Search

State-of-the-art deep learning models remain data-intensive, requiring a large training dataset to ensure the generalization ability. However, it is quite expensive or impractical to obtain massive training samples for condition monitoring practitioners. This paper proposes a simulation-driven domain adaptation method to circumvent the data deficiency issue using physical-based simulations. A bearing phenomenological model is developed to generate simulated vibration signals. In the frame of domain adaptation transfer learning, a Domain Adversarial Neural Network (DANN) is proposed utilizing the simulated data as the source domain. The DANN can align the coarse supervised source domain data and the fine supervised target domain data to conduct adversarial training. Experimental results indicate that the proposed method can reach high classification accuracy using a small amount of real data. Compared to non-adapted and other transfer learning models, the proposed method demonstrates superior performance for bearing fault diagnosis, which is promising for the industry.

The cyclic spectral coherence analysis has been proved to be a useful tool to reveal the hidden periodic or repetitive patterns, and how to determine an optimal demodulation band is crucial for characterizing the fault characteristic of the rolling element bearing (REB). Thus, a proportional selection scheme based on a bottom-up strategy is proposed in this study, which can determine an optimal demodulation band and realize bearing fault feature extraction under the conditions of poor frequency resolution and a lower signal-to-noise ratio (SNR). Firstly, compared with conventional frequency division structures, such as the 1/3-binary tree structure, the finer frequency band groups with the same bandwidth in each division level can be obtained automatically and strictly under poor frequency resolution conditions. Secondly, the richness of the REB fault information of each divided frequency band can be evaluated by the improved diagnostic feature indicator under a lower SNR condition. By using the scheme, the fault characteristic of the REB can be exposed. The efficacy of the proposed scheme is supported by simulations and experiments under some tough conditions.

Deep learning techniques have recently brought many improvements in the field of neural network training, especially for prognosis and health management. The success of such an intelligent health assessment model depends not only on the availability of labeled historical data but also on the careful samples selection. However, in real operating systems such as induction machines, which generally have a long reliable life, storing the entire operation history, including deterioration (i.e. bearings), will be very expensive and difficult to feed accurately into the training model. Other alternatives sequentially store samples that hold degradation patterns similar to real ones in damage behavior by imposing an accelerated deterioration. Labels lack and differences in distributions caused by the imposed deterioration will ultimately discriminate the training model and limit its knowledge capacity. In an attempt to overcome these drawbacks, a novel sequence-by-sequence deep learning algorithm able to expand the generalization capacity by transferring obtained knowledge from life cycles of similar systems is proposed. The new algorithm aims to determine health status by involving long short-term memory neural network as a primary component of adaptive learning to extract both health stage and health index inferences. Experimental validations are performed using the PRONOSTIA bearing degradation datasets.

International audience; The monitoring of rolling element bearing is indexed as a critical task for condition-based maintenance in various industrial applications. It allows avoiding unscheduled maintenance operations while decreasing their cost. For this purpose, various methodologies were developed to ensure accurate and efficient monitoring. In this context, this paper proposes an approach for bearing faults early diagnosis based on the variational mode decomposition (VMD), used as a notch filter for dominant mode cancellation, and a machine learning approach, namely the one-dimensional convolution neural network (1D-CNN), for detection and diagnosis purposes. Specifically, the proposed approach first performs features extraction using VMD for fault detection, and then triggers to multi-scale features extraction using CNN convolution and pooling layers for classification and diagnosis. The proposed bearing faults detection and diagnosis approach is evaluated, in terms of robustness and performances, using the well-known Case Western Reserve University experimental dataset. In addition, performances are evaluated versus well- established demodulation techniques, in terms of faults detection, and machine learning strategies, in terms of faults diagnosis. The achieved results show that the proposed VMD notch filter-based 1D-CNN approach is clearly promising for bearing degradations monitoring.

Fault diagnosis with formal languages can be performed in an interpretable way. However, the traditional formal languages cannot deal with noisy environments. Additionally, finding the optimal formal language for fault diagnosis is still a challenge due to the sparse reward issue. This article presents a novel method to find formal languages, written with signal spectral logic (SSL), to describe the fault behaviors among frequency domain for fault diagnosis. The formal language defined by SSL is robust to noise, acts as the fault diagnoser, and provides interpretabilities for human operators. Moreover, the fault diagnoser construction procedure has been formulated as a language generation process and an adversarial training technique is used to find the optimal formal language and avoid sparse reward issue existing in language generation problems. Some experiments with real rolling element bearing data and simulated signals demonstrate that our method is able to find formal languages to diagnose faults efficiently and accurately under noisy environments.

