Deep Spatiotemporal Convolutional-Neural-Network-Based Remaining Useful Life Estimation of Bearings

The remaining useful life(RUL)estimation of bearings is critical for ensuring the reliability of mechanical systems.Owing to the rapid development of deep learning methods,a multitude of data-driven RUL estimation approaches have been proposed recently.However,the following problems remain in existing methods:1)Most network models use raw data or statistical features as input,which renders it difficult to extract complex fault-related information hidden in signals;2)for current observations,the dependence between current states is emphasized,but their complex dependence on previous states is often disregarded;3)the output of neural networks is directly used as the estimated RUL in most studies,resulting in extremely volatile prediction results that lack robustness.Hence,a novel prognostics approach is proposed based on a time-frequency representation(TFR)subsequence,three-dimensional convolutional neural network(3DCNN),and Gaussian process regression(GPR).The approach primarily comprises two aspects:construction of a health indicator(HI)using the TFR-subsequence-3DCNN model,and RUL estimation based on the GPR model.The raw signals of the bearings are converted into TFR-subsequences by continuous wavelet transform and a dislocated overlapping strategy.Subsequently,the 3DCNN is applied to extract the hidden spatiotemporal features from the TFR-subsequences and construct HIs.Finally,the RUL of the bearings is estimated using the GPR model,which can also define the probability distribution of the potential function and prediction confidence.Experiments on the PRONOSTIA platform demonstrate the superiority of the proposed TFR-subsequence-3DCNN-GPR approach.The use of degradation-related spatiotemporal features in signals is proposed herein to achieve a highly accurate bearing RUL prediction with uncertainty quantification.

Nonlinear dynamic responses of sandwich functionally graded porous cylindrical shells embedded in elastic media under 1:1 internal resonance

In this article, the nonlinear dynamic responses of sandwich functionally graded(FG) porous cylindrical shell embedded in elastic media are investigated. The shell studied here consists of three layers, of which the outer and inner skins are made of solid metal, while the core is FG porous metal foam. Partial differential equations are derived by utilizing the improved Donnell's nonlinear shell theory and Hamilton's principle. Afterwards, the Galerkin method is used to transform the governing equations into nonlinear ordinary differential equations, and an approximate analytical solution is obtained by using the multiple scales method. The effects of various system parameters,specifically, the radial load, core thickness, foam type, foam coefficient, structure damping,and Winkler-Pasternak foundation parameters on nonlinear internal resonance of the sandwich FG porous thin shells are evaluated.

Short-time Fourier Transform Using Odd Symmetric Window Function

In this paper,a novel time–frequency(TF)analysis method,called the short-time Fourier transform using odd symmetric window function(OSTFT),is proposed using odd symmetric window function to replace the conventional even window function of STFT.Different from conventional STFT acquiring the amplitude maximum at time and frequency centers,OSTFT acquires the minimum amplitude of 0.Hence,OSTFT can obtain a TF representation(TFR)with high TF resolution by utilizing the leaked energy rather than restraining it.It is worth to mention that the proposed OSTFT can vitiate the effect of window size we choose on the TFR obtained.Furthermore,it also has a good performance on signals with complex instantaneous frequencies(IFs),even crossing IFs.Because we just replace the conventional window function of STFT,the time-consuming of the proposed OSTFT is at the same level as the conventional STFT.The effectiveness of proposed OSTFT has been validated on two complex multi-component simulated numerical signals and a signal collected from the brown bat.

Vibration Characteristics of Rotor System with Loose Disc Caused by the Insufficient Interference Force

