Rolling Bearings Research Articles

Predictive framework for remaining useful life of roller bearings: Utilizing fractional generalized Pareto degradation model in performance evaluation

Research on predictive frameworks for the Remaining Useful Life (RUL) of mechanical systems is limited. This study proposes a comprehensive RUL prediction framework for roller bearings, integrating early failure assessment, adaptive failure threshold determination, and an adaptive drift fractional-order Pareto degradation model. To tackle data insufficiency, multi-sensor fusion combines time-domain, frequency-domain, and time–frequency features. The framework normalizes the health indicator (HI) using Mahalanobis distance and a sigmoid function, and employs the MD-CUSUM technique for early fault detection. Dynamic threshold updating is achieved through BOX-COX transformation and Chebyshev inequality, establishing confidence intervals. The model demonstrates significant results: lowest RMSE of 5.4267, MAE of 3.7857, highest SOR of 0.90936, HD of 0.93543, PICP of 57.143%, and the narrowest MPIW of 9.45, showing enhanced accuracy and reduced uncertainty. Validation with the XJTU-SY bearing dataset confirms improved performance.

Remaining Useful Life Prediction of Rolling Bearings Based on Adaptive Continuous Deep Belief Networks and Improved Kernel Extreme Learning Machine

ABSTRACTRolling bearings are crucial components in a wide variety of machinery. Monitoring their conditions and predicting their remaining useful life (RUL) is vital to prevent unexpected breakdowns, optimize maintenance schedules, and reduce operational costs. This article proposes an approach based on adaptive continuous deep belief networks (ACDBN) and improved kernel extreme learning machine (KELM) to predict the RUL of rolling bearings. In the proposed approach, the ACDBN model is used for extracting hidden fault features and the distance between the initial health state and the real‐time degradation state is used to construct a health indicator (HI). Then, a hybrid kernel extreme learning machine prediction model optimized by the sparrow search algorithm (SSA‐KELM) is proposed to estimate the RUL using the extracted HIs. The SSA is used to find the optimal parameters of the KELM model. The proposed method has been assessed using existing bearing datasets. The obtained results indicate that the proposed method successfully improves RUL prediction accuracy compared to existing approaches in the literature.

SRCAE-STCBiGRU: a fused deep learning model for remaining useful life prediction of rolling bearings

SRCAE-STCBiGRU: a fused deep learning model for remaining useful life prediction of rolling bearings

Small-sample health indicator construction of rolling bearings with wavelet scattering network: An empirical study from frequency perspective

As a critical issue of diagnostics and health management (PHM), health indicator (HI) construction aims to describe the degradation process of bearings and can provide essential support of domain knowledge for early fault detection and remaining useful life prediction. In recent years, various deep neural networks, with end-to-end modeling capability, have been successfully applied to the HI construction for rolling bearings. In small-sample environment, however, the degradation features would not be extracted well by deep learning techniques, which may raise insufficient tendency and monotonicity characteristics in the obtained HI sequence. To address this concern, this paper proposes a HI construction method based on wavelet scattering network (WSN) and makes an empirical evaluation from frequency perspective. First, degradation features in different frequency bands are extracted from vibration signals by using WSN to expand the feature space with different scales and orientations. Second, the frequency band with the optimal scale and orientation parameters is selected by calculating the dynamic time wrapping (DTW) distance between the feature sequences of each frequency band and the root mean square (RMS) sequence. With the feature subset from the determined frequency band, the HI sequence can be built by means of principal component analysis (PCA). Experimental results on the IEEE PHM Challenge 2012 bearing dataset show that the proposed method can work well with only a small amount of bearing whole-life data in obtaining the HI sequences with high monotonicity and correlation characteristics. More interestingly, the critical frequency band whose information supports decisively the HI construction can be clarified, raising interpretability in a frequency sense and enhancing the credibility of the obtained HI sequence as well.

Predicting resistance of bearing steels to surface initiated fatigue: Application to hybrid contact