Rolling bearing is widely used in rotary machine; the unexpected failure affects the normal operation of the machine leads to production and financial losses to the industry. Hence, Rolling Element Bearing (REB) fault diagnosis is crucial. In this paper, raw bearing vibration signal is divided into the various intrinsic mode functions (IMFs) using Empirical Mode Decomposition (EMD). Statistical features are extracted from the first three IMFs of each conditions of the bearing. These features were given as inputs to K-NN classifier to classify different condition of the bearing.

Mechanical rotating components such as bearing and gear are widely used in various industrial occasions. Fault diagnosis of rotating component can guarantee operational reliability and reduce maintenance cost of system. Although many fault diagnosis methods have been developed based on artificial intelligence methods, most of them ignore the existing signal processing knowledge in rotating element fault diagnosis domain. Meanwhile, these methods just assume that working conditions such as load or rotating speed keep constant, which may cause low accuracy once operating condition changes. To address these problems, an enhanced intelligent fault diagnosis method for rotating component is proposed based on multifeature and convolutional neural network (CNN). The vibration signals are first transformed into angular domain by resampling technique. Then, the angular domain signals are converted to obtain corresponding envelope and squared envelope spectrum features, which are fused into red–green–blue color image form to enhance sample features and enlarge differences among various health states. Finally, a CNN is constructed to accomplish fault recognition. Experimental results show that the average diagnosis accuracies of planetary gearbox and rolling element bearing datasets can reach 96.7% and 95.65%, which demonstrate that the multifeature approach is effective in fault diagnosis of rotating components in scenarios with different rotating speeds.

This paper presents an approach to studying rolling element bearing damage under the interference of impact loading. In the experimental part, a series of bearing tests was performed by using the twin-disc test device with artificially damaged bearings. This was followed by analysis of the measured acceleration response data in impact-free condition as well as under the influence of the impact loading. The results showed successful detection of the bearing outer race damage by using typical bearing damage detection approaches regardless whether the impact loading was applied to the system or not. In turn, recognition of the bearing rolling element damage required specific signal processing. acceptedVersion

As bearing fault signals are nonlinear and nonstationary, which usually submerged in strong background noise, an approach was proposed for the identification of bearing fault based on enhanced cross-correlation spectrum. First, the optimal analysis frequency band was determined according to the spectral Gini index to avoid the blindness selection, which suppressed the interference of irrelevant components in advance and improved the signal-to-noise ratio. The signal fusion technique was employed which fused several optimal filtered signals to make full use of the fault features in each signal, which solved the problem for the proper selection of measuring points. Finally, the ensemble average cross-correlation spectrum was applied to further reduce the influence of background noise, while the bearing fault features could be detected. Both simulation and experimental signal were employed to demonstrate the effectiveness of the proposed approach, the comparative analysis was also employed, results show the effectiveness of this approach while the background noise is suppressed obviously, and the fault information can be easily distinguished.

Aiming at solving the difficulty in extracting early weak fault features of rolling element bearing (REB), a feature extraction method by combing variational mode extraction (VME) with multi-objective information fusion band-pass filter (MIFBF) is proposed. This method is based on the advantage of the VME in filtering out the interference signals and the enhancement effect of the MIFBF on the impact characteristic signals. Firstly, VME is used as the signal preprocessing method to filter out the interference noise. Then, the filtered signal is taken as the target signal, and then MIFBF is used to analyze it to obtain the optimal band-pass filter parameters-center Frequency and bandwidth, thereby constructing the optimal band-pass filter. Finally, further filter the target signal based on the constructed optimal band-pass filter and apply envelope demodulation spectral (EDS) analysis on the filtered signal. At last, satisfactory fault feature extraction effect can be achieved. To obtain key parameters of the optimal band-pass filter, this paper proposes a method based on time and frequency domain fusion indexes to balance the extraction of impact characteristics and cyclostationary characteristics while REB failure occurs. Compared with the existed single index based on time domain or frequency domain, better results could be obtained. Effectiveness and superiority of the proposed method are verified through simulation and experiments.

