Bearing Fault Research Articles

Tilting Pad Thrust Bearing Fault Diagnosis Based on Acoustic Emission Signal and Modified Multi-Feature Fusion Convolutional Neural Network

Tilting pad thrust bearings are widely utilized in large rotating machinery such as steam turbines and hydraulic turbines. Defects in their shaft tiles directly impact lubrication characteristics, thereby influencing the overall safety performance of the entire unit. To address this issue, this paper presents a fault diagnosis method for tilting pad thrust bearings using a modified multi-feature fused convolutional neural network (MMFCNN). Initially, an experimental bench for diagnosing faults in tilting pad thrust bearings was developed to collect multi-channel acoustic emission (AE) signals from both normal and faulty pads. Subsequently, the squeeze-and-excitation (SE) module was employed to reallocate the weights of each channel and fuse the features of multi-channel signals. Learning was then conducted on the signal fused with multiple features using the inverse-add module and spanning convolution. Next, a comparative analysis was carried out among the CNN1D, ResNet, and DFCNN models, and the MMFCNN model proposed in this study. The results show that under consistent operating conditions, the MMFCNN model achieves an average fault diagnosis accuracy of 99.58% when utilizing AE signal data from tilting pad thrust bearings in four states as inputs. Furthermore, when different operational conditions are introduced, the MMFCNN model also outperforms other models in terms of accuracy.

Physics-guided degradation trajectory modeling for remaining useful life prediction of rolling bearings

Remaining useful life (RUL) prediction has great significance in reducing operating costs and enhancing the maintainability and safety of rolling bearings. Recently, significant progress has been achieved in this field by leveraging deep learning approaches. However, advanced deep learning methods suffer from a black-box nature that makes them lack interpretability. Moreover, the prediction results may sometimes not follow laws of physics. To tackle this drawback, a physics-guided degradation trajectory modeling method is proposed for RUL prediction of rolling bearings, where physical knowledge is embedded in input preparation, model construction, and output specification. Specifically, phase space reconstruction is leveraged to construct the physics-guided inputs, transforming the degradation prediction of rolling bearings into the variation estimation of phase space trajectories. The degradation trajectory prediction strategy is derived from the physics-guided inputs and the decreased value of RUL is predicted through a compact one-dimensional convolutional neural network with the nonnegative bounded function. Simulation studies are performed with a theoretical model of defective rolling bearings to illustrate the effectiveness of the constructed physics-guided inputs. In addition, comparative analyses with advanced RUL prediction approaches under unseen working conditions and across different machines are conducted. The results demonstrate that the proposed method outperforms other approaches in both accuracy and robustness, showing its superior ability in RUL prediction of rolling bearings.

SAFO: Serial Amplitude Modulation–Frequency Modulation Operator for Multichannel Multicomponent Signal Decomposition and Its Application to Rolling Bearing Fault Diagnosis

SAFO: Serial Amplitude Modulation–Frequency Modulation Operator for Multichannel Multicomponent Signal Decomposition and Its Application to Rolling Bearing Fault Diagnosis

Fault characterization enhancement method for rolling bearings based on combination of weighted indicator screening and IMOMEDA

As one of the core components of aeroengine, it is very important to find out the fault of intermediate bearing in time for the safety of aeroengine. Early bearing fault features are easily buried by background noise, which makes it difficult to extract fault features. To solve this problem, a fault feature enhancement method based on weighted index screening and IMOMEDA is proposed in this article. First, employ wavelet packet decomposition to analyze the vibration signal; and the kurtosis, correlation coefficient, and permutation entropy are fused into weighted indexes by entropy weight method to screen the node components with high signal-to-noise ratio for reconstruction; then, envelope autocorrelation analysis is used to optimize the deconvolution cycle parameter T of the multipoint optimal minimum entropy deconvolution adjusted; and then the optimized IMOMEDA is used to filter out the noise components of reconstructed signals to achieve the enhancement of the fault features; and finally, envelope demodulation is performed to identify the fault feature information. After the simulation signal verification, the signal peak factor processed by this method is increased from 7.3 to 9.7, so the fault characteristics are effectively enhanced. The data of different fault types measured by the aeroengine intermediate bearing fault simulation test bench are analyzed. The results show that the method can effectively filter out the strong background noise in the vibration signal of different fault types, enhance the weak fault characteristics related to the bearing fault, and can be used as one of the effective methods for the fault diagnosis of the intermediate bearing of the aeroengine.

