Failure Of Rolling Element Bearings Research Articles

Simulation-driven Bearing Fault Diagnosis for Condition Monitoring without Faulty Data

The failure of rolling element bearings in complex mechanical systems is a significant cause of mechanical failures, leading to decreased productivity and safety risks. Deep learning has shown promising results in bearing fault diagnosis, but the predictive performance depends on highquality data. Domain adaptation has been studied to solve this problem, but it still has limitations when applied to real-world industrial applications. In this study, we propose a deep learning-based domain generalization framework for bearing fault diagnosis using the bearing simulation model and adversarial data augmentation method. The proposed framework was validated on a real bearing fault dataset and showed promising results in improving diagnostic performance in cases where fault data cannot be obtained or when dealing with unlearned target domains. This approach has the potential to improve industrial maintenance systems by obtaining improved generalization performance in the absence of fault datasets.

Impulsive components separation using minimum-determinant KL-divergence NMF of bi-variable map for bearing diagnosis

In vibration-based monitoring techniques, the periodic impulse response is a typical manifestation of the localized failure of rolling element bearings (REBs). Cyclic spectral help reveal the cyclostationarity of such signals by bridging the carrier frequency and the cyclic frequency, which are represented in bi-variable maps. By integrating the components within the optimal band over the carrier axis, the Enhanced Envelope Spectrum (EES) or Squared Envelope Spectrum (SES) can be obtained and provide attractive analysis tools for the bi-variable map. In this paper, as a complement of integration analysis of the bi-variable map, Non-negative Matrix Factorization is adopted to separate the cyclic components in bi-frequency map. In addition to using KL-divergence as the fidelity term, the minimum determinant regularization term is crafted to enforce the sparseness, uniqueness, and identifiability of the solution. With numerical simulation and bearing fault experiments, the result shows that the number of cyclic components is estimated correctly and that the periodic impulsive components are separated and enhanced from the bi-variable map, especially in the cases that cyclic interferences mask the fault feature.

Numerical study on butterfly wings around inclusion based on damage evolution and semi-analytical method

Butterfly wings are closely related to the premature failure of rolling element bearings. In this study, butterfly formation is investigated using the developed semi-analytical three-dimensional (3D) contact model incorporating inclusion and material property degradation. The 3D elastic field introduced by inhomogeneous inclusion is solved by using numerical approaches, which include the equivalent inclusion method (EIM) and the conjugate gradient method (CGM). The accumulation of fatigue damage surrounding inclusions is described using continuum damage mechanics. The coupling between the development of the damaged zone and the stress field is considered. The effects of the inclusion properties on the contact status and butterfly formation are discussed in detail. The model provides a potential method for quantifying material defects and fatigue behavior in terms of the deterioration of material properties.

An Automated Edge Computing-Based Condition Health Monitoring System: With an Application on Rolling Element Bearings

Abstract A failure of rolling element bearings is a frequent cause of machine breakdowns and results in a production loss due to the sudden failure. A regular condition health monitoring and an associated detection of bearing defects in the early stages can be used to predict such sudden failures. To monitor the bearing's condition, the generated vibration signature can be analyzed, since rotating machines have, in most instances, a unique vibration signature that relates to their health status. Presently, bearing analysis of many machines results in significant cost and complexity due to a large amount of vibration data that must be analyzed. A condition health monitoring system (CMS) was developed to automate and simplify the whole process from the vibration measurement to the analysis results. Additionally, the CMS is embedded into an Internet of Things (IoT) architecture. Thereby, a location-independent control of the CMS, the vibration data, and the analysis results is possible. The embedding of sensors can cause communication problems from the sensor to the cloud due to the low bandwidth of sensors and the amount of data that must be transmitted. To overcome this issue, an edge device that acts as a gateway between the vibration sensor and the cloud is the core of the CMS. It measures the vibration signal locally, analyzes it automatically, and publishes a feedback as to the bearing condition to the cloud.