The constant demand for improved energy efficiency in lubricated rolling/ sliding contacts, such as in rolling bearings, is generally addressed through increased power density and the use of low-viscosity (i.e., thin-film) lubricants. In these conditions, mixed lubrication regime can dominate with intensive asperity interactions between two contacting surfaces, thus increasing the risk of surface-initiated fatigue (e.g., in the form of micropitting). In the present study, the influence of steel mechanical properties on the micropitting accumulation in a hybrid contact (steel versus Si3N4 ceramics of different roughness) has been experimentally and theoretically investigated. It was found that higher roughness of Si3N4 surface could increase the risk of micropitting on the steel component. However, despite formation of some mild micropitting, the bearing steels tested here show good performance. For better understanding of the relationship between the mechanical properties of bearing steels and micropitting resistance, a new theoretical framework is introduced. The new micropitting model, based on the finding that the nucleation of a micropit is driven by (i) local accumulation of the plastic strain in mixed rough contact, and (ii) reduction of the total energy by debonding of the plastically deformed zone is introduced. The new micropitting model was shown to match with experiments across the range of test conditions. Micropitting performance maps based on the mechanical properties of steels (yield strength, work hardening and fracture toughness) have been obtained. The new theory has implications not only for predicting micropitting but also for predicting adhesive wear resistance of different steels.

Enhanced fault diagnosis of rolling bearings using an improved inception-lstm network

ABSTRACT Rolling bearings are integral to the operation of various mechanical systems, where their condition directly impacts equipment reliability and performance. Accurate fault diagnosis is essential for preventing unexpected failures and minimising downtime in industrial settings. This study presents a novel diagnostic model that combines Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks, enhanced by an improved Inception module. This integrated approach captures both spatial and temporal features of vibration signals, leading to superior diagnostic accuracy under diverse operating conditions. Experimental results using the Case Western Reserve University (CWRU) bearing dataset demonstrate an impressive average fault diagnosis accuracy of 99.83%, outperforming conventional models. The robustness of the proposed model against noise and its adaptability to varying operational environments underscore its practical value for real-world engineering applications, offering a reliable solution for maintaining the safety and efficiency of critical machinery.

Dynamic Modeling and Behavior of Cylindrical Roller Bearings Considering Roller Skew and the Influence of Eccentric Load

At high speeds, skew and skid may frequently occur for the rollers in cylindrical roller bearings, especially when under eccentric load, as the uneven load distribution along the generatrix of the roller further aggravates this phenomenon. In this paper, a dynamic model of a cylindrical roller bearing was established, taking into account roller skewing and interactions with the cage. Firstly, the interaction between the roller and the raceway was calculated by slicing the roller along its generatrix. Furthermore, the computation of the interaction between the roller and the cage is based on elastic theory, taking into account pocket clearance. Subsequently, the dynamic equations for both rollers and cage were derived. Based on this foundation, an investigation was conducted to reveal how rotational speed, radial loads, and moment loads affect roller slipping, skewing characteristics, and interactions with the cage under uneven load conditions. The findings indicate a direct proportionality between roller slipping and bearing speed while exhibiting an inverse relationship with load magnitude. Additionally, it was observed that both bearing speed and load have a direct influence on roller skewing angle. Moreover, normal interaction force between the roller and cage demonstrates a direct proportionality to bearing speed while inversely correlating with load magnitude.

Research on Fault Diagnosis of Rolling Bearing Based on Gramian Angular Field and Lightweight Model.

Due to the limitations of deep learning models in processing one-dimensional signal feature extraction, and high model complexity leading to low training accuracy and large consumption of computing resources, this paper innovatively proposes a rolling bearing fault diagnosis method based on Gramian Angular Field (GAF) and enhanced lightweight residual network. Firstly, the one-dimensional signal is transformed into a two-dimensional GAF image, fully preserving the signal's temporal dependency. Secondly, to address the parameter redundancy and high computational complexity of the ResNet-18 model, its residual blocks are improved. The second convolutional layer in the downsampling residual blocks is removed, traditional convolutional layers are replaced with depthwise separable convolutions, and the lightweight Efficient Channel Attention (ECA) module is embedded after each residual block. This further enhances the model's ability to capture key features while maintaining low computational cost, resulting in a lightweight model referred to as E-ResNet13. Finally, the generated GAF feature maps are fed into the E-ResNet13 model for training, and through a global average pooling layer, they are mapped to a fully connected layer for classifying the faults of rolling bearings. Verifying the superiority of the proposed GAF-E-ResNet13 model, experimental results show that the GAF image encoding method achieves higher fault recognition accuracy compared to other encoding methods. Compared with other intelligent diagnosis methods, the E-ResNet13 model demonstrates strong diagnostic performance and generalization capability under both a single condition and complex varying conditions, fully proving the innovation and practicality of this method.

