Rolling Bearing Fault Detection Research Articles

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.

Rolling bearing fault detection method based on teager energy operator

Abstract This study presents a method for detecting subtle faults in rolling bearings by integrating singular value decomposition (SVD) and local modal decomposition (LMD). Initially, the original vibration signals undergo reconstruction in phase space and subsequent denoising using SVD. Following this, LMD is employed to generate multiple local modal functions (IMFs), with the most sensitive IMF components selected based on kurtosis-mean square deviation criteria. The Teager energy operator (TEO) is then computed for these selected IMFs. Ultimately, Fourier transform analysis of TEO is utilized to identify faults in the inner and outer races of rolling bearings. Test results conducted on publicly available data from Case Western Reserve University validate the accurate identification of inner and outer race faults using this approach.

Fault diagnosis algorithm based on GADF-DFT and multi-kernel domain coordinated adaptive network

To address the problems of low detection accuracy of rolling bearings under different loads and the difficulty of effectively identifying the lack of labelled data, a rolling bearing fault diagnosis method combining GADF-DFT image coding and Multi-kernel domain coordinated adaptation network is proposed. Firstly, the vibration signal is converted into a two-dimensional image using GADF coding technology, and then the GADF image is converted into the frequency domain using discrete Fourier transform to extract deeper feature information. Combined with the multi-source domain adaptive method, the public feature extraction module is used to initially achieve feature mining of the image; the MK-MMD algorithm of the domain-specific adaptive module reduces the difference in feature distribution between the source and target domains; and the final classification difference minimization module reduces the problems caused by the classification errors that may be generated by the different domain classifiers due to the fact that the data samples are located near the category boundaries. The test uses the Case Western Reserve University dataset and divides the dataset with different operating conditions as the source and target domains, and the test results show that the proposed model demonstrates its effectiveness in responding to the complex operating condition changes in rolling bearing fault detection in multiple operating condition migration tasks, good adaptability and robustness, and is able to achieve accurate fault diagnosis under different operating conditions.

Early fault detection for rolling bearings: A meta‐learning approach

AbstractEarly fault detection (EFD) of rolling bearings aims at detecting the early symptoms of faults by monitoring small deviations of health states. Accurate EFD enables predictive maintenance and contributes to the stability of mechanical systems. In recent years, machine learning based methods have shown impressive performance on EFD. Most of the current machine learning‐based methods assume the availability for a large amount of data. However, in practice, the authors may only have a very limited amount of training data, which makes it hard to learn a reliable machine learning model. To address this concern, in this work, the authors propose to tackle EFD via meta learning. Specifically, the authors first formulate EFD as a few‐shot learning problem and then propose to tackle this problem with a metric‐based meta learning method. Furthermore, ensemble learning is further leveraged to improve the detection robustness. For the proposed method, the distribution difference from the working conditions and the bearings are considered. The experimental results on two bearing datasets show that the proposed method can achieve better EFD performance, that is, detecting incipient faults earlier while bringing in lower false alarms, compared with several frequently used EFD methods.

Bearing fault detection technology for automated machinery based on acoustic analysis

Traditional fault detection methods in acoustic signal feature extraction of rolling bearings often make the signal denoising process complex due to low signal-to-noise ratio and weak fault features, making this method difficult to meet real-time requirements. Therefore, a fault detection model based on Fast-Renoriented SIFT feature extraction is proposed, which can quickly extract a large number of features from the original signal without the need for noise reduction processing and can effectively improve the efficiency and accuracy of fault detection. At the same time, to adapt to the fault detection of rolling bearings under multiple working conditions, this study also proposes an adaptive extended word bag model that combines local kurtosis and local 2-dimensional information entropy features, improving the adaptability and flexibility of the new model. It obtained a 100% overall recognition rate and a fault detection time of no more than 0.5 seconds in a 5-fold cross-validation experiment, verifying the excellent recognition accuracy, stability, and operational efficiency of the detection model. Its recognition accuracy in the multi-working condition rolling bearing fault detection experiment was above 97%, which was improved by about 21.25% compared to the traditional word bag model and had significant advantages in fault recognition accuracy and computational efficiency.

