Showing papers on "Bearing (mechanical) published in 2021"
TL;DR: A new method is put forward that fuses multi-modal sensor signals, i.e. the data collected by an accelerometer and a microphone, to realize more accurate and robust bearing-fault diagnosis.
199 citations
TL;DR: The analysis results demonstrated that the proposed hybrid deep learning model can achieve higher detection accuracy than CNN and gcForest, which may be favorable to practical applications.
119 citations
TL;DR: It has been concluded that infrared thermography can be used in a non-contact way to automatically identify the faults that help to detect early warnings, irrespective of speeds and hence ensures reduced system shutdowns causing by bearing failure.
116 citations
TL;DR: The experimental results show that the proposed gated recurrent unit neural network with dual attention gates can effectively predict the RULs of rolling bearings, and it has higher prediction accuracy and convergence speed than the conventional prediction methods.
Abstract: In the mechatronic system, rolling bearing is a frequently used mechanical part, and its failure may result in serious accident and major economic loss. Therefore, the remaining useful life (RUL) prediction of rolling bearing is greatly indispensable. To accurately predict the RUL of the rolling bearing, a new kind of gated recurrent unit neural network with dual attention gates, namely, gated dual attention unit (GDAU), is proposed. With the acquired life-cycle vibration data of a rolling bearing, a series of root mean squares at different time instants are calculated as the health indicator (HI) vector. Next, the to-be HI sequence is predicted by GDAU according to the existing HI vector, and then the RUL of the rolling bearing is estimated. The experimental results show that the proposed GDAU can effectively predict the RULs of rolling bearings, and it has higher prediction accuracy and convergence speed than the conventional prediction methods.
108 citations
TL;DR: In this article, a data-driven approach for condition monitoring of generator bearing using temporal temperature data is presented, where four algorithms, the support vector regression machine, neural network, extreme learning machine, and the deep belief network are applied to model the bearing behavior.
Abstract: Wind turbines are widely installed as the new source of cleaner energy production. Dynamic and random stress imposed on the generator bearing of a wind turbine may lead to overheating and failure. In this paper, a data-driven approach for condition monitoring of generator bearings using temporal temperature data is presented. Four algorithms, the support vector regression machine, neural network, extreme learning machine, and the deep belief network are applied to model the bearing behavior. Comparative analysis of the models has demonstrated that the deep belief network is most accurate. It has been observed that the bearing failure is preceded by a change in the prediction error of bearing temperature. An exponentially-weighted moving average (EWMA) control chart is deployed to trend the error. Then a binary vector containing the abnormal errors and the normal residuals are generated for classifying failures. LS-SVM based classification models are developed to classify the fault bearings and the normal ones. The proposed approach has been validated with the data collected from 11 wind turbines.
106 citations
TL;DR: An emergent two dimensional discrete wavelet transform (2D-DWT) based IRT method has been proposed in this article for diagnosing the different bearing faults in IM, namely, inner and outer race defects, and lack of lubrication.
Abstract: Bearing is one of the most crucial parts in induction motor (IM) as a result there is a constant call for effective diagnosis of bearing faults for reliable operation. Infrared thermography (IRT) is appreciably used as a non-destructive and non-contact method to detect the bearing defects in a rotary machine. However, its performance is limited by insignificant information and string noise present in the infrared thermal image. To address this issue, an emergent two dimensional discrete wavelet transform (2D-DWT) based IRT method has been proposed in this article for diagnosing the different bearing faults in IM, namely, inner and outer race defects, and lack of lubrication. The dimensionality of the extracted features was reduced using principal component analysis (PCA) and thereafter the selected features were ranked in the order of most relevant features using the Mahalanobis distance (MD) method to achieve the optimal feature set. Finally these selected features have been passed to the complex decision tree (CDT), linear discriminant analysis (LDA) and support vector machine (SVM) for fault classification and performance evaluation. The classification results reveal that the SVM outperformed CDT and LDA. The proposed strategy can be used for self-adaptive recognition of bearing faults in IM which helps to avoid the unplanned and unwanted system shutdowns due to the bearing failure.
104 citations
TL;DR: The comparison results with the MSR show that the CMSR can obtain a higher output SNR, which is more beneficial to extract weak signal features and realize fault detection, and this method also has practical application value for engineering rotating machinery.