The rotating parts looseness is one of the common failures in rotating machinery.The current researches of looseness fault mainly focus on non-rotating components.However,the looseness fault of disc-shaft system,which is the main work part in the rotor system,is almost ignored.Here,a dynamic model of the rotor system with loose disc caused by the insufficient interference force is proposed based on the contact model of disc-shaft system with the microscopic surface topography,the vibration characteristics of the system are analyzed and discussed by the number simulation,and verified by the experiment.The results show that the speed of the shaft,the contact stiffness,the clearance between the disc and shaft,the damping of the disc and the rotational damping have an influence on the rotation state of the disc.When the rotation speed of the disc and the shaft are same,the collision frequency is mainly composed of one frequency multiplication component and very weak high frequency multiplication components.When the rotation speed of the disc and the shaft is close,the vibration of the disc occurs a beat vibration phenomenon in the horizontal direction.Simultaneously,a periodical similar beat vibration phenomenon also occurs in the waveform of the disc-shaft displacement difference.The collision frequency is mainly composed of a low frequency and a weak high frequency component.When the rotation speed of the disc and the shaft has great difference,the collision frequency is mainly composed of one frequency multiplication,a few weak high frequency multiplication components and a few low frequency multiplication component.With the reduction of the relative speed of the disc,the trajectory of the disc changes from circle-shape to inner eight-shape,and then to circle-shape.In the inner eight-shape,the inner ring first gradually becomes smaller and then gradually becomes larger,and the outer ring is still getting smaller.The obtained research results in this paper has important theoretical value for the diagnosis of the rotor system with the loose disc.

Multiple internal resonances of rotating composite cylindrical shells under varying temperature fields

Composite cylindrical shells,as key components,are widely employed in large rotating machines.However,due to the frequency bifurcations and dense frequency spectra caused by rotation,the nonlinear vibration usually has the behavior of complex multiple internal resonances.In addition,the varying temperature fields make the responses of the system further difficult to obtain.Therefore,the multiple internal resonances of composite cylindrical shells with porosities induced by rotation with varying temperature fields are studied in this paper.Three different types of the temperature fields,the Coriolis forces,and the centrifugal force are considered here.The Hamilton principle and the modified Donnell nonlinear shell theory are used to obtain the equilibrium equations of the system,which are transformed into the ordinary differential equations(ODEs)by the multi-mode Galerkin technique.Thereafter,the pseudo-arclength continuation method,which can identify the regions of instability,is introduced to obtain the numerical results.The detailed parametric analysis of the rotating composite shells is performed.Multiple internal resonances caused by the interaction between backward and forward wave modes and the energy transfer phenomenon are detected.Besides,the nonlinear amplitude-frequency response curves are different under different temperature fields.

Preface

Diagnostics and condition monitoring are aspects of a field of engineering that aims to manage the health of engineering machinery and structures,but which has increasing applicability to many other complex systems in the world from the medical care of people,through the reliability of supply chain logistics to the secure operation of a large and complex organisation.The management of health generally requires a number of sequential steps,which are usually as follows:the detection of any abnormality in the procedural operation of the system,the location of the abnormal behaviour within the system,an assessment of the severity of the abnormality and identification of the potential system vulnerabilities that result from it,a detailed diagnosis of the problem including the identification of its root cause,and finally a predictive assessment of the future prospects for the continued operation of the system,including any remedial action that should be taken to minimise impact and maximise continued operational performance.

意义建构理论视角下谦卑领导对团队创新绩效的作用机制研究

知识经济时代,信息越来越分散,领导者如何激励团队成员进行主动性的知识创造,对于促进知识传播和创新具有重要意义。作为一种“自下而上”的领导风格,谦卑领导成为团队创新绩效的重要影响因素。基于意义建构理论,本研究构建了谦卑领导通过外部知识获取/内部知识分享和团队学习行为影响团队创新绩效的双链式中介模型,并通过64个团队配对样本进行实证分析。结果表明,谦卑领导对团队外部知识获取和团队内部知识分享均有显著正向影响,团队知识分享/团队知识获取和团队学习行为在二者关系中起到双链式中介传导作用。研究结果为知识经济时代团队领导更广泛地看待知识管理现象,促进团队知识潜力发展,并最终提升团队创新绩效有一定的理论意义和实践启发。

Nonlinear characteristic investigation of magnetorheological damper-rotor system with local nonlinearity