Federated transfer learning-based distributed fault diagnosis method for rolling bearings

Abstract Current methods for bearing fault diagnosis often fall short in addressing data privacy concerns and typically rely on one-to-one transfer strategies, which are inadequate for achieving knowledge transfer in distributed environments. To address this issue, a distributed fault diagnosis method for rolling bearings based on federated transfer learning is proposed. This method ensures data privacy while integrating fault knowledge from multiple domains, thereby enabling more efficient knowledge transfer. Specifically, a Domain Adversarial Neural Network (DANN) is introduced as the base model within the federated learning framework. Additionally, Maximum Mean Discrepancy (MMD) is incorporated into the DANN to enhance the transfer of fault knowledge. Finally, a dynamic weighting parameter update method based on MMD is designed to evaluate the feature discrepancies between source and target domains, thereby updating the parameters of the federated framework and achieving global model aggregation. Experimental results on two bearing datasets demonstrate that the proposed method excels in both distribution alignment and fault diagnosis.

A hybrid method for fault diagnosis of rolling bearings

Abstract Traditional diagnostic methods often have insufficient accuracy and noise reduction, which leads to diagnostic errors. To address these issues, this paper proposes an advanced fault diagnosis model that combines the variational mode decomposition (VMD) improved by a Variable-Objective Search Whale Optimization Algorithm (VSWOA) with a Pelican Optimization (PO)-boosted Kernel Extreme Learning Machine (KELM) algorithm. The application of the method is shown here in the fault diagnosis of rolling bearings. The proposed VSWOA enhances the performance of VMD by incorporating a Sobol sequence, nonlinear time-varying factors, a multi-objective initial search strategy, and an elite Cauchy chaos mutation strategy, significantly improving noise reduction in vibration signals. Fault information is precisely extracted using waveform factors, sample entropy, and advanced composite multiscale fuzzy entropy, which enables effective feature screening and dimensionality reduction. The POA fine-tunes the KELM parameters, increasing the classification accuracy. The effectiveness of the model is verified through experimental evaluations using bearing data with injected Gaussian noise (from Case Western Reserve University) and the SpectraQuest datasets, where significant improvements in noise reduction and fault detection accuracy are achieved.

A fault detection of aero-engine rolling bearings based on CNN-BiLSTM network integrated cross-attention

Abstract Aero-engine rolling bearings are essential for engine health, in which disruptive failures can be prevented and reduce great losses in air flight. To improve the efficiency of fault detection, an improved network, named CNN- BiLSTM -Cross-Attention (CBLCA) was proposed. The Bidirectional Long Short-Term Memory (BiLSTM) layer captures the temporal features as the input data. The cross-attention mechanism is integrated with the Convolutional Neural Networks (CNN) layer and the BiLSTM layer respectively. More important feature information can be identified with the CBLCA model. The proposed model was also validated with the open-sourced aero-engine rolling bearings data set. To improve the identification accuracy, a novel method that combines Fast Fourier Transform (FFT) and Variational Mode Decomposition (VMD) is used for the data preprocessing. Each original signal sample is transformed into a feature set containing richer information, and the number of features significantly increased in the entire dataset. Compared with some existing LSTM models, such as LSTM, BiLSTM, CNN-BiLSTM, and CNN-LSTM, the classification accuracy was increased by 55%, 54%, 5%, and 7%, respectively. The processing method for vibration signals and the CBLCA model can improve the accuracy and reliability of fault diagnosis for aero-engine rolling bearings.

Adaptive fast iterative filter Holo-spectrum analysis and its applications to fault diagnosis of rolling bearing

As a signal demodulation analysis technique, Holo–Hilbert spectral analysis (HHSA) excels in capturing the intricate cross-scale coupling dynamics present in nonlinear and non-stationary vibration signals. Nonetheless, HHSA suffers from a lack of rigorous mathematical foundation, is subject to modal mixing constraints, and exhibits limited noise robustness. To address the aforementioned issues, this study presents an innovative nonlinear and non-stationary signal demodulation technique, referred to as adaptive fast iterative filter Holo-spectrum analysis (AFIFHSA). Also, an adaptive fast iterative filtering (AFIF) algorithm incorporated within AFIFHSA is designed to dynamically achieve a nonlinear and non-stationary signal decomposing. From that, several approximate narrowband signals, possessing physical significance at an instantaneous frequency, and a trend term can be obtained. Furthermore, the marginal spectrum (MS) obtained by AFIFHSA can be utilized to represent the effectiveness of fault characteristic identification. Lastly, the simulation and measured data are utilized to showcase AFIFHSA’s exceptional capabilities in recognizing high-resolution and eximious modulation relationships. The analysis outcomes additionally illustrate that AFIFHSA, as proposed, showcases superior performance in fault identification and robustness with comparison to other conventional approaches.