Abstract: In recent years, methods for detecting motor bearing faults have attracted increasing attention. However, it is very difficult to detect the faults from weak motor bearing signals under the strong noise. Stochastic resonance (SR) is a popular signal processing method, which can process weak signals with the noise, but the traditional SR is burdensome in determining its parameters. Therefore, in this paper, a new advancing coupled multi-stable stochastic resonance method, with two first-order multi-stable stochastic resonance systems, namely CMSR, is proposed to detect motor bearing faults. Firstly, the effects of the output signal-to-noise ratio (SNR) for system parameters and coupling coefficients are analyzed in-depth by numerical simulation technology. Then, the SNR is considered as the fitness function for the seeker optimization algorithm (SOA), which can adaptively optimize and determine the system parameters of the SR by using the subsampling technique. An advancing coupled multi-stable stochastic resonance method is realized, and the pre-processed signal is input into the CMSR to detect the faults of motor bearings by using Fourier transform. The faults of motor bearings are determined according to the output signal. Finally, the actual vibration data of induction motor bearings are used to prove the effectiveness of the proposed CMSR. The comparison results with the MSR show that the CMSR can obtain a higher output SNR, which is more beneficial to extract weak signal features and realize fault detection. At the same time, this method also has practical application value for engineering rotating machinery.
90 citations
TL;DR: In this paper, a new bearing fault diagnosis method combining singular value decomposition (SVD) and the squared envelope spectrum (SES) is proposed for early-stage defects of rolling element bearings (REBs) using vibration signals.
83 citations
TL;DR: In this article, the micro interface lubrication mechanisms of water-lubricated bearing are revealed and they apply to bearings with different dimensions and they lay solid foundation for further optimal design of surface topography of such bearings.
73 citations
TL;DR: In this article, a fault diagnosis method based on generalized composite multiscale weighted permutation entropy (GCMWPE), supervised Isomap (S-Iso), and marine predators algorithm-based support vector machine (MPA-SVM) was proposed.
Abstract: The rolling bearing vibration signals are complex, non-linear, and non-stationary, it is difficult to extract the sensitive features and diagnose faults by conventional signal processing methods. This paper focuses on the sensitive features extraction and pattern recognition for rolling bearing fault diagnosis and proposes a novel intelligent fault-diagnosis method based on generalized composite multiscale weighted permutation entropy (GCMWPE), supervised Isomap (S-Iso), and marine predators algorithm-based support vector machine (MPA-SVM). Firstly, a novel non-linear technology named GCMWPE was presented, allowing the extraction of bearing features from multiple scales and enabling the construction of a high-dimensional feature set. The GCMWPE uses the generalized composite coarse-grained structure to overcome the shortcomings of the original structure in multiscale weighted permutation entropy and obtain more stable entropy values. Subsequently, the S-Iso algorithm was introduced to obtain the main features and reduce the GCMWPE set dimensionality. Finally, a combination of GCMWPE and S-Iso set was input to the MPA-SVM for diagnosis and identification. The marine predators algorithm (MPA) was used to obtain the optimal SVM parameters. The effectiveness of the proposed fault diagnosis method was confirmed through two bearing fault diagnosis experiments. The results have shown that the proposed method can be used to correctly diagnose bearing states with high diagnostic accuracy.
66 citations
TL;DR: In this article, a skidding dynamics model of a rolling bearing is established to reveal the skidding characteristics of a bearing and its dynamic performance plays a significant effect on the stability, reliability and even safety of the machine.
TL;DR: In this article, a hybrid model based on extendable useful life (EUL) under continuous monitoring and bearing status classification is proposed Statistical properties of typical time domain features extracted from vibration and acoustic emission are studied Correlations of these parameters with bearing status are reviewed and feasible parameters are evaluated for bearing status quantification.
TL;DR: A high accuracy and high sensitivity have been achieved in the detection and classification of three-body abrasion due to particle contamination and a deep learning approach based on convolutional neural networks was used for multi-class classification into three different wear failure modes, namely running-in, inadequate lubrication and particle-contaminated oil.
TL;DR: This article proposes a strategy that can not only suppress the ZSC, but also eliminate the ripple of CMV in the three-phase OW-PMSM system, by synchronously adjusting the duration time of each phase within every control cycle based on the reference voltage of zero-sequence control loop.