Bearing Fault Detection Using Scalogram and Switchable Normalization-Based CNN (SN-CNN)

Bearings play a vital role in all rotating machinery, and their failure is one of the significant causes of machine breakdown leading to a profound loss of safety and property. Therefore, the failure of rolling element bearings should be detected early while the machine fault is small. This paper presents the model that detects bearing failures using the continuous wavelet transform and classifies them using a switchable normalization-based convolutional neural network (SN-CNN). State-of-the-art accuracy was achieved with the proposed model using the Case Western Reserve University (CWRU) bearing dataset, which serves as the primary dataset for validating various algorithms for bearing failure detection. Batch normalization techniques were also employed and compared to the proposed model. The spectrogram images were also used as input for further comparison. Using switchable normalization, the proposed model achieved the testing accuracy in between 99.44% and 100% for different batch sizes and datasets.

Optimal Sub-Band Analysis Based on the Envelope Power Spectrum for Effective Fault Detection in Bearing under Variable, Low Speeds.

Early identification of failures in rolling element bearings is an important research issue in mechanical systems. In this study, a reliable methodology for bearing fault detection is proposed, which is based on an optimal sub-band selection scheme using the discrete wavelet packet transform (DWPT) and envelope power analysis techniques. A DWPT-based decomposition is first performed to extract the characteristic defect features from the acquired acoustic emission (AE) signals. The envelope power spectrum (EPS) of each sub-band signal is then calculated to detect the characteristic defect frequencies to reveal abnormal symptoms in bearings. The selection of an appropriate sub-band is essential for effective fault diagnosis, as it can reveal intrinsically explicit information about different types of bearing faults. To address this issue, we propose a Gaussian distribution model-based health-related index (HI) that is a powerful quantitative parameter to accurately estimate the severity of bearing defects. The most optimal sub-band for fault detection is determined using two dimensional (2D) visualization analysis. The efficiency of the proposed approach is validated using several experiments in which different defect conditions are identified under variable, and low operational speeds.

Bearing Fault Diagnosis by a Robust Higher-Order Super-Twisting Sliding Mode Observer.

An effective bearing fault detection and diagnosis (FDD) model is important for ensuring the normal and safe operation of machines. This paper presents a reliable model-reference observer technique for FDD based on modeling of a bearing’s vibration data by analyzing the dynamic properties of the bearing and a higher-order super-twisting sliding mode observation (HOSTSMO) technique for making diagnostic decisions using these data models. The HOSTSMO technique can adaptively improve the performance of estimating nonlinear failures in rolling element bearings (REBs) over a linear approach by modeling 5 degrees of freedom under normal and faulty conditions. The effectiveness of the proposed technique is evaluated using a vibration dataset provided by Case Western Reserve University, which consists of vibration acceleration signals recorded for REBs with inner, outer, ball, and no faults, i.e., normal. Experimental results indicate that the proposed technique outperforms the ARX-Laguerre proportional integral observation (ALPIO) technique, yielding 18.82%, 16.825%, and 17.44% performance improvements for three levels of crack severity of 0.007, 0.014, and 0.021 inches, respectively.

Intelligent bearing performance degradation assessment and remaining useful life prediction based on self-organising map and support vector regression

Rolling element bearings are critical components of rotating machines since the failure of rolling element bearings may cease the functioning of the entire equipment. The damages observed due to bearing failures are expeditious in nature and hence the need to develop an effective prognostic methodology becomes a requisite to prevent the sudden machinery breakdown. The performance degradation assessment and accurate determination of remaining useful life are the two key issues in prognostics of rolling element bearings. This paper proposes a degradation indicator based on self-organising map for the performance degradation assessment of bearings and later support vector regression is utilised to estimate the remaining useful life of bearings. The time-domain and frequency domain features extracted from the raw bearing vibration signals are supplied to the self-organising map classifier to achieve the degradation index termed as self-organising map-minimum quantisation error evolution in the paper. For estimating the remaining useful life of bearings, first the central trend of minimum quantisation error is extracted to achieve the feature vector defined as bearing health index in this work. The bearing health index is then used as input and the life percentage of the bearing is set to output in order to build the support vector regression prediction model for remaining useful life estimation of bearings. The proposed method is validated on the vibration signatures collected in a bearing test rig. The results show that the advocated method can efficiently track the evolution of deterioration and predict the remaining useful life of bearings.