Unknown fault detection of rolling bearings guided by global–local feature coupling

Fault diagnosis technology can effectively prevent the occurrence of faults and reduce safety hazards, which is of great significance in nuclear power, aerospace, manufacturing, and other fields. Given the stringent demands of safe and reliable equipment operation in practical production environments, acquiring a comprehensive set of fault samples becomes challenging. At present, many deep learning-based methods have been researched on this problem. However, these methods do not account for the identification of novel faults that may emerge. In this paper, we propose a novel global and local feature joint learning method for unknown fault detection, which addresses this problem by applying the knowledge learned by the supervised feature extraction process to the unsupervised clustering process. In particular, we propose a dual-branch framework for detecting unknown faults, which is based on multi-scale coupled feature extraction. This framework establishes correlations between features at different scales and employs the coupling of global and local features to facilitate the detection of unknown faults. Additionally, we propose a causal modeling method for global and local features, aiming to uncover the true causal relationship among global and local features and fault categories. Moreover, we propose a consistent prediction method to ensure the coherence of prediction results between the global and local branches. We evaluate the performance of our model using the CWRU, PU, and RB datasets, and the results demonstrate its superiority over state-of-the-art methods in terms of clustering accuracy and normalized mutual information.

Development of Diagnosis System for Rolling Bearings Faults on Real Time Based on FPGA

The real-time monitoring of events in an industrial plant is an advanced technique that presents the real conditions of operation of the machinery responsible for the manufacturing process. A predictive maintenance program includes various rotating machinery condition monitoring techniques of the machine to determine the conditions of failure. To increase the operational reliability and to reduce preventive maintenance, it is necessary an efficient tool for analysis and process monitoring, in real time, enabling the detection of incipient faults for rolling bearings. Over the past few years there has been a major technological developments related to digital system, including innovations in both hardware and software. These innovations enable the development of new design methodologies that take into account the ease of future modifications, upgrades and expansions of the designed system. This paper presents a study of new design tools for embedded digital systems based on open hardware architecture with reconfigurable logic. Will be discussed a case study in the area of fault detection of rolling bearings, as well as its implementation and testing.

Deep learning based on rolling bearing fault diagnosis method

With the continuous development of industrial automation, rolling bearings play a crucial role in many fields as key mechanical components, and the fault diagnosis of rolling bearings have great significance. This paper discusses a deep learning based rolling bearing fault diagnosis method, aiming to improve the accuracy and efficiency of fault detection. Firstly, the vibration signals of rolling bearings are pre-processed to extract the feature information that helps fault diagnosis. Then, the features were automatically learned and classified by using BP neural network. Finally, the effectiveness and robustness of the method were verified through experiments. Compared with the traditional fault diagnosis method, the deep learning-based rolling bearing fault diagnosis method has higher accuracy and practicality, which provides strong support for the fault detection and preventive maintenance of rolling bearings.

Weak fault feature extraction of rolling bearing based on multi-system coupled cascaded stochastic resonance system

The accurate extraction of weak signal features under strong noise background plays a crucial role in the fault detection of rolling bearings. In order to promote the ability of stochastic resonance (SR) system to detect weak signals and improve the output performance of the system, a multi-system coupled cascaded SR (MCCSR) system is investigated and applied to the fault detection of rolling bearings. Firstly, a MCCSR system is constructed by exploiting the positive synergistic effect between multiple systems, which consists of a triangular-topology coupled system composed of three SR subsystems and a cascaded SR system with topology output as input. This system makes full use of the advantages of coupled system and cascaded system in weak signal detection. In terms of parameter optimization, a stepwise multi-parameter optimization strategy is proposed, which adopts different optimization methods for different parameters, and avoids the inconsistency between error and step factor by improving the least mean square algorithm. Finally, through the comparative analysis of numerical simulation and experimental signals, it is verified that the proposed method can effectively enhance the weak signal features and improve the system output signal-to-noise ratio, which can better serve for rolling bearing fault detection.