For the fatigue failure and tribological property of a rolling element bearing, the contact load variation plays a significant role while the most loaded position of the bearing outer raceway takes the greatest risk of failure. This paper focuses on the variation in contact load on the most loaded position of the outer raceway of a gearbox bearing in high-speed train. Under operation conditions of different input speeds and torques, the dynamic contact load distribution in a gearbox bearing of high-speed train was measured by instrumenting the bearing with strain gauges. The most loaded position was identified accordingly and the features and reasons of the variation in contact load on this position were suggested. Three factors were found to have varying degrees of impact on the contact load variation under different gear meshing conditions: modal vibration of the cage or shaft, radial geometrical differences among the rollers and vibration of the gearbox housing.

Objectives: The main objectives of this manuscript are to investigate and diagnose rolling element bearing defects in its inception time. Methods: Vibration signal generated by the induction motor contains a series of frequency components that have rich and viable information about bearing health conditions. Recently, the Maximum Energy Concentration (MEC), measure of time-frequency spectrum has been employed to investigate the small variations in low frequency biomedical signal spectrum. In this paper, the above technique has been modified and applied to study the bearing defects of induction motor using vibration signal and it is termed as Adaptive Modified Morlet Wavelet (AMMW) transform. Initially, this proposed method was validated on two medium frequency synthetic time series signals in terms of MEC measurement at different Signal to Noise Ratio (SNR). Results: The simulated results have depicted that the AMMW method provides excellent timefrequency localization capability over other time-frequency methods like Morlet wavelet transform, modified Morlet wavelet transform, adaptive S-transform and adaptive modified S-transform. Then, this method has been applied to the standard database of vibration signal to determine interquartile power for fault detection purpose and also fault index parameter termed as IFI has been analyzed to detect small variation in vibration signals.

The remaining useful life (RUL) prediction of a rolling element bearing is important for more reasonable maintenance of machinery and equipment. Generally, the information of a failure can hardly be acquired in advance while running and the degradation process varies in terms of different faults. Thus, fault identification is indispensable for a multi-condition RUL prediction, where, however, the fault identification and RUL prediction are separated in most studies. A new hybrid scheme is proposed in this paper for the multi-condition RUL prediction of rolling element bearings. The proposed scheme contains both classification and regression, where the 2D-DCNN based classifier and predictors are built concerning typical fault conditions of a bearing. For the online prediction, the raw signals are spanned in the time-frequency domain and then transferred into images as the input of the scheme. The classifier is used to monitor the vibration of rolling bearings for online fault recognition and excite the corresponding predictor for RUL prediction once a fault is detected. The output from the predictor is amended by the proposed adaptive delay correction method as the final prediction results. A demonstration is performed based on the XJTU-SY datasets and the results are compared with those from the state-of-the-art methods, which proves the superiority of the proposed scheme in improving the accuracy and linearity of RUL prediction. The time cost of the proposed online prediction scheme is also investigated and the results indicate high time effectiveness.

Aiming at the problem that it is difficult to separate and extract the composite fault features of rolling-element bearings, a composite fault diagnosis method combining robust local mean decomposition (RLMD), sparrow search algorithm (SSA), maximum second-order cyclostationarity blind deconvolution (CYCBD), is proposed. First, the RLMD is used to decompose the product function of the signal, and the two indicators, the excess and the correlation coefficient are then used as evaluation criteria to select the appropriate components for reconstruction. The reconstructed signal is then inputted into the SSA-optimized CYCBD algorithm, by specifying the objective function parameter which separates the faults and obtains multiple single fault signals with optimal noise reduction. Finally, envelope demodulation analysis is used for the multiple single fault signals, to obtain the characteristic frequencies of the corresponding faults, so as to complete the fault separation and feature extraction of composite faults. In order to verify the effectiveness of the method, the initial signals and the actual signals generated by the computer shall be used. The algorithm is verified using the XJTU-SY rolling-element bearing dataset, which shows the good performance of the method.