Damping-regulated GSR array method for multi-harmonic fault diagnosis under variable speed conditions

Abstract Bearing fault diagnosis under variable speed conditions is essential due to the complexities introduced by speed fluctuations. The accurate detection of multi-harmonic faults is critical for ensuring reliability in intricate operating environments. From the perspective of the beneficial effects of noise, in this study we propose a novel damping-regulated generalized stochastic resonance (GSR) array method designed for multi-harmonic fault diagnosis under variable speed conditions. First, we employ computed order tracking (COT) to transform non-stationary time-domain signals into stationary signals in the angular domain. A damping-regulated GSR oscillator is then introduced within this domain, forming the basis of our GSR array. By analyzing the system stationary response, we reveal the diagnostic performance in theory to assess the array’s capacity for enhancing multi-harmonic fault characteristics. Through simulations and experimental validation, our method demonstrates superior diagnostic accuracy, particularly in variable speed scenarios. It excels in preserving and enhancing weak multi-harmonic fault characteristics while offering significant advantages in high diagnostic robustness. These findings provide significant potential for practical applications in fault diagnostics across various engineering systems.

A wavelet packet transform-based domain adaptation method for rolling bearings fault diagnosis with fusion of local and global time–frequency features

Abstract Fault diagnosis transfer learning models commonly employ deep neural networks (DNN) to analyze time-frequency features. However, excessively deep neural networks can result in diminished generalization capabilities, leading to subpar performance of the model across various working conditions. Furthermore, inappropriate domain adaptation strategies significantly constrain the accuracy of the model. To address this issue, a Robustly Optimized Residual-Network and Vision Transformer Domain Adaptation Model (RoReViTDAM) is proposed in this article, combining Wavelet Packet Transform (WPT), residual networks, and self-attention mechanisms. Firstly, the WPT is utilized to construct Multi-band Wavelet Coefficient Matrix (MWCM) and corresponding Multi-band Wavelet Coefficient Time-Frequency Feature Matrix (MWSM) with small size and feature aggregation. Subsequently, a shallow Robustly Optimized Residual Network (RoResNet) is designed to effectively extract features from MWCM, considering the spatial distance dependencies of features. Additionally, ViT (Vision Transformer) is employed for time-frequency global feature extraction from MWSM. Furthermore, Domain Adversarial Neural Network (DANN) and Multi-Kernel Maximum Mean Discrepancy (MK-MMD) are employed to extract domain-invariant features from signals of different operating conditions and fault types. At last, three fault diagnosis experiments are conducted in multi-condition scenarios of bearings. The experimental results illustrate the superiority and effectiveness of the proposed model.

A Bearing Fault Diagnosis Model Based on a Simplified Wide Convolutional Neural Network and Random Forrest

Bearings play a crucial role in the complex mechanical systems of ships, and their operational status is closely related to vibration signals. Therefore, analyzing bearing signals plays an important role in the field of fault diagnosis. In order to solve the problems of low accuracy and slow response speed in fault diagnosis through vibration signals at mixed speeds, this paper introduces an improved Simple Window Deep Convolutional Neural Network with Random Forest (SWDCNN-RF) model on traditional Wide Convolutional Neural Network (WDCNN). It was verified through the publicly available dataset of ball bearings from Western Reserve University in the United States. It was found that the improved model increased speed by 38.51% and accuracy from 97.5% to 99.6% at epoch = 50, and also achieved faster convergence and smaller fluctuations during training. This study is of great significance for determining the occurrence time and type of bearing faults, and provides criteria for reliability evaluation and fault diagnosis of equipment using bearings.

Multi-Patch Time Series Transformer for Robust Bearing Fault Detection with Varying Noise

In time-series studies involving bearing sensor data, Gaussian noise and white noise techniques are commonly employed to evaluate model robustness. However, these conventional noise techniques are limited in their applicability to real-world industrial environments. This paper proposes three novel noise techniques—electrical interference noise, harmonic noise, and random shock noise—that more accurately reflect the complex noise encountered in industrial settings. Additionally, a new deep learning model, MultiPatchTST, is introduced, demonstrating robust performance under various noise conditions. Experimental results reveal that Gaussian noise has minimal impact on model performance, whereas the proposed noise techniques significantly affect performance, providing a more realistic evaluation of noise robustness. The proposed MultiPatchTST model achieves superior performance across all metrics in the presence of all four noise types, confirming its robustness and reliability.