Evaluation and analysis of abrasive wear resistance of hybrid roller bearings under lubricant contamination

Abrasive wear is a common failure mode of rolling bearings. To enhance the resistance of traditional all-steel bearings to abrasive wear, the structure of traditional all-steel cylindrical thrust roller bearings has been appropriately adjusted, and hybrid roller bearings have been formed by replacing some of the steel rollers in the traditional bearings with ceramic rollers. The frictional performance of hybrid roller bearings under lubricant contamination conditions was analyzed, and the wear reduction mechanism was discussed. The results show that under the condition of lubricant contamination, replacing the appropriate amount of ceramic rollers not only helps reduce friction and decrease wear but also contributes to improving the overall temperature rise of the bearing. Furthermore, when the total amount of contaminants is constant, a specific threshold exists for the impact of replacing the number of rolling elements on reducing mass loss. For a substantial reduction in wear with hybrid roller bearings, the replacement rollers should constitute at least one-fifth of the total number of rollers. The wear reduction observed in hybrid roller bearings results from a combination of crushing and refinement, grinding and finishing, and self-healing mechanisms.

Influence of surface texture on pocket pairs lubrication performance of cylindrical roller bearings

PurposeThis paper aims to reveal the influence of the grooved texture parameters on the lubrication performance of circular pocket-roller pairs in cylindrical roller bearings.Design/methodology/approachIn this paper, the thermal elastohydrodynamic lubrication mathematical model of the grooved texture circular pocket-roller pair was established, the finite difference method and successive over-relaxation method were used to solve the model, the influence of texture quantity, texture depth and texture area ratio on circumferential bearing capacity, friction coefficient, maximum temperature rise, stiffness and damping of the circular pocket-roller pairs were analyzed.FindingsThe results show that texture quantity, texture depth and texture area ratio significantly influence the static and dynamic characteristics of circular pocket-roller pairs. The suitable surface groove texture parameters can dramatically improve the circumferential bearing capacity, reduce the friction coefficient, inhibit the maximum temperature rise and increase the stiffness and damping of the circular pocket-roller pairs.Originality/valueThe research in this paper can provide a theoretical basis for the optimization design of pockets in cylindrical roller bearings to reduce friction and vibration.

Review of rolling bearings performance degradation trend prediction

Review of rolling bearings performance degradation trend prediction

Research on Small Sample Rolling Bearing Fault Diagnosis Method Based on Mixed Signal Processing Technology

The diagnosis of bearing faults is a crucial aspect of ensuring the optimal functioning of mechanical equipment. However, in practice, the use of small samples and variable operating conditions may result in suboptimal generalization performance, reduced accuracy, and overfitting for these methods. To address this challenge, this study proposes a bearing fault diagnosis method based on a symmetric two-stream convolutional neural network (CNN). The method employs hybrid signal processing techniques to address the issue of limited data. The method employs a symmetric parallel convolutional neural network (CNN) for the analysis of bearing data. Initially, the data are transformed into time–frequency maps through the utilization of the short-time Fourier transform (STFT) and the simultaneous compressed wavelet transform (SCWT). Subsequently, two sets of one-dimensional vectors are generated by reconstructing the high-resolution features of the faulty samples using a symmetric parallel convolutional neural network (CNN). Feature splicing and fusion are then performed to generate bearing fault diagnosis information and assist fault classification. The experimental results demonstrate that the proposed mixed-signal processing method is effective on small-sample datasets, and verify the feasibility and generality of the symmetric parallel CNN-support vector machine (SVM) model for bearing fault diagnosis under small-sample conditions.

Research on ACMD-ICYCBD method for rolling bearing fault feature extraction

Aiming at the difficulty in obtaining the eigenfrequency of the vibration component of rolling bearing faults in a strong background noise environment and the problem of extraction efficiency, the adaptive chirp mode decomposition (ACMD) combined with Improved maximum second-order cyclostationary blind deconvolution (ICYCBD) fault feature extraction algorithm is proposed. Firstly, to improve the signal-to-noise ratio, the original signal is adaptively decomposed using the ACMD method, and the optimal components are selected based on the principle of maximizing the correlation gini coefficient index. Secondly, to improve the accuracy of parameter setting and extraction efficiency, an improved CYCBD method is proposed to estimate the cyclic frequency set of CYCBD using the proposed enhanced energy harmonic product spectrum (EEHPS) method for the optimal components, the envelope spectrum peak factor index is improved by proposing the envelope spectral period pulse factor (EPPF) index, and the filtering length of the CYCBD is selected adaptively using the step search to obtain the optimized filtered signal. Finally, the envelope spectrum analysis is carried out to extract the fault information accurately. The simulation signals and experimental data show that the method can quickly and accurately extract the fault characteristics of rolling bearings under strong background noise, and the comparison with other methods shows the effectiveness and superiority of the proposed method.