Abstract: The zero-sequence loop inherently exists in the structure of the three-phase open-end winding permanent-magnet synchronous machine (OW-PMSM) system with common dc bus, which provides the circulating path for zero-sequence current (ZSC) and consequently causes the torque ripple and extra system loss. Besides, the common-mode voltage (CMV), which can cause many negative effects, such as the bearing current and the failure of motor bearing, is another concern in the driving system. However, the simultaneous suppression of ZSC and the CMV control of the devices is investigated mainly in the application of the induction machine system and the multiphase OW-PMSM system, whereas it is ignored in the application of the three-phase OW-PMSM system, in which the suppression of ZSC is more complex considering the effect of flux linkage harmonic components. On this basis, this article proposes a strategy that can not only suppress the ZSC, but also eliminate the ripple of CMV in the three-phase OW-PMSM system. In the proposed method, the suppression of ZSC is achieved by synchronously adjusting the duration time of each phase within every control cycle based on the reference voltage of zero-sequence control loop, which can simplify the conventional rearrangement-based scheme. Furthermore, the duration time of each phase is distributed into each switch device of the two inverters, with the principle of eliminating the ripple of CMV. Consequently, both the suppression of ZSC and the CMV control can be achieved easily. Furthermore, the influence of dead time on the CMV control is analyzed, indicating that the CMV control is irrelevant with the dead time. Moreover, the effective modulation range of the proposed technique is analyzed. Finally, the experimental validation is conducted on a three-phase OW-PMSM system with common dc bus.
TL;DR: In this article, a physics-informed deep learning approach was proposed for bearing condition monitoring and fault detection, which consists of a simple threshold model and a deep convolutional neural network (CNN) model.
TL;DR: This research presents a novel and scalable approach that combines reinforcement learning and reinforcement learning for rolling bearing fault diagnosis and shows real-time improvements in the accuracy and efficiency of these methods.
Abstract: Machine learning methods are widely used for rolling bearing fault diagnosis. Most of them are based on a basic assumption that training and testing data are adequate and follow the same distributi...
TL;DR: In this paper, a comprehensive model considering the kinematics of the bearing components, the Hertzian contact between the rolling elements and raceways, the interaction between the bearing and cage, the hydro-dynamic lubrication, and the thermal effects is introduced to study and forecast the over-skidding and skidding mechanisms.
TL;DR: A new feature extraction method based on co-occurrence matrices for bearing vibration signals was proposed instead of the conventional feature extraction methods, as in the literature, and the success rate is 87.50% for data sets used to test the proposed approach.
Abstract: Recently, precise and deterministic feature extraction is one of the current research topics for bearing fault diagnosis. For this aim, an experimental bearing test setup was created in this study....
TL;DR: In this paper, an ensemble data-driven approach is proposed to predict the remaining useful life (RUL) of bearings, which is regarded as one of the critical approaches to avoid failure of bearings and their systems.
Abstract: Bearing is a key component in rotary machines. Their failures may cause the abrupt shutdown of these machines, which would result in substantial economic losses. Therefore, the prediction of the remaining useful life (RUL) of bearings is regarded as one of the critical approaches to avoid failure of bearings and their systems. In this article, an ensemble data-driven approach is proposed to predict the RUL of bearings. It uses feature extraction, an attention mechanism, and uncertainty analysis. First, the features embedded in the bearings’ vibration signals are extracted. Second, a stacked gated recurrent unit (GRU) is constructed to predict the bearing RUL. A novel attention mechanism based on dynamic time warping (DTW) is developed to improve the performance of information extraction, and a Bayesian approach is employed to analyze the prediction uncertainty. Finally, the proposed approach is validated using two benchmark-bearing data sets. The results show that the proposed approach can predict the bearing RUL effectively, and the prediction uncertainty can also be evaluated.
TL;DR: A fault diagnosis model based on composite multiscale permutation entropy and reverse cognitive fruit fly optimization algorithm optimized extreme learning machine (RCFOA-ELM) is proposed that can effectively improve the accuracy of fault classification and provide a new solution for rolling bearing fault diagnosis.
TL;DR: In this article, a dual-rotor-bearing-coupling misalignment system with blade-casing rubbing is investigated based on numerical simulation and experimental measurement, and the results reveal the nonlinear vibration characteristics of the dualrotor bearing-Coupling system with local blade-case rubbing fault.