Purpose The purpose of this paper is to reveal the friction and wear performance of grooves textured cylindrical roller thrust bearings with different groove dimensions under starved lubrication. Design/methodology/approach The groove dimensions include: width of grooves (WOG, 50 µm, 100 µm and 150 µm), depth of grooves (DPOG, 7 µm, 11 µm and 15 µm) as well as groove deflection angle (GDA, 45°). A fiber laser marking system was used to prepare groove patterns on the raceways of shaft washers. The friction and wear properties of grooves textured bearings were researched through a vertical universal wear test rig using a customized roller bearing tribo-pair under starved lubrication. Static finite element analyses were conducted to reveal their surface stresses. Through the comprehensive comparison and analyses, the influence mechanism of grooves on the tribological behavior of cylindrical roller thrust bearings was proposed and discussed. Findings When grooves textured bearings run under starved lubrication, their average coefficients of friction (COFs) and wear losses are all significantly reduced and much lower than those of smooth group. The influence of DPOG on the COF curves is significant, while the influence of WOG on the COF curves is a little weak. The influence of groove dimensions on the surface stresses of grooves textured bearings is weak, whether the WOG or DPOG. In this work, when the WOG is 100 µm and the DPOG is 15 µm, its average COF and wear loss are both the lowest, 0.0066 and 0.61 mg, respectively. Compared with the data of smooth group, its friction coefficient is reduced by 75.3% and its mass loss is reduced by 95.8%, showing a significant improvement in this condition. Originality/value This work can provide a valuable reference for the raceway design and reliability optimization of rolling element bearings.

The main objective of the present study is to experimentally investigate and figure out the effect of deep cryogenic treatment in improving dynamic behaviors in terms of damping of a rotating shaft supported by rolling element bearings. An AISI 4140 steel for rotating shafts was selected for the experiments because it is the most widely used material in most industries for a wide range of applications such as machinery components, crankshafts, motor shafts, axle shafts, and railway locomotive traction motor shafts. Untreated, conventionally heat treated, deep cryogenic treated, and deep cryogenic treated and tempered shafts were used for the experiments to observe damping behavior changes of the shafts. Deep cryogenic treated and deep cryogenic treated and tempered shafts were cooled from pre-tempering temperature to - 140 degrees C and held for tempering hold times of 12, 24, 36, and 48 hours. So, ten sets of shafts were employed for the experiment. The vibration data was captured for each of the shafts for five different shaft running speeds 600, 1200, 1800, 2400 and 3000 rpm. The results showed that damping ability of the deep cryogenic treated shaft at a hold time of 36 hours was superior to that of the others shafts. Scientific Research Projects Coordination (BAP) of the University of Duzce [BAP-2015.06.05.351] The authors would like to acknowledge the support of Scientific Research Projects Coordination (BAP) of the University of Duzce for the project entitled BAP-2015.06.05.351. WOS:000686428400008 2-s2.0-85113813843

This research paper presents the analysis of vibration generated due to the race waviness of rolling element bearing which is under external dynamic load. The waviness on races is modeled either by sinusoidal function or superposition of sinusoidal functions of different orders. The mathematical model developed to analyze elastic deformations at race-rollers contacts and nonlinear contact forces for a defect free bearing under static external loading is modified and extended to take dynamic load into account. The radial waviness on races at these contacts is introduced in order to determine un-balanced excitations causing the bearing to vibrate. The set of equations that represent the dynamics of system have been obtained by inputting the excitations to linear vibratory model of bearing system. The vibration response of bearing has been obtained by numerical simulation of state space model representing dynamics of bearing system. The spectra of vibration response are analyzed for different amplitude contribution of dynamic component of external load. The numerical results have been obtained for NJ204 bearing with known values of order and amplitude of waviness. For harmonic loading, the additional spectral components have been observed for each order of waviness on both outer and inner race. The additional spectral components have also been observed for outer race waviness modeled as superposition of different order waviness models, whereas for inner race, there is some amplitude variation in spectral components. The vibration frequencies resulting from various orders of waviness on races have been validated with those of prior researchers.

This article presents the experimental and statistical methodology for localized fault analysis in the rotor-bearing system. These defects on outer race, on inner race, and on a combination of ball and outer race are considered. In this study speed, load and defects were considered as the essential process variables to understand their significance and effects on vibration response for the rotor-bearing system. Three factors at three levels were considered for experimentation, and the experiment was designed for L27 based on design of experiments (DOE) methodology. From the experiments, the vibration response results are recorded in terms of root mean square value for the analysis. Response surface methodology (RSM) is used for identifying the interaction effect of varying process parameters upon the response of vibrations by response surface plot. The rotor-bearing test setup is used for experimentation and is analyzed by using DOE. This study establishes the prediction of fault in the rotor-bearing system in combined parametric effect analysis and its influence with DOE and RSM.

