Bearing Fault Detection Research Articles

Wheel Bearing Fault Detection for Automobiles using Wheel Speed Sensor

Wheel bearing fault diagnosis and monitoring for automobiles are getting more attention as the market shifts towards electric vehicles (EV) and autonomous vehicles (AV). Electric drivetrains are expected to have longer lifespan than the reliable life of low-cost wheel bearing designs. The probability of having a failed wheel bearing in an EV thus increases. In an AV setting, most of the rides are taken by people who are unfamiliar to the vehicle, removing the human as a viable sensor to detect wheel bearing defects. Fleet managers will be required to pay a high price to inspect potential defects. The implementation of a wheel bearing fault detection system can help reduce these costs.
 In our previous work, a method for injecting wheel bearing brinelling failure was developed. In this paper, we propose a wheel bearing brinelling fault detection algorithm using wheel speed sensor. The proposed algorithm detects wheel bearing faults by identifying peaks in the wheel speed spectrum at bearing critical frequencies. To implement this approach, we leverage phase domain transform technique to normalize against effect of varying speed. A wavelet filtering technique is employed to enhance the bearing fault signatures prior transforming the wheel speed signal into frequency domain. Post-processing techniques are developed to smooth the high-variance wheel speed spectra and improve the detectability of the wheel bearing faults. A regression model is then developed to calculate the wheel bearing health indicators. The results show that the proposed algorithm achieves high performance in detecting wheel bearing faults.

Improving the Performance of Bearing Fault Detection on a Turbo Shaft Engine

Health and usage monitoring systems (aka “HUMS”) have typically not been associated with CFR Title 14 type 27 normal category rotorcraft (weighing less than 7000 pounds, with a seating capacity of 9 or less) due in part to the cost of such systems relative to the asset value. This paper describes performance improvements to HUMS bearing diagnostics methodologies to enhance functionality and improve the business case for HUMS. From a system perspective, this implementation allowed the detection of a real-world turboshaft bearing fault prior to a catastrophic engine failure. The system requirements used vibration data that was resampled to control for changes in shaft rate to reduce smearing of the spectrum. A spectral estimation algorithm is then performed to improve the measured energy associated with the bearing defect. Bearing energy was calculated using the envelope analysis, where a novel approach is taken to correctly select the best window. Finally, a process for thresholding and alerting was implemented, the results being that the aircraft was grounded at the appropriate time and a repair was effectuated prior to engine failure.

Fault Injection Method and Ground-truth Development to Enable a Low-cost Bearing Fault Monitoring System in the Automotive Industry

Bearing condition monitoring is a widely studied field, but applications to the automotive industry have received little attention as the bearing failure rates are typically low in traditional internal combustion engine vehicles with 200 – 300k mile lifespans. The rapid advancement of electric and autonomous vehicles enables vehicles with million-mile lifespans. This implies that the reliable life of existing bearing designs is exceeded throughout the vehicle life, which can potentially lead to vehicle failure. To enable the development of a bearing fault detection and prognostics system, healthy and faulty bearing data must be collected, and the ground-truth states of the health of bearings need to be determined for algorithm refinement and validation. This work explores the fault injecting options, and ground-truthing together with their limitations. Two methods based on precision machining and seeded spalling are developed and used to inject inner race faults in a ball bearing. A non-invasive ground-truthing method is proposed to quantify the state of health of the fault injected bearings in which bench test data is collected under various speed and load conditions. The vibration signals from the bench tests are used to calculate the root-square of the area under the acceleration Power Spectral Density curve (known as GRMS) for each speed and load condition. To remove the dependency of the results on load and speed conditions, a speed-load-GRMS plot is generated, and a plane is fitted to the data for each fault level. Next, the volume under the plot is calculated, yielding a single cumulative GRMS value for each fault level. This value is used as the ground-truth health of bearing for each fault level. For the bearing with the faults injected using precision machining fault injection, the obtained ground-truth values are 1.56, 3.68, and 4.36 times larger than the same figure for the healthy bearing for the faults with the widths of 0.1 mm, 0.5 mm, and 2 mm, respectively. The observed correlation between the fault sizes and the calculated ground-truth values validates the proposed method which can provide a good separation among different health states of a bearing.

