successive holes can all lead to an increase in the screening effectiveness, while the rows of holes and the net spacing between two successive rows have less effect on the screening effectiveness. Further, the distance between the rigid foundation and wave barriers can has a limited impact on vibration mitigation effectiveness. According to the results, it’s suggested in the design that two-row of holes is recommended and the hole depth, radius and the nest spacing between two successive holes should take the values of 1.0, 0.15 and 0.1, respectively.

Seismic research technologies of power systems focus on the design, analysis and disaster mitigation before earthquakes.To quickly assist the emergency work after earthquakes, this paper proposed a post-earthquake evaluation method facing porcelain cylindrical equipment that uses monitoring data to predict structural stress responses. This method establishes a stress response proxy model by integrating machine learning and swarm intelligence evolution technologies, then builds refined simulation model, and conducts response analyses to form structural response database. Based on this, the proxy model can be trained and evaluated.Once the structural responses can be monitored, the proxy model can supply the stress response rapidly after earthquakes to help the post-disaster detection. A case study was performed using 1100 kV transformer bushing, and the evaluation models were vali-dated by shaking table tests and theoretical model based on distributed parameter system. The results indicate that using acceleration monitoring data can accurately evaluate the base stress of porcelain cylindrical equipment. Particle swarm optimization can efficiently adjust the internal structures of evaluation models, further increasing the model accuracy. The accuracies of evaluation models were validated by both shaking table tests and theoretical model.

In this study, a novel semi-supervised fault diagnosis of planetary gearboxes based on manifold regularized support higher-order tensor machines (MRSHTM) is proposed. In the MRSHTM, CANDECOMP/PARAFAC (CP) decomposition is introduced to exploit the intrinsic structural information of tensor data, and tensor-based inverse multiquadric kernel function (Tensor-IMKF) is defined to construct a Laplacian operator. The constructed graph matrix can better describe the manifold structure between tensor data. Besides, the one-versus-rest (OVR) strategy is introduced into the MRSHTM model for multi-class fault diagnosis of planetary gearboxes. Hierarchical multiscale permutation entropy (HMPE) is adopted to extract the three-order tensor features “channel×hierarchical layer×scale”, and then the extracted HMPE values are fed into OVR-MRSHTM for automatic fault identification. The results suggest that the proposed method can achieve semi-supervised fault diagnosis of planetary gearboxes in tensor space.

Deep learning methods have shown great potential in the field of fault diagnosis of train wheelset bearings, but their effective implementation is based on the correct matching between various types of data and category labels. For data with a small number of label error samples, traditional deep learning methods are difficult to achieve the expected diagnostic effect. To address this issue, this paper proposes a fault diagnosis method combining box graph method and feature fusion model is proposed to address this issue. In this method, the outlier in each group of bearing signals is removed by box graph method, and the remaining data is expanded by the SMOTE method to restore to the original data size; Input the processed sample data into the improved feature fusion model for fault identification and classification. The experimental data of train wheel bearings was used for validation. The results showed that compared to directly using traditional neural network models for fault diagnosis, the diagnostic accuracy of the method proposed in this paper is higher, indicating that the method has better processing performance for bearing data with a small number of label error samples.

