Bearing fault diagnosis based on a multiple-constraint modal-invariant graph convolutional fusion network

Multisensor data fusionmethod can improve the accuracy of bearing fault diagnosis,in order to address the problems of single-sensor data types and the insufficient exploration of redundancy and complementarity between different modal data in most existing multisensor data fusion methods for bearing fault diagnosis,a bearing fault diagnosis method based on a Multiple-Constraint Modal-Invariant Graph Convolutional Fusion Network(MCMI-GCFN)is proposed in this paper.Firstly,a Convolutional Autoencoder(CAE)and Squeeze-and-Excitation Block(SE block)are used to extract features of raw current and vibration signals.Secondly,the model introduces source domain classifiers and domain discriminators to capture modal invariance between different modal data based on domain adversarial training,making use of the redundancy and complementarity between multimodal data.Then,the spatial aggregation property of Graph Convolutional Neural Networks(GCN)is utilized to capture the dependency relationship between current and vibration modes with similar time step features for accurately fusing contextual semantic information.Finally,the validation is conducted on the public bearing damage current and vibration dataset from Paderborn University.The experimental results showed that the delivered fusion method achieved a bearing fault diagnosis accuracy of 99.6%,which was about 9%–11.4%better than that with nonfusion methods.

Enhanced Detection of Rolling Element Bearing Fault Based on Stochastic Resonance

Early bearing faults can generate a series of weak impacts. All the influence factors in measurement may degrade the vibration signal. Currently, bearing fault enhanced detection method based on stochastic resonance（SR） is implemented by expensive computation and demands high sampling rate, which requires high quality software and hardware for fault diagnosis. In order to extract bearing characteristic frequencies component, SR normalized scale transform procedures are presented and a circuit module is designed based on parameter-tuning bistable SR. In the simulation test, discrete and analog sinusoidal signals under heavy noise are enhanced by SR normalized scale transform and circuit module respectively. Two bearing fault enhanced detection strategies are proposed. One is realized by pure computation with normalized scale transform for sampled vibration signal, and the other is carried out by designed SR hardware with circuit module for analog vibration signal directly. The first strategy is flexible for discrete signal processing, and the second strategy demands much lower sampling frequency and less computational cost. The application results of the two strategies on bearing inner race fault detection of a test rig show that the local signal to noise ratio of the characteristic components obtained by the proposed methods are enhanced by about 50% compared with the band pass envelope analysis for the bearing with weaker fault. In addition, helicopter transmission bearing fault detection validates the effectiveness of the enhanced detection strategy with hardware. The combination of SR normalized scale transform and circuit module can meet the need of different application fields or conditions, thus providing a practical scheme for enhanced detection of bearing fault.

Stochastic resonance of coupled time-delayed system with fluctuation of mass and frequency and its application in bearing fault diagnosis

The stochastic resonance behavior of coupled stochastic resonance(SR)system with time-delay under mass and frequency fluctuations was studied.Firstly,the approximate system model of the time-delay system was obtained by the theory of small time-delay approximation.Then,the random average method and Shapiro-Loginov algorithm were used to calculate the output amplitude ratio of the two subsystems.The simulation analysis shows that increasing the time-delay and the input signal amplitude appropriately can improve the output response of the system.Finally,the system is applied to bearing fault diagnosis and compared with the stochastic resonance system with random mass and random frequency.The experimental results show that the coupled SR system taking into account the actual effect of time-delay and couple can more effectively extract the frequency of the fault signal,and thus realizing the diagnosis of the fault signal,which has important engineering application value.

Attention mechanism based multi-scale feature extraction of bearing fault diagnosis

Effective bearing fault diagnosis is vital for the safe and reliable operation of rotating machinery.In practical applications,bearings often work at various rotational speeds as well as load conditions.Yet,the bearing fault diagnosis under multiple conditions is a new subject,which needs to be further explored.Therefore,a multi-scale deep belief network(DBN)method integrated with attention mechanism is proposed for the purpose of extracting the multi-scale core features from vibration signals,containing four primary steps:preprocessing of multi-scale data,feature extraction,feature fusion,and fault classification.The key novelties include multi-scale feature extraction using multi-scale DBN algorithm,and feature fusion using attention mecha-nism.The benchmark dataset from University of Ottawa is applied to validate the effectiveness as well as advantages of this method.Furthermore,the aforementioned method is compared with four classical fault diagnosis methods reported in the literature,and the comparison results show that our pro-posed method has higher diagnostic accuracy and better robustness.