On the thermally-induced failure of rolling element bearings

A comprehensive investigation of two types of thermal failure of rolling element bearings is carried out both experimentally and analytically. The first type is concerned with the thermal failure of rolling bearings that can occur at high rotational speeds and large radial loads, and the second type deals with spindle bearings of high-speed machine tools. Developed in the present study is a simulation technique that considers the thermal expansion of the bearing elements in the dynamic simulations of the rolling bearings during the thermal failure. The results of dynamic simulations for the first type of thermal failure show that the unstable motion of the cage can lead to an ultimate bearing seizure because of the cage failure due to the large rollers/cage contact forces and high wear rate of the cage. However, the surface damage and wear at the roller/raceways interface are not considerable. On the contrary, for the second type of thermal failure of rolling element bearings, the simulation results reveal that the minimum film thickness at the roller/raceways interface drops and mixed lubrication regime prevails. Subsequently, a severe surface damage and wear occur at the rollers/raceways contact surfaces and eventually the bearing fails. The cause of this failure is the thermal seizure of the spindle bearing due to the rapid rise of the thermally-induced preload inside the bearing assembly with no sign of cage instability.

Understanding white etching cracks in rolling element bearings: Formation mechanisms and influent tribochemical drivers

Among prevalent tribological failures in rolling element bearings, a peculiar rolling contact fatigue mode has been defined as white etching cracks, which correspond to broad subsurface three-dimensional branching crack networks bordered by white etching microstructure. White etching cracks tend to appear before other conventional rolling contact fatigue microstructural alterations and eventually lead to premature flaking or radial cracking that remains unpredictable using fatigue life estimations. Even though they have been reported for several decades, occurrences present no common evident denominator and are delicate to reproduce on laboratory test rigs without artificial hydrogen charging. In this study, white etching crack reproductions on two different standard endurance test rigs are compared and analyzed in order to, firstly, propose, an update of the understanding of white etching crack surface affected tribochemical formation mechanisms, and secondly, to identify influent tribochemical and mechanical drivers. If white etching cracks are associated to different combinations of specific non-self-sufficient macroscopic drivers depending on the application or the test rig, evidences demonstrate that they often come down to similar phenomena at a microscopic tribological scale that should all be mastered for efficient white etching crack reproduction and countermeasure design.

Investigations of Bearing Failures Associated with White Etching Areas (WEAs) in Wind Turbine Gearboxes

A critical problem for wind turbine gearboxes is failure of rolling element bearings where axial cracks form on the inner rings. Metallurgical analyses show that the failure mode is associated with microstructural alterations manifested by white etching areas (WEAs) and white etching cracks (WECs). This article presents field experience from operating wind turbines that compares performance of through-hardened and carburized materials. It shows that through-hardened bearings develop WEA/WECs and fail with axial cracks, whereas carburized bearings do not. In another comparison of two rotor bearings with different carburized metallurgies, one bearing developed WEA/WECs and failed by macropitting, whereas the other bearing did not develop WEAs or WECs and did not fail. The field experience shows that a carburized bearing that has a core with low carbon content, high nickel content, greater compressive residual stresses, and a higher amount of retained austenite provides higher fracture resistance and makes carburized bearings more durable than through-hardened bearings in the wind turbine environment.

Electric discharge damage in aircraft propulsion bearings

Rolling contact fatigue (RCF) is a common failure mode for rolling element bearings, however evidence of precursor failure modes or initiators can be lost or obscured by the subsequent severe damage caused by RCF. Although electric discharge damage (EDD) is not normally associated with rolling element bearings in aviation propulsion applications, it can occur and may not be instantly recognisable. Unlike factors such as lubricant cleanliness or misalignment, EDD does not normally form part of the life prediction of aviation propulsion rolling element bearings. Early identification of EDD and subsequent mitigation or elimination is, therefore, essential to prevent significant reduction in life or failure of rolling element bearings. This paper will review the phenomenon of EDD and discuss several recent examples observed in aviation propulsion systems.