A Qualified review of ML and DL algorithms for Bearing Fault Diagnosis

Moving machinery is the backbone of socio-economic development. The use of machines helps in increasing the production of everyday used items, and tools, that generate electricity and mechanical energy, and provides easy and fast transportation and help by saving human efforts, energy, and time. The mechanical industry is totally dependent on the bearing and it is considered bread and butter of the system. Bearing failure is about 40% of the total failures of induction motors which is why it is a crucial challenge to predict the failure and helps prevent future downtime events through maintenance schedules with the latest techniques and tools of. This paper presents a review of how DL techniques and algorithms outsmarted ML for bearing fault detection and diagnosis and summarizes the accuracy results generated by most common DL algorithms over classical ML algorithms. Additionally this paper reasons different criteria for which DL algorithms have been proved efficient for building productive model in the field of bearing fault detection. Furthermore, some of the most famous datasets by different universities have been discussed and accuracy results are provided by reviewing algorithms on the CWRU dataset by different researchers and comparison chart is listed in the results section.

Bearing fault diagnosis method based on Gramian angular field and ensemble deep learning

Inspired by the successful experience of convolutional neural networks (CNN) in image classification, encoding vibration signals to images and then using deep learning for image analysis to obtain better performance in bearing fault diagnosis has become a highly promising approach. Based on this, we propose a novel approach to identify bearing faults in this study, which includes image-interpreted signals and integrating machine learning. In our method, each vibration signal is first encoded into two Gramian angular fields (GAF) matrices. Next, the encoded results are used to train a CNN to obtain the initial decision results. Finally, we introduce the random forest regression method to learn the distribution of the initial decision results to make the final decisions for bearing faults. To verify the effectiveness of the proposed method, we designed two case analyses using Case Western Reserve University (CWRU) bearing data. One is to verify the effectiveness of mapping the vibration signal to the GAFs, and the other is to demonstrate that integrated deep learning can improve the performance of bearing fault detection. The experimental results show that our method can effectively identify different faults and significantly outperform the comparative approach.

Automatic incipient fault detection and health state assessment of rolling element bearings using pruned exact linear time method

Rolling element bearing is a crucial component in rotating machines. The existence of faults in the bearing causes sudden failures, resulting in catastrophic failure of machines. Incipient fault detection and health state assessment are the essential tasks in condition-based maintenance to avoid machine failures. This paper proposes a new method, the Pruned Exact Linear Time, for identifying the incipient faults and following damage states in the bearing. When a fault initiates in the bearing, there is an increment in the vibration response. This increment can be quantified by the degradation indicator computed from the vibration signal. The Variational Mode Decomposition technique is used to de-noise the vibration signals. Various statistical features are derived from the de-noised signal, and the best feature subset is chosen by the Recursive Feature Elimination method. Then, the significant bearing life degradation indicator is computed using the Reconstruction Independent Component Analysis method by fusing selected features. A novel index is formulated for computing the percentage of failure, which can identify whether the bearing is in a mild failure state or medium failure state. The efficiency of the proposed framework is demonstrated using experimental bearing datasets.

Low-cost DSP-based vibration data acquisition system for bearing fault detection in electric motors

This paper presents a low-cost DSP-based vibration data acquisition system with a high data storage capacity for fault detection in electric motors. An outer race bearing fault detection case study for a three-phase induction machine using vibration signals provided by the DSP-based acquisition system from a test rig is detailed. The fault detection scheme is performed based on Fast Fourier Transform (FFT) and Discrete Wavelet Transform (DWT).

Fault Detection of Wind Turbine Generator Bearing Using Attention-Based Neural Networks and Voting-Based Strategy

Wind turbines are usually exposed to dynamic and adverse environmental conditions, and faults on mechanical parts are major threats to the economic efficiency of wind farms. Condition monitoring and fault detection play important roles in the reliability of wind turbines. However, due to the dynamic characteristics and fluctuations of variables, precise modeling for representing the complex operation state of key components is still challenging. This study presents a data-driven fault detection method for generator bearing of wind turbines. A dual-stage attention-based recurrent neural network is first trained to model the normal behavior of generator bearing temperature. Based on long short-term memory, an input attention mechanism is specially designed to allocate the variable importances specific to each timeslot. Since the variable importance is distinct at different timeslots, the time-varying correlations among variables in wind turbines can be captured, and hence, the performance of variable estimation can be remarkably improved. In addition, this article focuses on the challenges faced by the practice of wind power industries, where the evolution rates of impending faults are usually unknown and distinct, and proposes a novel voting-based detection strategy. The proposed strategy conducts multiple processes to evaluate a potential fault from different perspectives and, hence, can guarantee the robustness and performance of fault detecting. Eventually, field supervisory control and data acquisition data from real wind farms are utilized for validation and extensive comparison studies verify the effectiveness of this proposed scheme.