Morphology Similarity Distance for Bearing Fault Diagnosis Based on Multi-Scale Permutation Entropy

Bearings are crucial components in rotating machines,which have direct effects on industrial productivity and safety.To fast and accurately identify the operating condition of bearings,a novel method based on multi⁃scale permutation entropy(MPE)and morphology similarity distance(MSD)is proposed in this paper.Firstly,the MPE values of the original signals were calculated to characterize the complexity in different scales and they constructed feature vectors after normalization.Then,the MSD was employed to measure the distance among test samples from different fault types and the reference samples,and achieved classification with the minimum MSD.Finally,the proposed method was verified with two experiments concerning artificially seeded damage bearings and run⁃to⁃failure bearings,respectively.Different categories were considered for the two experiments and high classification accuracies were obtained.The experimental results indicate that the proposed method is effective and feasible in bearing fault diagnosis.

Multi-scale morphology analysis of acoustic emission signal and quantitative diagnosis for bearing fault

Monitoring of potential bearing faults in operation is of critical importance to safe operation of high speed trains.One of the major challenges is how to differentiate relevant signals to operational conditions of bearings from noises emitted from the surrounding environment.In this work,we report a procedure for analyzing acoustic emission signals collected from rolling bearings for diagnosis of bearing health conditions by examining their morphological pattern spectrum（MPS） through a multi-scale morphology analysis procedure.The results show that acoustic emission signals resulted from a given type of bearing faults share rather similar MPS curves.Further examinations in terms of sample entropy and Lempel-Ziv complexity of MPS curves suggest that these two parameters can be utilized to determine damage modes.

WDBM: Weighted Deep Forest Model Based Bearing Fault Diagnosis Method

In the research field of bearing fault diagnosis,classical deep learning models have the problems of too many parameters and high computing cost.In addition,the classical deep learning models are not effective in the scenario of small data.In recent years,deep forest is proposed,which has less hyper parameters and adaptive depth of deep model.In addition,weighted deep forest(WDF)is proposed to further improve deep forest by assigning weights for decisions trees based on the accuracy of each decision tree.In this paper,weighted deep forest model-based bearing fault diagnosis method(WDBM)is proposed.The WDBM is regard as a novel bearing fault diagnosis method,which not only inherits the WDF’s advantages-strong robustness,good generalization,less parameters,faster convergence speed and so on,but also realizes effective diagnosis with high precision and low cost under the condition of small samples.To verify the performance of the WDBM,experiments are carried out on Case Western Reserve University bearing data set(CWRU).Experiments results demonstrate that WDBM can achieve comparative recognition accuracy,with less computational overhead and faster convergence speed.

Bearing Fault Diagnosis Method of Wind Turbine Based on Improved Anti-Noise Residual Shrinkage Network

Aiming at the difficulty of rolling bearing fault diagnosis of wind turbine under noise environment,a new bearing fault identification method based on the Improved Anti-noise Residual Shrinkage Network(IADRSN)is proposed.Firstly,the vibration signals of wind turbine rolling bearings were preprocessed to obtain data samples divided into training and test sets.Then,a bearing fault diagnosis model based on the improved anti-noise residual shrinkage network was established.To improve the ability of fault feature extraction of the model,the convolution layer in the deep residual shrinkage network was replaced with a Dense-Net layer.To further improve the anti-noise ability of the model,the first layer of the model was set as the Drop-block layer.Finally,the labeled data samples were used for training model and the trained model was applied to the test set to output the fault diagnosis results.The results showed that the proposed method could achieve the fault diagnosis of wind turbine bearing more accurately in the high noise environment through comparison and verification.