A Study of Noise Effect in Electrical Machines Bearing Fault Detection and Diagnosis Considering Different Representative Feature Models

As the field of fault diagnosis in electrical machines has significantly attracted the interest of the research community in recent years, several methods have arisen in the literature. Also, raw data signals can be acquired easily nowadays, and, thus, machine learning (ML) and deep learning (DL) are candidate tools for effective diagnosis. At the same time, a challenging task is to identify the presence and type of a bearing fault under noisy conditions, especially when relevant faults are at their incipient stage. Since, in real-world applications and especially in industrial processes, electrical machines operate in constantly noisy environments, a key to an effective approach lies in the preprocessing stage adopted. In this work, an evaluation study is conducted to find the most suitable signal preprocessing techniques and the most effective model for fault diagnosis of 16 conditions/classes, from a low-workload (computational burden) perspective using a well-known dataset. More specifically, the reliability and resiliency of conventional ML and DL models is investigated here, towards rolling bearing fault detection, simulating data that correspond to noisy industrial environments. Diverse preprocessing methods are applied in order to study the performance of different training methods from the feature extraction perspective. These feature extraction methods include statistical features in time-domain analysis (TDA); wavelet packet decomposition (WPD); continuous wavelet transform (CWT); and signal-to-image conversion (SIC), utilizing raw vibration signals acquired under varying load conditions. The noise effect is examined and thoroughly commented on. Finally, the paper provides accumulated usual practices in the sense of preferred preprocessing methods and training models under different load and noise conditions.

Early Fault Detection of Rolling Bearings Based on Time-Varying Filtering Empirical Mode Decomposition and Adaptive Multipoint Optimal Minimum Entropy Deconvolution Adjusted.

Due to the early formation of rolling bearing fault characteristics in an environment with strong background noise, the single use of the time-varying filtering empirical mode decomposition (TVFEMD) method is not effective for the extraction of fault characteristics. To solve this problem, a new method for early fault detection of rolling bearings is proposed, which combines multipoint optimal minimum entropy deconvolution adjusted (MOMEDA) with parameter optimization and TVFEMD. Firstly, a new weighted envelope spectrum kurtosis index is constructed using the correlation coefficient and envelope spectrum kurtosis, which is used to identify the effective component and noise component of the bearing fault signal decomposed by TVFEMD, and the intrinsic mode function (IMF) containing rich fault information is selected for reconstruction. Then, a new synthetic impact index (SII) is constructed by combining the maximum value of the autocorrelation function and the kurtosis of the envelope spectrum. The SII index is used as the fitness function of the gray wolf optimization algorithm to optimize the fault period, T, and the filter length, L, of MOMDEA. The signal reconstructed by TVF-EMD undergoes adaptive filtering using the MOMEDA method after parameter optimization. Finally, an envelope spectrum analysis is performed on the signal filtered by the adaptive MOMEDA method to extract fault feature information. The experimental results of the simulated and measured signals indicate that this method can effectively extract early fault features of rolling bearings and has good reliability. Compared to the classical FSK, MCKD, and TVFEMD-MOMEDA methods, the first-order correlated kurtosis (FCK) and fault feature coefficient (FFC) of the filtered signal obtained using the proposed method are the largest, while the sample entropy (SE) and envelope spectrum entropy (ESE) are the smallest.