Rolling Bearing Fault Detection System and Experiment Based on Deep Learning.

The current situation of frequent small-scale accidents shows that the existing methods have not completely solved the problem of bearing failures, and new research methods need to be used to complete the study of bearing failures. To prevent the failure of rolling bearings and meet the need for timely detection of faults, this research is based on deep learning. Using the combination of deep transfer learning and metric learning methods, the identification and analysis of bearing multi-state vibration signals under different working conditions are carried out. The combination of SSAE-based similarity measurement criteria and deep transfer learning can reduce the differences between different domains. It is difficult to distinguish the data samples at the boundary and diagnose the problems that the physical meaning is difficult to understand. Through the bearing fault diagnosis analysis, the validity of the deep learning diagnosis model proposed in this paper is verified. The results show that the detection accuracy of the rolling bearing fault detection method based on LCM-SSAE is 0.6 percentage points higher than that of the rolling bearing fault detection method based on SSAE, which proves that the method is suitable for the fault detection of rolling bearing, and it also shows the effectiveness and robustness of the fault detection system of rolling bearing.

A Survey on Fault Diagnosis of Rolling Bearings

The failure of a rolling bearing may cause the shutdown of mechanical equipment and even induce catastrophic accidents, resulting in tremendous economic losses and a severely negative impact on society. Fault diagnosis of rolling bearings becomes an important topic with much attention from researchers and industrial pioneers. There are an increasing number of publications on this topic. However, there is a lack of a comprehensive survey of existing works from the perspectives of fault detection and fault type recognition in rolling bearings using vibration signals. Therefore, this paper reviews recent fault detection and fault type recognition methods using vibration signals. First, it provides an overview of fault diagnosis of rolling bearings and typical fault types. Then, existing fault diagnosis methods are categorized into fault detection methods and fault type recognition methods, which are separately revised and discussed. Finally, a summary of existing datasets, limitations/challenges of existing methods, and future directions are presented to provide more guidance for researchers who are interested in this field. Overall, this survey paper conducts a review and analysis of the methods used to diagnose rolling bearing faults and provide comprehensive guidance for researchers in this field.

A new adaptive parallel resonance system based on cascaded feedback model of vibrational resonance and stochastic resonance and its application in fault detection of rolling bearings

Accurate extraction of weak feature information in strong background noise is a key to detect and identify rolling bearing faults. Stochastic resonance (SR) and vibrational resonance (VR) have received extensive attention and research in weak signal detection by reason of their advantages of utilizing additional inputs (i.e. noise or high frequency harmonic signals) to enhance weak signals. Considering the advantages and disadvantages of SR and VR in weak signal detection, this paper combines the two to construct a cascaded feedback model of VR and SR, and utilize it to form a parallel resonance system, which improves the detection performance of weak signals through the ensemble average effect. Furthermore, a multi-parameter optimization strategy based on the improved whale optimization algorithm (WOA) is proposed for the parameter selection of the parallel resonance system. It uses the constructed measurement index independent of the prior knowledge as the fitness function to realize automatic adjustment of multi-parameter, and obtains the final output by weighted summation of the optimal results obtained by multiple iterations. Finally, the suggested method is analyzed by numerical simulation signal and experimental data of rolling bearings, and the effectiveness and superiority of the proposed method in the detection of weak fault features are verified.