Failure of roller bearings can cause downtime or a complete shutdown of rotating machines. Therefore, a well-timed detection of bearing defects must be performed. Modern condition monitoring demands simple but effective bearing failure diagnosis by integrating dynamic models with intelligence techniques. This paper presents an integration of Dimensional Analysis (DA) and Artificial Neural Network (ANN) to diagnose the size of the bearing faults. The vibration responses of artificially damaged bearings using Electrode Discharge Machining are collected using Fast Fourier Techniques on a developed rotor-bearing test rig. Two-performance indicators, actual error, and performance of error are used to evaluate the accuracy of models. The simplicity of the DA model and the performance of the ANN model predicting with 5.49% actual error and 97.79 performance of error band enhanced the accuracy of diagnosis compared to the experimental results. Moreover, ANN has shown good performance over experimental results and DA.

Transient impulses caused by local faults are critical informative indicators for rolling element bearing fault diagnosis. The methods for accurately extracting transient impulses while suppressing strong background noise and interference components have received extensive studies. In this article, a novel fault diagnosis scheme based on optimized wavelet packet denoising and modulation signal bispectrum is proposed, which takes advantage of the transient impulse enhancement of wavelet packet denoising and the demodulation ability of modulation signal bispectrum to diagnose bearing faults more accurately. First, the measured signals are decomposed into a series of time–frequency subspaces using wavelet packet transform. An optimal threshold value is selected based on the proposed threshold criterion by considering unbiased autocorrelation of envelope and Gini index of the transient impulses. Subsequently, the subspaces are denoised by the wavelet packet denoising with the optimized threshold value, and the master subspaces that containing the fault-related transient impulses are selected based on the Gini index indicator. Finally, the modulation signal bispectrum is utilized to further purify the signal and extract the modulation components contained in the transient impulses, and the suboptimal modulation signal bispectrum slices are selected based on the characteristic frequency intensity coefficient. The modulation signal bispectrum detector is then obtained by averaging the suboptimal modulation signal bispectrum slices to determine the type of the bearing faults. The proposed wavelet packet denoising-modulation signal bispectrum is validated based on the simulation and experimental studies. Compared with the variational mode decomposition and Teager energy operator, fast kurtogram as well as conventional modulation signal bispectrum, the proposed wavelet packet denoising-modulation signal bispectrum method has superior performance in extracting the fault feature of the incipient defects on different bearing components.

The wavelet transform is a state of the art time–frequency analysis method for rolling element bearing localized fault detection, using vibration signals. When these localized faults are present at more than one location of bearing, it is called “multi-fault.” Using wavelet transform fault detection with high severity is possible, but this method fails to detect the presence of fault as well as the location of a fault in multi-fault case and when the fault severity is low. The identification of the fault location, in rolling element bearing when more than one location of bearing contains a localized fault, is very useful for further root cause analysis; therefore, multi-fault detection is a challenge today. In the present work, a new morphological joint time–frequency adaptive kernel–based semi-smart framework is developed to address this challenge. In morphological joint time–frequency adaptive kernel, the kernel will adapt itself by analyzing the basic morphology of the bearing under observation and by considering the location of a fault. The simulation and experimental results show that morphological joint time–frequency adaptive kernel–based framework is able to detect low severity single fault as well as the location of the localized fault on rolling element bearing in the multi-fault case. Experimental results also show that the morphological joint time–frequency adaptive kernel framework is independent of bearing dimensions as well as machine operating conditions.

Rolling Element Bearings are one of the most ubiquitous machine elements used in various machineries in the manufacturing industry. Prognosis and estimation of residual life of rolling element bearing are very important for efficient implementation of health monitoring and condition-based maintenance. In this paper, a rolling element bearing fault or degradation trend prediction is modeled using Gaussian Process Regression (GPR) method. Various vibration features based on signal complexity, namely Shannon entropy, permutation entropy, and approximate entropy are estimated to obtain the bearing degradation trend. When fault or degradation occurs in rolling element bearing, there is a subtle change in the dynamics of the system and subsequently, there are changes in the features extracted from the vibration signal. In this paper, a comparative analysis of various kernel functions of the GPR model is carried out using accuracy-based metrics. In addition, the combination of goodness of metric (monotonicity (Mon), robustness (Rob), and prognosability (Pro)), namely hybrid metric, is proposed to select the efficient bearing degradation trend of features. Further, the GPR at ARD exponential kernel has been employed to make the prognosis of degradation trend in bearings with a 95% confidence interval (CI). The proposed methodology is validated through a mathematical model of the simulated vibration signal. Finally, from the simulated and experimental data, it is demonstrated that the entropy features have better performance than the statistical features.