Mechanism analysis of burn-up failure for needle roller bearing at high-speed gear in manual transmission

Adequate lubrication conditions of needle roller bearings are crucial for the normal operation of transmission. Addressing the burn-up and adhesion failures of needle roller bearings during transmission bench trials, two rigid body rotation physical models were developed to analyze and model the lubrication effect on the inner and outer surfaces of needle roller bearing. Through theoretical analysis and lubrication simulation, it was found that under the critical condition of d/D=0.553, the centrifugal force F on the outer ring surface of the needle bearing was related to the density of lubricant, rotational speed, as well as the angle and size of oil hole, rather than the position of the oil holes; Moreover, the centrifugal force F on the needle roller bearing at input shaft was related to the speed of transmission gear and was also associated with the clearance size between needle roller and cage. Theoretical analysis, simulation, and experimental results have demonstrated that the lubrication performance of needle roller bearings during operation at lower rotary speed gears was inferior to that at higher speed gears, and the lubrication was also less effective when the cage clearance was narrower. Employing angled oil inlets (holes), appropriately enlarging the diameter of oil inlets (holes), and increasing the clearance width between needle roller and cage will contribute to the improvement of lubrication effect.

ON THE ISSUE OF ENSURING THE OPERABILITY OF FREIGHT CAR AXLE BOXES

The article is devoted to analysing the reliability of freight car axle boxes. Wheelsets and axle boxes are especially important for ensuring the traffic safety of trains. The study examined cases of cars uncoupling from a train along the route due to technical faults resulting in delays in train movement. The analysis covered the period of 1995–2023. It found that failures of car axle boxes and automatic brakes predominated. After a significant decrease from 2000–2014, the number of roller axle box failures in recent years fluctuated at 30% of the total number of uncouplings. The analysis obtained a dependency for the change in car detachments due to roller bearing failures, calculated per 1,000 cars in the operating fleet. Despite fluctuations in the number of railcars, the safety level of operations in recent years remained virtually at the same level. The study highlighted that excessive heating accompanies damage in the elements of the axle box unit. It considered methods for detecting possible failures and their advantages and disadvantages. The authors identified the primary causes of axle box component failures. Cylindrical bearings and end mountings remain the most dangerous for failure probability. We demonstrated that the axle box design with cylindrical bearings has a fundamental flaw. Axial forces are perceived through the rolling friction of the ends of the rollers against the ring flanges, leading to the formation of ‘chevron’ scratches and dents on the roller ends, which are stress concentrators. The study established that the surface of the axle box housing is constantly in contact with the side frame of the freight car bogie and wears out. It results in pinching of the axle box body and contributes to the appearance of additional permanent forces acting on the axle box assembly. The end-bearing mounts are weakened and destroyed. Skewed rollers in the bearings cause fatigue failure of the rolling races. It leads to increased heating of the axle box assemblies and uncoupling of the car. A promising direction is the improvement of axle box bodies. Keywords: rolling stock, traffic safety, reliability, axle box assembly, roller bearing, axle box body, failure.

Temporal convolutional network with soft threshold and contractile self-attention mechanism for remaining useful life prediction of rolling bearings

Abstract RUL prediction is an effective approach to prevent system failures and cut maintenance expenditures. Due to the wide receptive field and the avoidance of future information leakage, temporal convolutional neural network (TCN) is widely applied for RUL estimation of bearings. However, the predictive performance of TCN is limited by the loss of degradation features and the breakdown of continuity of timing information. To overcome the above defects, hybrid temporal convolutional neural network with soft threshold and contractile self-attention mechanism (HTCN-SC) is presented. Firstly, the adaptive threshold is determined by the contraction self-attention mechanism with higher interpretability, which captures the contribution of different features to the estimation of RUL. Then, the soft threshold is employed to activate the degraded features. On the one hand, the degeneracy features endowed by the dilated causal convolution with obvious negative values are fully preserved. On the other hand, the noise components that are given low weights are completely suppressed compared to the original TCN. Finally, parallel branch composed of one-dimensional convolutional networks are used to supplement the continuity of time series. Degradation signals from different working conditions and bearings are employed to verify the performance of the HTCN-SC. The results indicate that HTCN-SC with accurate RUL estimation and generalization ability is an effective tool for rolling bearing health monitoring.