CVRgram for Demodulation Band Determination in Bearing Fault Diagnosis under Strong Gear Interference

Fault-related resonance frequency band extraction-based demodulation methods are widely used for bearing diagnostics.However,due to the high peaks of strong gear meshing interference,the classical band selection methods have poor performance and cannot work well for bearing fault type detection.As such,the CVRgram-based bearing fault diagnosis method is proposed in this paper.In the proposed method,inspired by the conditional variance(CV)index and root mean square(RMS),a novel index,named the CV/root mean square(CVR),is first proposed.The CVR index has high robustness for the interference of non-Gaussian or Gaussian noise and has the ability to determine the center frequency of the weak bearing fault-related resonance frequency band under strong interference.Secondly,motived by the Kurtogram,the CVRgram algorithm is developed for adaptively determining the optimal filtering parameters.Finally,the CVRgram-based bearing fault diagnosis method under strong gear meshing interference is proposed.The performance of the CVRgram-based method is verified by both the simulation signal and the experiment signal.The comparison analysis with the Kurtogram,Protrugram,and CVgram-based method shows that the proposed technique has a much better ability for bearing fault detection under strong noise interference.

Adaptive Variational Mode Decomposition for Bearing Fault Detection

Rolling element bearings are commonly used in rotary mechanical and electrical equipment. According to investigation, more than half of rotating machinery defects are related to bearing faults. However, reliable bearing fault detection still remains a challenging task, especially in industrial applications. The objective of this work is to propose an adaptive variational mode decomposition (AVMD) technique for non-stationary signal analysis and bearing fault detection. The AVMD includes several steps in processing: 1) Signal characteristics are analyzed to determine the signal center frequency and the related parameters. 2) The ensemble-kurtosis index is suggested to decompose the target signal and select the most representative intrinsic mode functions (IMFs). 3) The envelope spectrum analysis is performed using the selected IMFs to identify the characteristic features for bearing fault detection. The effectiveness of the proposed AVMD technique is examined by experimental tests under different bearing conditions, with the comparison of other related bearing fault techniques.

Bearing Fault Diagnosis Based on Graph Formulation and Graph Convolutional Network

Bearing fault diagnosis stands as a critical component in the maintenance of rotating machinery.Many prevalent deep learning techniques are tailored to Euclidean datasets such as audio,image,and video.However,these methods falter when confronting non-Euclidean datasets,notably graph representations.In response,here we introduce an innovative approach harnessing the graph convolutional network(GCN)to analyze graph data derived from vibration signals related to bearing faults.This enhances the precision and reliability of fault diagnosis.Our methodology initiates by deriving a periodogram from the unprocessed vibration signals.Subsequently,this periodogram is mapped into a graph format,upon which the GCN is engaged for classification purposes.We substantiate the efficacy of our approach through rigorous experimental assessments conducted on a collection of ten bearing sets.Within these experiments,an accelerometer chronicles vibration signals across varying load conditions.We probe into the diagnostic accuracy rates across diverse loads and signal-to-noise ratios.Furthermore,a comparative evaluation of our method against several established algorithms delineated in this study is undertaken.Empirical observations confirm that our GCN-based strategy registers an elevated diagnostic accuracy quotient.

Optimizing Bearing Fault Detection:CNN-LSTM with Attentive TabNet for Electric Motor Systems

Electric motor-driven systems are core components across industries,yet they’re susceptible to bearing faults.Manual fault diagnosis poses safety risks and economic instability,necessitating an automated approach.This study proposes FTCNNLSTM(Fine-Tuned TabNet Convolutional Neural Network Long Short-Term Memory),an algorithm combining Convolutional Neural Networks,Long Short-Term Memory Networks,and Attentive Interpretable Tabular Learning.The model preprocesses the CWRU(Case Western Reserve University)bearing dataset using segmentation,normalization,feature scaling,and label encoding.Its architecture comprises multiple 1D Convolutional layers,batch normalization,max-pooling,and LSTM blocks with dropout,followed by batch normalization,dense layers,and appropriate activation and loss functions.Fine-tuning techniques prevent over-fitting.Evaluations were conducted on 10 fault classes from the CWRU dataset.FTCNNLSTM was benchmarked against four approaches:CNN,LSTM,CNN-LSTM with random forest,and CNN-LSTM with gradient boosting,all using 460 instances.The FTCNNLSTM model,augmented with TabNet,achieved 96%accuracy,outperforming other methods.This establishes it as a reliable and effective approach for automating bearing fault detection in electric motor-driven systems.