An enhanced K-SVD denoising algorithm based on adaptive soft-threshold shrinkage for fault detection of wind turbine rolling bearing

Due to nonstationary operating conditions of wind turbines and surrounding harsh working environments, the impulse features induced by bearing faults are always overwhelmed by heavy noise, which brings challenges to accurately detect rolling bearing faults. Sparse representation exhibits excellent performance in nonstationary signal analysis, but it is closely bound up with the degree of similarity between the atoms in a dictionary and signals. Therefore, this paper investigates an enhanced K-SVD denoising method based on adaptive soft-threshold shrinkage to achieve high-precision extraction of impulse signals, and applies it to fault detection of generator bearing of wind turbines. An adaptive sparse coding shrinkage soft-threshold denoising is first proposed to remove noise and harmonic interference in the residual term of dictionary updating, so that the updated atoms show obvious impact characteristics. Furthermore, a soft-threshold shrinkage function with adaptive threshold is designed to further suppress clutter in atoms of the learned dictionary, so as to obtain an optimized dictionary for recovering impulse signals. Two actual engineering cases are selected for analysis, and the envelope spectrum correlation kurtosis corresponding to the results obtained by the proposed method is significantly higher than that of other comparison methods, thus verifying its superiority in detecting rolling bearing faults.

Rolling Bearing Fault Detection in the Range of Ultrasound

Rolling Bearing Fault Detection in the Range of Ultrasound

A systematic review of rolling bearing fault diagnoses based on deep learning and transfer learning: Taxonomy, overview, application, open challenges, weaknesses and recommendations

Rolling bearing fault detection is critical for improving production efficiency and lowering accident rates in complicated mechanical systems, as well as huge monitoring data, posing significant challenges to present fault diagnostic technology. Deep Learning is now an extraordinarily popular research topic in the field and a promising approach for detecting intelligent bearing faults. This paper aims to give a comprehensive overview of Deep Learning (DL) based on bearing fault diagnosis. The most widely used DL algorithms for detecting bearing faults include Convolutional Neural Network, Recurrent neural network, Autoencoder, and Generative Adversarial Network. It discusses a variety of transfer learning architectures and relevant theories while summarises, classifies, and explains several publications on the subject. The research area’s applications and problems are also addressed.

Spectrum Shape Based Roller Bearing Fault Detection and Identification

This article describes a combination of numerical methods of automated fault detection and identification in rolling bearings, even when there is a limited knowledge about inspected bearings and their characteristic frequencies in particular. The most important feature is global approach to solving the stated problem using several rolling bearing diagnostic methods and techniques which work simultaneously and complement each other. These include the well-known Hilbert-based envelope, Discrete Wavelet Transform and Fourier Transform as well as new Outer Envelope Filtering, Pulse Train Scanning, spectrum shape analysis and novel application of known signal processing tools like Fuzzy Logic and Continuous Wavelet Transform. Combination of partial results obtained by each path allows for judging the probabilities of different fault types (outer-race, inner-race and roller). Two techniques additionally used here are spectrum scanning by a train of harmonic pulses (Pulse Train Scanning) and an outer envelope filter which removes high frequency noise, but enhances the most important short time diagnostic symptoms, which are often removed by standard filtering. A combination of such solutions allows for unambiguous and precise localization of characteristic fault frequencies in a spectrum without initial knowledge about their expected values.

A Nonparametric Cumulative Sum-Based Fault Detection Method for Rolling Bearings Using High-Level Extended Isolated Forest

As an advanced anomaly detection method, extended isolated forest (EIF) has great potential for use in bearings. A low-dimensional data transformation method based on EIF with high extension level and an early fault detection strategy for rolling bearings based on feature combination are proposed due to the limitation of EIF extension level on low-dimensional data. First, a preliminary sequence base is generated through sliding windows, and the sequences related to the number of transformed dimensions are selected based on the dynamic time warped (DTW) distance of the original sequence to perform weighted DTW barycenter averaging (WDBA), which enables preprocessing of the maximum extension level of EIF. Then, to realize early fault detection of rolling bearings, the nonparametric cumulative sum (NCUSUM) algorithm is used to design a joint threshold discrimination scheme for root mean square (rms) and kurtosis of the new and original sequences. We tested the algorithm on fault simulation using FEMTO-ST and XJTU-SY bearing datasets. The results show that WDBA-EIF algorithm can better remove uncorrelated noise from time series at high extension level. Compared with several anomaly detection methods, under 60% datasets, it has the best detection result of an early weak anomaly in the running process of rolling bearings and has a low false alarm rate.