TL;DR: A digital twin model of life-cycle rolling bearing driven by the data-model combination is proposed, with the measured signals and the bearing fault dynamic model, and the evolution law of bearing defect during the life cycle is revealed by a back propagation neural network.
Abstract: The digital twin of a life-cycle rolling bearing is significant for its degradation performance analysis and health management. This paper proposes a digital twin model of life-cycle rolling bearing driven by the data-model combination. With the measured signals and the bearing fault dynamic model, the time-varying defect size is estimated, and the evolution law of bearing defect during the life cycle is revealed by a back propagation (BP) neural network. Then the excitations of evolutionary defects are introduced into the bearing dynamic model, so as to form a life-cycle bearing dynamic model in the virtual space. Finally the simulation data in the virtual space is mapped into the corresponding data in the physical space via an improved CycleGAN neural network with the smooth cycle consistency loss. By comparing the obtained digital twin result with the measured signal in the time-domain and frequency-domain, the effectiveness of the proposed model is verified.
TL;DR: Experimental results show that the method proposed in this paper can extract the characteristic frequency of faulty bearing under stronger noise interference.
TL;DR: A novel model named LSS which combines the advantages of long short-term memory (LSTM) network with statistical process analysis to predict the fault of aero-engine bearings with multi-stage performance degradation with higher prediction accuracy is proposed.
TL;DR: A detector based on Ensemble Average of Autocorrelated Envelopes (EAAE) is proposed to identify the early occurrence faults in rolling element bearings, of which the fault induced vibration signals are inevitably contaminated or masked by both additive background noise and random phase noise.
TL;DR: A rolling element bearing fault diagnosis approach based on principal component analysis and adaptive deep belief network with Parametric Rectified Linear Unit activation layers is proposed, which results in an optimal DBN structure with high accuracy and convergence rate.
TL;DR: In this paper, a novel quadratic function-based deep convolutional auto-encoder is developed in order to predict the remaining useful life (RUL) of bearing.
Abstract: As one of the most important components of machinery, once the bearing has a failure, serious catastrophe may happen. Hence, for avoiding the catastrophe, it is valuable to predict the remaining useful life (RUL) of bearing. Health indicators (HIs) construction plays a greatly important role in the data-driven RUL prediction. Unfortunately, most of the existing HIs construction methods need prior knowledge and few of them construct HIs from raw vibration signals. For dealing with the above issues, a novel quadratic function-based deep convolutional auto-encoder is developed in this work. The raw bearing vibration signals are first preprocessed by low-pass filtering. Then the cleaned vibration signals are input into the quadratic function-based DCAE neural networks for constructing HIs of bearings. Compared with AE, DNN, KPCA, ISOMAP, PCA and VAE, it is revealed that the proposed methodology can construct a better HI from the raw bearing vibration signal in terms of comprehensive performance. Several comparative experiments have been implemented, and the results indicate that the HI constructed by quadratic function-based DCAE neural network has stronger predictive power than the traditional data-driven HIs.
TL;DR: Experiments indicate that, compared with other methods, the new method exhibits a higher accuracy in the multistate classification of rolling bearings under actual operating conditions when driven by the limited data with noise labels.
Abstract: The fault characteristics of the rolling bearings of wind turbine gearboxes are unstable under actual operating conditions. Problems such as inadequate fault sample data, imbalanced data types, and noise labels (error labels) in historical data occur. Consequently, the accuracy of wind turbine gearbox bearing fault diagnosis under actual operating conditions is insufficient. Hence, a new method for the fault diagnosis of wind turbine gearbox bearings under actual operating conditions is proposed. It uses an improved label-noise robust auxiliary classifier generative adversarial network (rAC-GAN) driven by the limited data. The improved rAC-GAN realizes a batch comparison between the generated and real data to ensure the quality of the generated data and improve the generalization capability of the model in scenarios of actual operating conditions. It can be used to generate a large number of multitype fault data that satisfy the characteristics of the probability distribution of real samples and display higher robustness to label noises. Experiments indicate that, compared with other methods, the new method exhibits a higher accuracy in the multistate classification of rolling bearings under actual operating conditions when driven by the limited data with noise labels.
TL;DR: The analysis results of rolling bearing signals show that APMD has excellent ability to identify and extract PCs and is a valid method for rolling bearing fault diagnosis.