Fault Diagnosis Method of Rolling Bearing Based on MSCNN-LSTM

Deep neural networks have been widely applied to bearing fault diagnosis systems and achieved impressive success recently.To address the problem that the insufficient fault feature extraction ability of traditional fault diagnosis methods results in poor diagnosis effect under variable load and noise interference scenarios,a rolling bearing fault diagnosis model combining Multi-Scale Convolutional Neural Network(MSCNN)and Long Short-Term Memory(LSTM)fused with attention mechanism is proposed.To adaptively extract the essential spatial feature information of various sizes,the model creates a multi-scale feature extraction module using the convolutional neural network(CNN)learning process.The learning capacity of LSTM for time information sequence is then used to extract the vibration signal’s temporal feature information.Two parallel large and small convolutional kernels teach the system spatial local features.LSTM gathers temporal global features to thoroughly and painstakingly mine the vibration signal’s characteristics,thus enhancing model generalization.Lastly,bearing fault diagnosis is accomplished by using the SoftMax classifier.The experiment outcomes demonstrate that the model can derive fault properties entirely from the initial vibration signal.It can retain good diagnostic accuracy under variable load and noise interference and has strong generalization compared to other fault diagnosis models.

Quantitative Diagnosis of Fault Severity Trend of Rolling Element Bearings

The condition monitoring and fault diagnosis of rolling element bearings are particularly crucial in rotating mechanical applications in industry. A bearing fault signal contains information not only about fault condition and fault type but also the severity of the fault. This means fault severity quantitative analysis is one of most active and valid ways to realize proper maintenance decision. Aiming at the deficiency of the research in bearing single point pitting fault quantitative diagnosis, a new back-propagation neural network method based on wavelet packet decomposition coefficient entropy is proposed. The three levels of wavelet packet coefficient entropy（WPCE） is introduced as a characteristic input vector to the BPNN. Compared with the wavelet packet decomposition energy ratio input vector, WPCE shows more sensitive in distinguishing from the different fault severity degree of the measured signal. The engineering application results show that the quantitative trend fault diagnosis is realized in the different fault degree of the single point bearing pitting fault. The breakthrough attempt from quantitative to qualitative on the pattern recognition of rolling element bearings fault diagnosis is realized.

Connected Components-based Colour Image Representations of Vibrations for a Two-stage Fault Diagnosis of Roller Bearings Using Convolutional Neural Networks

Roller bearing failure is one of the most common faults in rotating machines.Various techniques for bearing fault diagnosis based on faults feature extraction have been proposed.But feature extraction from fault signals requires expert prior information and human labour.Recently,deep learning algorithms have been applied extensively in the condition monitoring of rotating machines to learn features automatically from the input data.Given its robust performance in image recognition,the convolutional neural network(CNN)architecture has been widely used to learn automatically discriminative features from vibration images and classify health conditions.This paper proposes and evaluates a two-stage method RGBVI-CNN for roller bearings fault diagnosis.The first stage in the proposed method is to generate the RGB vibration images(RGBVIs)from the input vibration signals.To begin this process,first,the 1-D vibration signals were converted to 2-D grayscale vibration Images.Once the conversion was completed,the regions of interest(ROI)were found in the converted 2-D grayscale vibration images.Finally,to produce vibration images with more discriminative characteristics,an algorithm was applied to the 2-D grayscale vibration images to produce connected components-based RGB vibration images(RGBVIs)with sets of colours and texture features.In the second stage,with these RGBVIs a CNN-based architecture was employed to learn automatically features from the RGBVIs and to classify bearing health conditions.Two cases of fault classification of rolling element bearings are used to validate the proposed method.Experimental results of this investigation demonstrate that RGBVI-CNN can generate advantageous health condition features from bearing vibration signals and classify the health conditions under different working loads with high accuracy.Moreover,several classification models trained using RGBVI-CNN offered high performance in the testing results of the overall classification accuracy,precision,recall,and F-score.

Bearing Fault Numerical Model for the Closed-slot Rotor Submersible Motor

A numerical model of the bearing fault of a motor with a closed-slot rotor using the finite element method(FEM)is proposed.The rotor’s radial motion can be regarded as static eccentric at the defect time points and healthy at other time points.The frequency of the harmonic component is analyzed corresponding to bearing fault in stator current according to the radial movement of the motor shaft.Moreover,the relative permeability variation region is established to achieve the radial motion of the rotor with bearing fault.Firstly,the relative permeability variation region is established in the health and static eccentric models.Then,the defect time points are estimated and the static eccentricity model by transient field is analyzed.Finally,the relative permeability of the variable region in the static eccentric model is imported into the variable region of the health model at the defect time points.The simulation results show that the air gap flux density of the bearing fault model is different from that of the health model and static eccentric models.In addition,the stator current contains harmonic components of the bearing fault.The analysis results prove the applicability of the proposed model.