Aiming at the problems that the existing convolutional neural network cannot fully extract the correlation features between rolling bearing time domain signals, the large number of samples required for model training and the insufficient generalization, A new method for diagnosing multi-condition faults of rolling bearings based on an enhanced convolutional neural network model is proposed. The length of the bearing single-revolution fault characteristic signal is calculated according to the rolling bearing speed and sampling frequency, then the complete information of the single-revolution time domain signal is encoded by Gramian Angular Difference Field coding technology to generate the corresponding feature image, enabling the neural network can visually learn the time domain signal correlation features. The 7×7 deep convolutional layer of the ConvNeXt model is reconstructed by using the asymmetric convolution in the ACNet network model: that is, two 3×3, one 1×3 and one 3×1 asymmetric small convolution kernel are used to reconstruct the 7×7 convolutional layer in the form of a multi-branch structure combination, which enhances the feature extraction efficiency of the ConvNeXt model. The data augmentation module and learning rate decay strategy of the ConvNeXt model are improved to raise the generalization of the ConvNeX model under small-sample training, to build an enhanced deep convolutional neural network model IConvNeXt. Different fault diameters of Case Western Reserve University, composite rolling bearing faults of Southeast University and variable speed bearing fault data sets of Ottawa, Canada are used for experimental verification, the results show that the proposed IConvNeXt model achieves a fault diagnosis rate of 100% for different fault diameters and composite faults of rolling bearings, and a fault diagnosis rate of 99.63% for variable speed bearings. The proposed method is experimentally compared with RP+ResNet, RP+ IConvNeXt, time-frequency graph+DCNN, MLCNN-LSTM, MTF+ IConvNeXt and other methods, the results were condicted to validate that the fault diagnosis effect of the proposed model is better than that of other methods under less sample training and has strong generalization performance.

Aiming at the complex mechanism and operation characteristics of the mechanical system of diesel engine on RO-RO passenger ship, a multi-component nonlinear vibration signal decomposition method and vibration characteristic parameter extraction strategy are proposed. The signal is decomposed into intrinsic mode function (IMF) components via empirical mode decomposition and restrained the end effects in empirical mode decomposition (EMD) by symmetrical extension. The sensitive component is extracted by the correlation. The characteristic frequencies are matched according to the amplitude demodulation spectrum via 4 order high energy operator (4th-HEO). The proposed method is illustrated by the three-dimensional vibration signals under 0 load and 100 load operation conditions of the RO-RO passenger ship. The results show that the analysis performance of vertical signals are better than other axial signals. And for the 9-cylinder 4-stroke diesel engine, the amplitudes of 0.5 times and 4.5 times are very prominent, and are more sensitive to load changes.

Aiming at the problem that the existing damage identification methods are difficult to track the structural damage in real time and require a large amount of calculation, a model order reduction and online damage identification method based on the combination of recursive proper orthogonal decomposition (RPOD) and strong tracking extended Kalman filter (STEKF) is proposed.The structural damage identification under dynamic load is studied. The RPOD method is used to update online and construct the reduced-order model reflecting the structure state in real time, which solves the problem of large calculation and difficult convergence of dynamic analysis of multi-degree of freedom structures under unknown loads. Meanwhile, the evolution of damage is tracked and located. The STEKF method is used to track the state vector of the reduced-order model and identify the reduced-order model parameters degraded by damage. The feasibility of the proposed method is verified by numerical simulation of a six-story shear frame and model test of a three-story steel frame. The results show that the proposed method can accurately construct the reduced-order model and track the time-varying history of the reduced-order model parameters. Meanwhile, it can effectively identify the location and extent of the damage of the shear building structure, even when dealing with high levels of noise, it retains high accuracy.

In order to reveal the damage distribution law of self-centering braced structure under simultaneous and successive actions of earthquake and wind, the damage of beams and columns and global damage of a 50-story steel frame self-centering braced tube structure are analyzed under earthquakes with different intensities and wind with a return period of fifty years. The results indicate that under earthquake alone and under simultaneous action of earthquake and wind, beams in braced tube are damaged earlier than those in exterior frame, and the damage of columns in braced tube develops faster than those in exterior frame. The simultaneous action of earthquake and wind increases the damage degree of columns and increases the maximum damage value of columns at the 31st story by 78.6%. As earthquake intensity increases, the amplification influence of the action of wind on structural global damage increases gradually. Under successive action of earthquake with different intensities and wind with a return period of fifty years, the action of wind increases the average value of maximum residual deformation ratio of seismic damaged structure. When the peak ground acceleration is 10 m/s², the average value of maximum residual deformation ratio of seismic damaged structure is increased from 0.325% to 0.330% caused by wind load. However, wind load has less influence on the damage state of the most severely damaged member in seismic damaged structure.