Transfer learning using Convolution Neural Networks (CNNs) has improved the state of the art results in many research studies. Rolling element bearing fault diagnosis is a domain that has been researched extensively using different data mining and machine learning techniques. In this paper, we prove that deep CNNs, when trained on Bispectrum images of fault signals using transfer learning, provide highly accurate and reliable results for fault diagnosis that are at par with the state of the art results. These transfer learning based models are able to quickly learn patterns from visual features of a vibration signal’s bispectrum, eliminating the need for manual feature extraction. In this paper, four pretrained networks – Alexnet, VGG-19, GoogLeNet, ResNet-50 – have been fine tuned on bispectrum images prepared from vibration signals of machine ball bearing elements. Each network has been trained with 3 optimizers – Stochastic Gradient Descent (SGD), Adaptive Moment Estimation (Adam) and Adamax. These models are able to obtain high classification accuracy within a few epochs. We also visualise and analyze the feature maps associated with intermediate convolution layers for one of these models.

Vibration signals of rolling element bearing’s early weak fault are often submerged by some interference components. To extract early weak fault features accurately, a weak fault feature enhancement method of rolling element bearing based on resonance sparse decompositionand multi-objective information frequency band selection is proposed. This method makes full use of resonance sparse decomposition in filtering the interferences and multi-objective information frequency band selection in enhancing impulsive and cyclostationary features of rolling element bearing’s early weak fault simultaneously. First, resonance sparse decomposition is used as the preprocessing program of multi-objective information frequency band selection to filter the interferences (such as rotating frequency with its harmonics) of rolling element bearing’s early weak fault characteristic components. Then, the filtered components containing main fault information are analyzed by MIFBS to establish the best band-pass filter. Finally, the envelope spectrum analysis is applied on the filtered vibration signal, and the fault characteristic frequency with its harmonics is extracted. To achieve the optimal output parameters of multi-objective information frequency band selection, fusion indexes based on time- and frequency-domain estimators are proposed and used to balance the enhancement of impulsive and cyclostationary characteristics of rolling element bearing’s early weak fault signals. Compared with most of the existing methods mainly based on single time-domain estimators or frequency-domain estimators to improve performance spectral kurtosis, the proposed fusion indexes overcome their defects and could enhance the impulsive and cyclostationary features of rolling element bearing’s early weak fault simultaneously. Effectiveness of the proposed method is verified through simulation and experiment. Besides, its advantage over the other related methods is also presented.

Inspired by previous achievements, different time-domain features for diagnosis of rolling element bearings are investigated in this study. An experimental test rig is prepared for condition monitoring of angular contact bearing by using an acoustic emission sensor for this purpose. The acoustic emission signals are acquired from defective bearing, and the sensor takes signals from defects on the inner or outer race of the bearing. By studying the literature works, different domains of features are classified, and the most common time-domain features are selected for condition monitoring. The considered features are calculated for obtained signals with different loadings, speeds, and sizes of defects on the inner and outer race of the bearing. Our results indicate that the clearance, sixth central moment, impulse, kurtosis, and crest factors are appropriate features for diagnosis purposes. Moreover, our results show that the clearance factor for small defects and sixth central moment for large defects are promising for defect diagnosis on rolling element bearings.

Conventional signal processing techniques for fault detection are usually aimed at constant speed conditions. Nowadays, order tracking, especially tacho‐less order tracking is regarded as a valuable tool for extracting fault features under variable rotational speed conditions. Therefore, a teager energy operator (TEO)‐based modified Laplacian of Gaussian filter is proposed to enhance the fault detection behaviour of tacho‐less order tracking. This method consists of four steps: First, multi‐synchrosqueezing transform is employed to estimate and then extract the instantaneous rotation frequency of a shaft. Second, based on the extracted curve, the non‐stationary domain signal is converted into a quasi‐stationary domain by the re‐sampling technique. Third, the obtained quasi‐stationary domain signal is de‐noise by the novel method. Finally, fault characteristic orders are extracted via envelope order spectrum analysis method and the performance is significantly improved. The results of numerical simulation and experimental investigations are performed to validate the superiority of the novel method for fault extraction under variable rotational speed conditions.