Magnetorheological(MR)dampers show superior performance in reducing rotor vibration,but their high nonlinearity will cause nonsynchronous response,resulting in fatigue and instability of rotors.Herein,we are devoted to the investigation of the nonlinear characteristics of MR damper mounted on a flexible rotor.First,Reynolds equations with bilinear constitutive equations of MR fluid are employed to derive nonlinear oil film forces.Then,the Finite Element(FE)model of rotor system is developed,where the local nonlinear support forces produced by MR damper and its coupling effects with the rotor are considered.A hybrid numerical method is proposed to solve the nonlinear FE motion equations of the MR damper-rotor system.To validate the proposed model,a rotor test bench with two dual-coil MR dampers is constructed,upon which experimental studies on the dynamic characteristics of MR damper-rotor system are carried out.The effects of different system parameters,including rotational speed,excitation current and amount of unbalance,on nonlinear dynamic behaviors of MR damper-rotor system are evaluated.The results show that the system may appear chaos,jumping,and other complex nonlinear phenomena,and the level of the nonlinearity can be effectively alleviated by applying suitable excitation current and oil supply pressure.

Dynamic, thermal, and vibrational analysis of ball bearings with over-skidding behavior

The term“over-skidding”indicates that the cage rotational speed ratio exceeds the theoretical value as ball purely rolls on the raceway.Different from the skidding phenomenon that occurs in low-load and high-speed bearing,over-skidding usually occurs in large-size angular contact bearings,and it is still difficult to suppress under high load conditions.The main forms of damage to the raceway by over-skidding are spinning and gyro slip.To further explore the vibration characteristics and thermal effects of this phenomenon,a set of over-skidding tests of an angular contact bearing with a bore diameter of 220 mm were conducted on an industrial-size test bench.Through the experiment,the influence of axial load,rotational speed,and lubrication conditions on the occurrence of over-skidding were determined.Based on a previous dynamics model,the heat generation and thermal network models were integrated in the present study to predict the over-skidding and its thermal behavior.The model was validated in terms of the measured degree of over-skidding and temperature rise.The results showed that the degree of over-skidding reaches up to 12%of the theoretical value,and the friction power loss of the ball-pocket accounts for 30%of the total power loss.The analysis of the vibration signal showed a strong correlation between the bearing vibration characteristics and over-skidding behavior,thereby providing a way to indirectly measure the degree of over-skidding.

Dynamic characteristics of large permanent magnet direct‐drive generators considering electromagnetic–structural coupling effects

Dynamic characteristics of large permanent magnet direct‐drive generators(PMDGs)considering electromagnetic–structural coupling effects are analyzed in this study.Using the conformal mapping method,the scalar magnetic potential of the air gap magnetic field considering the slot effect is calculated.On the basis of the discrete current element and magnetic equivalent circuit model,the local magnetic saturation effect of the stator and rotor is quantitatively simulated and the air gap magnetic field intensity distribution is obtained via numerical simulation.A series of uniformly distributed equivalent electromagnetic springs are introduced to develop an electromagnetic–structural coupling finite element PMDG model.The proposed air gap field analysis method is verified by the finite element analysis results.On the basis of the test platform for the Goldwind 1.5MW PMDG,both modal and dynamic response tests for the stator/rotor coupling system are conducted,and the results are compared with the natural frequencies,mode shapes,and vibration responses obtained using the numerical model.The effects of the air gap length and rotor speed on the natural frequencies of the coupling system are analyzed.The proposed model has the potential to accurately evaluate the PMDG vibration energy,avoiding resonance points,and maintaining stable operations of the unit.

Power fluctuation and power loss of wind turbines due to wind shear and tower shadow

The magnitude and stability of power output are two key indices of wind turbines. This study investigates the effects of wind shear and tower shadow on power output in terms of power fluctuation and power loss to estimate the capacity and quality of the power generated by a wind turbine. First, wind speed models, particularly the wind shear model and the tower shadow model, are described in detail. The widely accepted tower shadow model is modified in view of the cone-shaped towers of modem large-scale wind turbines. Power fluctuation and power loss due to wind shear and tower shadow are analyzed by performing theoretical calculations and case analysis within the framework of a modified version of blade element momentum theory. Results indicate that power fluctuation is mainly caused by tower shadow, whereas power loss is primarily induced by wind shear. Under steady wind conditions, power loss can be divided into wind farm loss and rotor loss. Wind farm loss is constant at 3a（3a- 1）R^2/（8H^2）. By contrast, rotor loss is strongly influenced by the wind turbine control strategies and wind speed. That is, when the wind speed is measured in a region where a variable-speed controller works, the rotor loss stabilizes around zero, but when the wind speed is measured in a region where the blade pitch controller works, the rotor loss increases as the wind speed intensifies. The results of this study can serve as a reference for accurate power estimation and strategy development to mitigate the fluctuations in aerodynamic loads and power output due to wind shear and tower shadow.