An iot based bearing fault detection using convolutional neural network

An iot based bearing fault detection using convolutional neural network

Wind turbine blade bearing fault detection with Bayesian and Adaptive Kalman Augmented Lagrangian Algorithm

As a critically supporting and rotational component for wind turbines, blade bearings need special health monitoring for safe operation in actual industrial conditions. One of the main difficulties of the wind turbine blade bearing condition monitoring is noisy signals generated under fluctuating slow speed with heavy loads. This is because blade bearing rotation speed is influenced by blade flipping and external disturbances, and this influence is time-varying. This paper proposes a new method, Bayesian and Adapted Kalman Augmented Lagrangian (BAKAL), to filter the signal under this time-varying condition. The new method uses a two-step search (coarse and fine search) to deal with the filtering process based on Bayesian Augmented Lagrangian (BAL) framework. In addition, both linear and nonlinear effects and their sparsity are considered for model construction. Finally, the smearing problem in the frequency spectrum is dealt with through signal resample in the order domain for superior performance of fault diagnosis. The proposed BAKAL algorithm is strictly validated in several experiments under approximately fixed speed and variable speed within the condition of heavy loadings. The experiments use an industrial and rotational wind turbine blade bearing with natural defects, which has been served in an actual wind power plant for over 15 years. The experimental results demonstrate the effectiveness of the proposed method.

Acoustic detection of bearing faults through fractional harmonics lock-in amplification

Bearing failure detection in induction motors is vital to prevent breakdowns and dead times in industrial processes. Non-invasive detection of bearing faults can be performed by analyzing sound signals and the associated spectral signature. However, measured sound often contains signals from other surrounding sources of vibrations, which can pollute analysis or degrade SNR. This work proposes the use of Lock-In Amplification (LIA) for bearing fault detection in order to overcome SNR limitations and allow specific detection.In practice, LIA is synchronized to the motor shaft frequency through a fractional Phase-locked loop (PLL) frequency synthesizer, delivering characteristic frequencies for bearing faults, which are fractional harmonics of the shaft rotation frequency, and phase-locked to the motor under study.Comparison with Fourier spectrum measurements confirms the superior ability of LIA for extracting small signals from noise, but also allows specific detection, even in the presence of external sources vibrating at the same frequencies as the motor of interest. Additionally, fractional harmonics LIA was able to detect characteristic frequencies from fault-free conditions, paving the way for earlier failure detection and continuous monitoring.This work is the first reported experimental implementation of fractional harmonics LIA for bearing fault detection purposes.

Fault diagnosis of bearings in multiple working conditions based on adaptive time-varying parameters short-time Fourier synchronous squeeze transform

Rolling bearings are commonly used components in rotating machinery and play a vital role. When the bearing fails, if it cannot be found and repaired in time, it will cause great economic losses. Time-frequency analysis has been widely used for bearing fault signals under non-stationary operating conditions, but the existing methods have problems such as poor adaptability under multiple operating conditions. At the same time, the low time-frequency resolution and poor energy aggregation also affect the fault feature extraction effect. Aiming at these problems, this paper proposes a bearing fault detection method, which combines empirical mode decomposition and adaptive time-varying parameter short-time Fourier synchronous squeezing transform (AFSST), it solves the problem of adapting to signals under multiple operating conditions; A weighted least squares estimation time-varying parameter algorithm is proposed, which improves the calculation speed by 29% under the premise of ensuring the calculation accuracy; A time-varying index of energy effective compression ratio is proposed to accurately measure the time-varying energy aggregation of time-frequency analysis methods. Using short-time Fourier transform, continuous wavelet transform, wavelet synchrosqueezed transform, and AFSST to analyze the simulated FM signal, the results show that the AFSST transform has better time-frequency resolution and higher energy-efficient compression rate globally. Through the verification of the fault experimental data of rolling bearings, the diagnosis method proposed in this paper can accurately extract the bearing fault characteristics, has a good diagnosis ability in the multi-working operating environment, and has strong robustness and anti-noise interference.

Application of Teager-Kaiser Energy Operator in the Early Fault Diagnosis of Rolling Bearings.

Rolling bearings are key components that support the rotation of motor shafts, operating with a quite high failure rate among all the motor components. Early bearing fault diagnosis has great significance to the operation security of motors. The main contribution of this paper is to illustrate Gaussian white noise in bearing vibration signals seriously masks the weak fault characteristics in the diagnosis based on the Teager–Kaiser energy operator envelope, and to propose improved TKEO taking both accuracy and calculation speed into account. Improved TKEO can attenuate noise in consideration of computational efficiency while preserving information about the possible fault. The proposed method can be characterized as follows: a series of band-pass filters were set up to extract several component signals from the original vibration signals; then a denoised target signal including fault information was reconstructed by weighted summation of these component signals; finally, the Fourier spectrum of TKEO energy of the resulting target signal was used for bearing fault diagnosis. The improved TKEO was applied to a vibration signal dataset of run-to-failure rolling bearings and compared with two advanced diagnosis methods. The experimental results verify the effectiveness and superiority of the proposed method in early bearing fault detection.