Maintenance planning plays a critical role in the process industry, where any unplanned maintenance may lead to a significant loss. Condition monitoring happens to aid maintenance planning and has become an inherent part of the maintenance activity. Physical parameters such as vibration, acoustic emission, current, etc., are used for condition monitoring, out of which vibration is the most preferred parameter and is widely used in the industry. Vibration data is measured near to bearings, which themselves are monitored for their condition, and hence rolling element bearing (REB) is the focus of this study. REBs are monitored for the presence of a fault in them as well as for their severity. Fault diagnosis of REB using harmonic product spectrum (HPS) is proposed in this study. The proposed methodology's novelty lies in the signal pre-processing step, whose output is fed to the HPS method, which is used for defective raceway identification. The efficacy of HPS is assessed with parameter optimized Variational mode decomposition (VMD) and classical bandpass filtering method as pre-processors. It is observed that the HPS delivers better diagnostic results with the VMD method than the bandpass filtering method. Non-dominated sorting particle swarm optimization algorithm is deployed for parameter optimization of VMD. HPS combined with VMD as pre-processor forms an autonomous HPS(AHPS) algorithm, whose input is measured signal and output is defect frequency. The process is so designed that a raw signal, when fed to the algorithm, delivers the result as identification of a defective raceway. Unlike previously developed methods, the proposed method needs no manual intervention. Results obtained from simulated signals and signals recorded through experiments validate that the proposed methodology can be used effectively for fault diagnosis of REB.

In the past decades, data-driven methods for the machinery fault diagnosis problem have been developed successfully, especially for the tasks where the training data and the testing data are from the same distribution. In the real industrial scenarios, because of the diversity of the practical factors, the training data and the testing data are generally from different distributions which leads to data distribution discrepancy. Most existing well-established methods basically cannot well address this problem. In this paper, a new multi-layer domain adversarial learning strategy is proposed for transfer learning. Adversarial training in multiple layers is implemented to achieve domain fusion under varying operating conditions. The experiments on the real-world rolling element bearing dataset are carried out for validation, and promising testing accuracies is achieved in different tasks, which are higher than the other popular methods. The experimental results verify the validity of the proposed method on the problem of the cross-domain fault diagnosis, and the applicability in the real industrial scenarios.

Purpose Fault diagnosis methods based on blind source separation (BSS) for rolling element bearings are necessary tools to prevent any unexpected accidents. In the field application, the actual signal acquisition is usually hindered by certain restrictions, such as the limited number of signal channels. The purpose of this study is to fulfill the weakness of the existed BSS method. Design/methodology/approach To deal with this problem, this paper proposes a blind source extraction (BSE) method for bearing fault diagnosis based on empirical mode decomposition (EMD) and temporal correlation. First, a single-channel undetermined BSS problem is transformed into a determined BSS problem using the EMD algorithm. Then, the desired fault signal is extracted from selected intrinsic mode functions with a multi-shift correlation method. Findings Experimental results prove the extracted fault signal can be easily identified through the envelope spectrum. The application of the proposed method is validated using simulated signals and rolling element bearing signals of the train axle. Originality/value This paper proposes an underdetermined BSE method based on the EMD and the temporal correlation method for rolling element bearings. A simulated signal and two bearing fault signal from the train rolling element bearings show that the proposed method can well extract the bearing fault signal. Note that the proposed method can extract the periodic fault signal for bearing fault diagnosis. Thus, it should be helpful in the diagnosis of other rotating machinery, such as gears or blades.

The most frequent faults in rotating electrical machines occur in their rolling element bearings. Thus, an effective health diagnosis mechanism of rolling element bearings is necessary from operational and economical points of view. Recently, convolutional neural networks (CNNs) have been proposed for bearing fault detection and identification. However, two major drawbacks of these models are (a) their lack of ability to capture global information about the input vector and to derive knowledge about the statistical properties of the latter and (b) the high demand for computational resources. In this paper, short time Fourier transform (STFT) is proposed as a pre-processing step to acquire time-frequency representation vibration images from raw data in variable healthy or faulty conditions. To diagnose and classify the vibration images, the image classification transformer (ICT), inspired from the transformers used for natural language processing, has been suitably adapted to work as an image classifier trained in a supervised manner and is also proposed as an alternative method to CNNs. Simulation results on a famous and well-established rolling element bearing fault detection benchmark show the effectiveness of the proposed method, which achieved 98.3% accuracy (on the test dataset) while requiring substantially fewer computational resources to be trained compared to the CNN approach.