Investigating the Effect of Vibration Signal Length on Bearing Fault Classification Using Wavelet Scattering Transform

Bearing condition monitoring and prognosis are crucial tasks for ensuring the proper operation of rotating machinery and mechanical systems. Vibration signal analysis is one of the most effective techniques for bearing condition monitoring and prognosis. In this study, the wavelet scattering transform, derived from wavelet transforms and incorporating concepts from scattering transform and convolutional network architectures, was utilized to extract quantitative features from vibration signals. The number of wavelet scattering coefficients obtained from vibration signals of different lengths varied due to the use of predefined wavelet and scaling filters in the wavelet scattering network. Additionally, these wavelet scattering coefficients are associated with different scattering paths within the corresponding wavelet scattering networks. Eight different lengths of vibration signals, associated with fifteen classes of rolling element bearing faults and conditions, were investigated using wavelet scattering feature extraction. The classes of rolling element bearing faults and conditions included normal bearings as well as ball and inner race faults with various fault diameters ranging from 0.007 inches to 0.028 inches. For the specific dataset validated, the computational results indicated that excellent bearing fault classification was achievable using wavelet scattering feature vectors derived from vibration signals with lengths of at least 6000 samples.

Feature Enhancement-Based Few-Shot Bearing Surface Defect Image Classification Method

The application of intelligent bearing surface defect classification based on deep neural networks remains challenging in real factories, due to the scarcity of defect samples. Under real-working conditions, with less training samples, the paper proposes a few-shot bearing defect image classification network which can recognize different bearing surface defects image, including notch, reddish rust, scratching, incising, conformity, pitting and mill scale. Based on general metric learning neural network framework, a local feature extraction layer is designed, which calculates the auto-correlation vector of global feature in a sliding region to enhance detail features. Additionally, a similar feature attention module emphasizes the the regions of similarity between the query set and the class prototype center to overcome the influence of background noise on classification. To validate the effectiveness of the proposed network, comparative experiments were conducted using the benchmark dataset miniImageNet, achieving classification accuracies of 59% in the 5-way 1-shot setting and 76% in the 5-way 5-shot setting respectively. Furthermore, to assess its performance in a real-factory condition, a self-made dataset of bearing defects from a factory was employed. The proposed network achieved a remarkable classification accuracy of 88% in the 5-way 5-shot setting. These experimental results confirm the practical application value of our few-shot bearing surface defect image classification network, demonstrating its ability to accurately recognize various bearing defects with limited training samples.

An Investigation of the Effect of the Fault Degree on the Dynamic Characteristics and Crack Propagation of Rolling Bearings

The incipient faults of rolling bearings are dynamically propagating in the service status. However, the bearing material and the size of the fault will affect its expansion trend and direction. Furthermore, bearings manufactured from different materials behave differently when they fail. Therefore, the influence of the fault degree on the dynamic characteristics and crack propagation of rolling bearings is investigated in this paper. First, a dynamic model of the bearing, both under fault-free conditions and with varying fault sizes on the outer ring, is established by considering the actual working conditions of rolling bearings. Then, the reliability of the dynamic model is verified theoretically and experimentally. Finally, the study examined the slip behavior of rolling elements, the variation trends in the maximum shear stress and principal stresses on the outer ring, and the direction of crack propagation under different fault severities. The results indicate that (1) the severity of roller slip becomes more pronounced with the expansion of the fault size; (2) material differences will affect the timing of macro-slip during faults; (3) crack propagation tends to initiate at the edge of the fault exit, with the propagation rate increasing as fault severity escalates; and (4) tensile stress was observed in the first principal stress, which accelerates crack bifurcation at the faulted edge, while both the second principal stress and third principal stress exhibit compressive stress, playing a suppressive role in crack bifurcation at the faulted edge. These findings provide a theoretical basis for further research on the evolution of faults in rolling bearings.

Machinery fault diagnosis based on a transfer learning model with local sparse structure of PMDCNN-LSTM

Aiming at the fault diagnosis problem in practical applications where the performance of the model degrades when applied to new domains or tasks due to limited and approximate training data, a transfer learning (TL) model with a local sparse structure based on a parallel multichannel convolutional neural network (CNN) and a long-short-term memory network (LSTM) (PMDCNN-LSTM) is proposed. The model is combined with CNN, LSTM, and TL to achieve parameter sharing by training learning parameters in the source domain and transferring them to the target domain. At the same time, the model is transformed from training data to real-world applications by integrating samples from the target domain, which improves generalization while maintaining the ability to identify faults in the training set. The advantages of CNN and LSTM are combined in this model compared with the traditional TL model, which can efficiently process data containing spatio-temporal features while maintaining sensitivity to time-series variations, exhibiting higher performance and advantages in complex data processing. It is shown by the experimental results that excellent performance and generalization ability can still be maintained by the model in high-noise environments, which is considered to be of great significance for research on bearing and gear fault diagnosis.