Bearing Vibration Signal Fault Diagnosis Based on LSTM-Cascade CatBoost

<p>Bearing is one of the most concerned parts in the field of fault diagnosis. At present, there are numerous excellent algorithms applied to bearing fault detection. This paper proposes a new fault bearings diagnosis model named LSTM-Cascade CatBoost, which can directly classify bearing vibration signals in the case of multiple granularity and high dimensions without signal processing. The model is based on gcForest, whose complexity can be adjusted automatically to the size of data set and it uses LSTM to extract features of time series signals instead of multi-grained scanning for improving the model’s feature extraction ability. CatBoost is used as the base classifier of cascade forest to improve the classification accuracy of the model. Experimental results show the fact that this model is highly accurate in CWRU and XJTU-SY datasets. Besides, it not only proves that the feature extraction ability of LSTM is significantly better than that of multi-grained scanning, but CatBoost as a base classifier can further improve the accuracy of cascade forest.</p> <p> </p>

Research on Bearing Fault Detection Algorithm Based on Convolution Neural Network and SVM

Research on Bearing Fault Detection Algorithm Based on Convolution Neural Network and SVM

Bearing Fault Detection Based on Improved Multiscale Dispersion Entropy and Single Value Classification

In view of the difficulty in extracting bearing fault feature data and the problem of fault detection affected by the inaccurate determination of classical domains when using Extenics for fault detection. In this paper, a bearing fault detection method based on improved multi-scale dispersion entropy and single value classification is proposed. The improved multi-scale dispersion entropy is used to extract the characteristic parameters representing the running state of the bearing, the Fisher ratio is used for feature selection and dimension reduction, and the matter element models of the four running states of the bearing are established. The classical domains of the characteristic parameters of the four running states of the bearing are determined by using the single value classification method, and the matter elements of the four running states of the bearing are established. The running state of the bearing is determined by calculating the comprehensive correlation value between the samples to be tested and the matter elements in different running states. The results show that the accuracy of the proposed method is 99.25%. In order to verify the effectiveness of the feature extraction method proposed in this paper, the improved multi-scale dispersion entropy is compared with the refined composite multiscale dispersion entropy method. In order to verify the superiority of the classical domain determination method proposed in this paper, the single value classification method is compared with the mathematical statistics method. The experimental results verify the effectiveness of the methods proposed in this paper.

Adaptive Band Extraction Based on Low Rank Approximated Nonnegative Tucker Decomposition for Anti-Friction Bearing Faults Diagnosis Using Measured Vibration Data

Condition monitoring and fault diagnosis are topics of growing interest for improving the reliability of modern industrial systems. As critical structural components, anti-friction bearings often operate under harsh conditions and are contributing factors of system failures. Efforts have been cast on bearing diagnostics under the sensor fusion and machine learning framework, whilst challenges remain open on the identification of incipient faults. In this paper, exploiting multi-way representations and decompositions of measured vibration data, a novel band separation method based on the factorization of spectrogram tensors using the low rank approximated nonnegative Tucker decomposition (LRANTD) is proposed and applied to identify detailed fault signatures from the spectral, temporal, and spatial dimensions, flexible for extracting multi-sensor features and multi-dimensional correlations. With the proposed method, informative frequency bands of the latent vibrational components can be automatically extracted, in accordance with the inherent temporal patterns that can be conveniently fed for spectral analysis and fault discrimination. Furthermore, an improved cross-spectrum can be calculated from multi-channel vibrations via LRANTD with enhanced fault features. Based on the real-world vibration data of the accelerated bearing life tests, detailed experimental studies and thorough comparisons to the conventional benchmarks have verified the effectiveness of the reported diagnostic methodology. The proposed method significantly improves the presence of the bearing frequency peaks distinctly over the background noises in the spectrum and hence improves the bearing defect detection process.