Deep Domain-Adversarial Anomaly Detection With One-Class Transfer Learning

Despite the big success of transfer learning techniques in anomaly detection, it is still challenging to achieve good transition of detection rules merely based on the preferred data in the anomaly detection with one-class classification, especially for the data with a large distribution difference. To address this challenge, a novel deep one-class transfer learning algorithm with domain-adversarial training is proposed in this paper. First, by integrating a hypersphere adaptation constraint into domain-adversarial neural network, a new hypersphere adversarial training mechanism is designed. Second, an alternative optimization method is derived to seek the optimal network parameters while pushing the hyperspheres built in the source domain and target domain to be as identical as possible. Through transferring one-class detection rule in the adaptive extraction of domain-invariant feature representation, the end-to-end anomaly detection with one-class classification is then enhanced. Furthermore, a theoretical analysis about the model reliability, as well as the strategy of avoiding invalid and negative transfer, is provided. Experiments are conducted on two typical anomaly detection problems, i.e., image recognition detection and online early fault detection of rolling bearings. The results demonstrate that the proposed algorithm outperforms the state-of-the-art methods in terms of detection accuracy and robustness.

CORRIGENDUM: Rolling bearing fault detection of electric motor using time domain and frequency domain features extraction and ANFIS

CORRIGENDUM: Rolling bearing fault detection of electric motor using time domain and frequency domain features extraction and ANFIS

Adaptive Swarm Decomposition Algorithm for Compound Fault Diagnosis of Rolling Bearings

The feature extraction of compound faults is still considered the bottle neck task of machinery fault diagnosis. In this article, a novel adaptive swarm decomposition (ASWD) algorithm based on fine to coarse (FTC) segmentation is proposed for compound fault detection of rolling bearings. Firstly, the number of oscillating components that affects the results of ASWD is automatically determined by the order statistics filter and energy spectrum segmentation method without any prior knowledge. Secondly, the Teager energy kurtosis (TEK) of successively extracted components is employed as the indicator to evaluate the effectiveness of iterations. This not only setups the swarm decomposition (SWD) threshold, but also improves the performance of periodic impulses separation. Finally, ASWD is applied to intelligently separate the different oscillating components and suppress the redundant decomposition. The testing results of ASWD on the simulation and real cases indicate that ASWD can effectively extract compound fault impulses from multicomponent vibration signals. The comparison between SWD and other decomposition methods further verifies the superiority of ASWD. The characteristic frequency intensity coefficient (CFIC) of ASWD is increased by 34.2%, 49.2%, and 56.5% in the three cases, respectively, than SWD, variational mode decomposition (VMD), and ensemble empirical mode decomposition (EEMD).

Impact of noise model on the performance of algorithms for fault diagnosis in rolling bearings

Condition monitoring of rolling bearings is attracting much interest since most of the production slowdowns depends on the damaging of these components. Several algorithms for fault detection have appeared in the technical literature in the last decade. In most cases, performance is assessed over both synthetic and experimental data. Unfortunately, the computer simulations adopt signal models that are trivial and are not able to predict the actual performance on the field. In this work we propose a framework suitable to fairly, quantitatively and objectively compare different algorithms for fault detection in rolling bearings. The vibration signal is obtained through computer simulations. The signal entailed by the damage is generated through the model at “impact-delay-line” already available in the technical literature. The machine noise is generated as a wideband component with the possible superposition of narrowband components. The wideband component has been modeled as additive white Gaussian noise, additive white noise drawn from an α-stable distribution and additive noise stemming from an autoregressive process. Narrowband components are modeled through trains of Gaussian pulses. The performance of three well known algorithms for fault detection are compared in terms of capability in identifying the theoretical cyclic frequencies related to a damage. In these scenarios the behavior of fault detectors are definitely far from that predicted by classical wideband noise models like, for instance, additive white Gaussian noise.