A Comprehensive 3-Steps Methodology for Vibration-Based Fault Detection,Diagnosis and Localization in Rotating Machines

In any industry,it is the requirement to know whether the machine is healthy or not to operate machine further.If the machine is not healthy then what is the fault in the machine and then finally its location.The paper is proposing a 3-Steps methodology for the machine fault diagnosis to meet the industrial requirements to aid the maintenance activity.The Step-1 identifies whether machine is healthy or faulty,then Step-2 detect the type of defect and finally its location in Step-3.This method is extended further from the earlier study on the 2-Steps method for the rotor defects only to the 3-Steps methodology to both rotor and bearing defects.The method uses the optimised vibration parameters and a simple Artificial Neural Network(ANN)-based Machine Learning(ML)model from the earlier studies.The model is initially developed,tested and validated on an experimental rotating rig operating at a speed above 1st critical speed.The proposed method and model are then further validated at 2 different operating speeds,one below 1st critical speed and other above 2nd critical speed.The machine dynamics are expected to be significantly different at these speeds.This highlights the robustness of the proposed 3-Steps method.

PARAMETERS OPTIMIZATION OF CONTINUOUS WAVELET TRANSFORM AND ITS APPLICATION IN ACOUSTIC EMISSION SIGNAL ANALYSIS OF ROLLING BEARING

Morlet wavelet is suitable to extract the impulse components of mechanical fault signals. And thus its continuous wavelet transform （CWT） has been successfully used in the field of fault diagnosis. The principle of scale selection in CWT is discussed. Based on genetic algorithm, an optimization strategy for the waveform parameters of the mother wavelet is proposed with wavelet entropy as the optimization target. Based on the optimized waveform parameters, the wavelet scalogram is used to analyze the simulated acoustic emission （AE） signal and real AE signal of rolling bearing. The results indicate that the proposed method is useful and efficient to improve the quality of CWT.

Analysis of weak signal detection based on tri-stable system under Levy noise

Stochastic resonance system is an effective method to extract weak signal.However,system output is directly influenced by system parameters.Aiming at this,the Levy noise is combined with a tri-stable stochastic resonance system.The average signal-to-noise ratio gain is regarded as an index to measure the stochastic resonance phenomenon.The characteristics of tri-stable stochastic resonance under Levy noise is analyzed in depth.First,the method of generating Levy noise,the effect of tri-stable system parameters on the potential function and corresponding potential force are presented in detail.Then,the effects of tri-stable system parameters w,a,b,and Levy noise intensity amplification factor D on the resonant output can be explored with different Levy noises.Finally,the tri-stable stochastic resonance system is applied to the bearing fault detection.Simulation results show that the stochastic resonance phenomenon can be induced by tuning the system parameters w,a,and b under different distributions of Levy noise,then the weak signal can be detected.The parameter intervals which can induce stochastic resonances are approximately equal.Moreover,by adjusting the intensity amplification factor D of Levy noise,the stochastic resonances can happen similarly.In bearing fault detection,the detection effect of the tri-stable stochastic resonance system is superior to the bistable stochastic resonance system.

Research and application of stochastic resonance in quad-stable potential system

To solve the problem of low weak signal enhancement performance in the quad-stable system,a new quad-stable potential stochastic resonance(QSR)is proposed.Firstly,under the condition of adiabatic approximation theory,the stationary probability distribution(SPD),the mean first passage time(MFPT),the work(W),and the power spectrum amplification factor(SAF)are derived,and the impacts of system parameters on them are also extensively analyzed.Secondly,numerical simulations are performed to compare QSR with the classical Tri-stable stochastic resonance(CTSR)by using the genetic algorithm(GA)and the fourth-order Runge–Kutta algorithm.It shows that the signal-to-noise ratio(SNR)and mean signal-to-noise increase(MSNRI)of QSR are higher than CTSR,which indicates that QSR has superior noise immunity than CTSR.Finally,the two systems are applied in the detection of real bearing faults.The experimental results show that QSR is superior to CTSR,which provides a better theoretical significance and reference value for practical engineering application.