Rolling Bearing Fault Diagnosis Based on CBAM-WDCNN

Rolling bearing fault diagnosis methods based on convolutional neural networks (CNNs) have demonstrated excellent capabilities in processing complex nonlinear signals. Among them, the one-dimensional convolutional neural network greatly accelerates the signal processing process by directly processing the one-dimensional time-series signal and simplifying the input of the model. However, the traditional 1D convolutional neural network has problems such as poor feature recognition. In this paper, we propose a model based on the fusion of the attention mechanism CBAM and the deep convolutional neural network with a wide kernel in the first layer (WDCNN), which emphasizes the key features and suppresses the irrelevant noises, and significantly improves the model's recognition ability for fault features. The improved method is compared with other diagnostic algorithms to verify its higher accuracy and more stable training process.

State-of-the-Art Detection and Diagnosis Methods for Rolling Bearing Defects: A Comprehensive Review

State-of-the-Art Detection and Diagnosis Methods for Rolling Bearing Defects: A Comprehensive Review

Anomaly detection method of traction motor bearing based on multi-scale sub-band fuzzy entropy manifold fusion index

Detecting early faults in traction motor bearings poses significant challenges due to weak signals and difficulties in identifying fault initiation points with sufficient sensitivity. This paper introduces a novel anomaly detection method based on a multi-scale sub-band fuzzy entropy manifold fusion index (MFMI). The proposed method decomposes vibration signals across multiple scales to capture local features of bearing health, calculates sub-band fuzzy entropy to quantify fault characteristics, and uses locality preserving projection to retain nonlinear structural features while reducing dimensionality. Validation experiments using full-cycle acceleration life vibration signals demonstrate the superior performance of the proposed method. For instance, in the traction motor case, the proposed index detected early damage at the 189th time point, outperforming other indicators that detected damage after the 200th time point. The proposed method also shows higher sensitivity to early degradation trends while maintaining stability during normal operation. These results highlight the practical applicability of the method for early anomaly detection in traction motor bearings, offering earlier and more reliable fault detection compared to traditional methods.

Normalization-Guided and Gradient-Weighted Unsupervised Domain Adaptation Network for Transfer Diagnosis of Rolling Bearing Faults Under Class Imbalance

Normalization-Guided and Gradient-Weighted Unsupervised Domain Adaptation Network for Transfer Diagnosis of Rolling Bearing Faults Under Class Imbalance

MCRCNet: A Bearing Fault Diagnosis Method for Unknown Faults Based on Transfer Learning

MCRCNet: A Bearing Fault Diagnosis Method for Unknown Faults Based on Transfer Learning

Subway Train Bearing Fault Detection Under Variable Speed Based on Speed-Guided ACMD and Order Tracking

Subway Train Bearing Fault Detection Under Variable Speed Based on Speed-Guided ACMD and Order Tracking

Unsupervised Contrastive Learning-Based Single Domain Generalization Method for Intelligent Bearing Fault Diagnosis

Unsupervised Contrastive Learning-Based Single Domain Generalization Method for Intelligent Bearing Fault Diagnosis

Rolling Bearing Fault Diagnosis Based on a Synchrosqueezing Wavelet Transform and a Transfer Residual Convolutional Neural Network.

This study proposes a novel rolling bearing fault diagnosis technique based on a synchrosqueezing wavelet transform (SWT) and a transfer residual convolutional neural network (TRCNN) designed to address the difficulties of feature extraction caused by the non-stationarity of fault signals, as well as the issue of low fault diagnosis accuracy resulting from small sample quantities. This approach transforms the one-dimensional vibration signal into time-frequency diagrams using an SWT based on complex Morlet wavelet basis functions, which redistributes (squeezes) the values of the wavelet coefficients at different localized points in a time-frequency plane to the estimated instantaneous frequencies. This allows the energy to be more fully concentrated in actual corresponding frequency components. This strategy improves both the time-frequency aggregation and the resolution, which better reflects the eigenvalues of non-stationary signals. In this process, transfer learning and a residual structure are used in the training of a convolutional neural network. The resulting time-frequency diagrams, acquired using the steps discussed above, are then input to the TRCNN for diagnosis. A series of validation experiments confirmed that applying the TRCNN structure made it possible to achieve high diagnostic accuracy, even when training the network with only a small number of fault samples, as all 12 fault types from the test dataset were diagnosed correctly. Further simulation experiments demonstrated that our proposed method improved fault diagnosis accuracy compared to that of conventional techniques (with increases of 1.74% over RCNN, 1.28% over TCNN, 1.62% over STFT, 1.73% over WT, 2.83% over PWVD, and 1.39% over STFA-PD). In addition, diagnostic accuracy reached 100% during the application of three-time transfer learning, validating the effectiveness of the proposed method for rolling bearing fault diagnosis.