Bearing condition diagnosis and prognosis using applied nonlinear dynamical analysis of machine vibration signal

In this paper, we introduce a modified form of the correlation integral developed by Grassberger and Procaccia referred to as the partial correlation integral, which can be computed in real time. The partial correlation integral algorithm is then used to analyze machine vibration data obtained throughout a life test of a rolling element bearing. From the experimental results, the dimensional exponent (an approximation of the correlation dimension) as computed from the partial correlation integral algorithm tends to increase as time progresses and the useful remaining life of the bearing is decreasing. The dimensional exponents of a healthy bearing and a bearing close to failure are statistically different. We also propose a computational scheme for bearing condition monitoring (diagnosis and prognosis) using the dimensional exponent integrated with a surrogate data testing technique. As a result, we can characterize the condition of the bearing from the results of the surrogate data test and furthermore, we provide some preliminary evidence that the dimensional exponent can be used to predict the failure of rolling element bearings in rotating machinery from real-time vibration data.

Online Vibration Monitoring Of Ball Bearing DamageUsing an Experimental Test Rig

Rolling element bearings in rotating machinery are vulnerable mechanical components that are prone to premature failure with dire consequences. Early warning of impending bearing failures is vital to the safety and reliability of high-speed turbomachinery. At present, vibration monitoring is one of the most common procedures used to detect online damage and early failure of rolling element bearings. This paper presents results from an experimental rotor-bearing test rig with known induced damage in the supporting ball bearings. Two types of damages are examined, namely, that of 1) the ball rolling element and 2) the inner race of the bearing. Four online vibration signature analyzing schemes are used: 1) time averaging, 2) frequency domain analysis, 3) joint time-frequency analysis (Wigner-Ville and wavelet transforms), and 4) chaotic vibration analysis (modified Poincare diagrams). The results from the various procedures are compared for efficiency in early pinpointing of damage and reliability in the detection. Based on this study, it was found that, in the case of race damage in which interactions with the damage area take place in a periodic manner during shaft rotation, vibration signals resulting from contacts with the damage location can generally be found within a typical window of data sampling, and damage can be identified using all four selected online procedures. For the case of ball-element damage, the vibration signals are chaotic in nature due to the unpredictable rotational patterns of the ball elements. Under such conditions, interactions with the ball-element damage area may not occur within the particular data sampling period, and as a result, damage cannot be detected by the traditional time-frequency procedures. It is also demonstrated that, in this study, the modified Poincare map approach using a relatively large amount of sampling data can be effective in identifying damage in the ball rolling element.

Application of statistical moments to bearing failure detection

Failure of rolling element bearings plays a significant role in the breakdown of industrial machinery. However, the analysis of signals resulting from measurements taken from outer casings of equipment has proven to be an effective and powerful tool for the early detection of failure in bearings. Although a number of techniques have been developed over the years directed at warning of impending failure, in most cases these methods are only effective in the later stages of damage development. This paper looks at variations of the statistical moment analysis method that show potential for damage detection at a much earlier stage. This approach has several advantages over other methods in that measurements taken are essentially independent of load and speed. Data for the analysis is relatively easy to collect using an accelerometer mounted near the bearing of interest and can then be processed on a micro-computer using suitable software. An important part of the processing is separating out unwanted data from other energy sources within the machine. This is achieved by developing selective digital filtering within the software. In this paper data from damaged and undamaged bearings are compared on the basis of analyzing both rectified and unrectified signals.

The application of vibration signature analysis and acoustic emission source location to on-line condition monitoring of anti-friction bearings

The mechanical noise, vibration and heat generated by defects in fabrications, process plant and machinery are a measure of component condition, and once characterised allow on-line assessment of defect severity. Unexpected failures which might threaten the entire production process can thereby be avoided. There is an increasing need for reliable and robust condition monitoring instrumentation with on-line data processing, particularly as the trend contunues towards larger production units with little or no standby capacity for use in the event of unscheduled stoppages due to mechanical failure. This article describes the detection of incipient failure of rolling element bearings by kurtosis and the location of fatigue cracks in slowly rotating bearings by acoustic emission