Fault feature extraction of planet gear in wind turbine gearbox based on spectral kurtosis and time wavelet energy spectrum

Planetary transmission plays a vital role in wind turbine drivetrains, and its fault diagnosis has been an important and challenging issue. Owing to the complicated and coupled vibration source, time-variant vibration transfer path, and heavy background noise masking effect, the vibration signal of planet gear in wind turbine gearboxes exhibits several unique characteristics： Complex frequency components, low signal-to-noise ratio, and weak fault feature. In this sense, the periodic impulsive components induced by a localized defect are hard to extract, and the fault detection of planet gear in wind turbines remains to be a challenging research work. Aiming to extract the fault feature of planet gear effectively, we propose a novel feature extraction method based on spectral kurtosis and time wavelet energy spectrum （SK-TWES） in the paper. Firstly, the spectral kurtosis （SK） and kurtogram of raw vibration signals are computed and exploited to select the optimal filtering parameter for the subsequent band-pass filtering. Then, the band-pass filtering is applied to extrude periodic transient impulses using the optimal frequency band in which the corresponding SK value is maximal. Finally, the time wavelet energy spectrum analysis is performed on the filtered signal, selecting Morlet wavelet as the mother wavelet which possesses a high similarity to the impulsive components. The experimental signals collected from the wind turbine gearbox test rig demonstrate that the proposed method is effective at the feature extraction and fault diagnosis for the planet gear with a localized defect.

Application of stress wave theory for pyroshock isolation at spacecraft-rocket interface

A large number of pyroshock devices are employed in spacecraft and rockets to realize stage separation and appendage deployment.Release of pyroshock devices induces high-level transient shock responses which tend to cause fatal damages in electronic equipment made of crystals and brittle materials.This paper aims to provide methods to isolate pyroshock and guarantee the safety of such equipment against high-frequency shocks.Firstly,stress wave transmission mechanism in stepped rods is investigated,upon which optimal area rate for shock isolation is achieved.Then,two spacecraft-rocket interface structures for pyroshock isolation,namely isolation hole and interim segment,are proposed.Both numerical simulations and experiments are carried out to validate the two shock isolation strategies.It is revealed that the interim segment structure shows better pyroshock isolation performance at the cost of increasing the weight of launching system whereas isolation hole is an optimal choice to reduce pyroshock response without causing weight increase.

Deep convolutional tree-inspired network:a decision-tree-structured neural network for hierarchical fault diagnosis of bearings

The fault diagnosis of bearings is crucial in ensuring the reliability of rotating machinery.Deep neural networks have provided unprecedented opportunities to condition monitoring from a new perspective due to the powerful ability in learning fault-related knowledge.However,the inexplicability and low generalization ability of fault diagnosis models still bar them from the application.To address this issue,this paper explores a decision-tree-structured neural network,that is,the deep convolutional tree-inspired network(DCTN),for the hierarchical fault diagnosis of bearings.The proposed model effectively integrates the advantages of convolutional neural network(CNN)and decision tree methods by rebuilding the output decision layer of CNN according to the hierarchical structural characteristics of the decision tree,which is by no means a simple combination of the two models.The proposed DCTN model has unique advantages in 1)the hierarchical structure that can support more accuracy and comprehensive fault diagnosis,2)the better interpretability of the model output with hierarchical decision making,and 3)more powerful generalization capabilities for the samples across fault severities.The multiclass fault diagnosis case and cross-severity fault diagnosis case are executed on a multicondition aeronautical bearing test rig.Experimental results can fully demonstrate the feasibility and superiority of the proposed method.

Special issue： Wind turbine dynamic modeling, condition monitoring and diagnosis

As one of the typical renewable energy sources, wind energy has experienced an immense growth with respect to turbine size and market share, and this growth has led to a rapid development in wind-powered equipment in the last 10 years. In 2016, China was the world＇s largest regional market for new wind power development, with a capacity addition of 23.3 GW, which almost equaled the global cumulative capacity in 2001.