Case Western Reserve University Bearing Research Articles

Bearing Intelligent Fault Diagnosis in the Industrial Internet of Things Context: A Lightweight Convolutional Neural Network

The advancement of Industry 4.0 and Industrial Internet of Things (IIoT) has laid more emphasis on reducing the parameter amount and storage space of the model in addition to the automatic and accurate fault diagnosis. In this case, this paper proposes a lightweight convolutional neural network (LCNN) method for intelligent fault diagnosis of bearing, which can largely satisfy the need of less parameter amount and storage space as well as high accuracy. First, depthiwise separable convolution is adopted, and a LCNN structure is constructed through an inverse residual structure and a linear bottleneck layer operation. Secondly, a novel decomposed Hierarchical Search Space is introduced to automatically explore the optimal LCNN for bearing fault diagnosis in the context of the IIoT. In the meantime, the real-time monitoring and fault diagnosis of the model are also deployed. In order to verify the validity of the designed model, Case Western Reserve University Bearing fault dataset and MFPT bearing fault dataset are adopted. The results demonstrate the great advantages of the model. The LCNN model can automatically learn and select the appropriate features, highly improving the fault diagnosis accuracy. Meanwhile, the computational and storage costs of the model are largely reduced, which contributes to its being applied to the mechanical system in the IIoT context.

Fault Diagnosis Based on Non-Negative Sparse Constrained Deep Neural Networks and Dempster–Shafer Theory

Fault diagnosis is an important technology to ensure the safe and reliable operation of equipment. Deep learning driven by big data brings new opportunities for fault diagnosis. Due to the diversity and complexity of the actual fault data distribution, a fault diagnosis algorithm based on non-negative sparse constrained deep neural networks (NSCDNN) and Dempster-Shafer theory (DST) is proposed in this paper. The deep neural network is trained by non-negative constraint and sparse constraint, which can learn part-based representation of fault data. The improved DST is combined with the classification confidence and accuracy of NSCDNN model, which can deal with the uncertainty of information from different sensors. Experimental results of the data provided by Case Western Reserve University Bearing Data Center show that the proposed NSCDNN-DST algorithm can improve the accuracy of fault diagnosis effectively.

Bearing Fault Feature Selection Method Based on Weighted Multidimensional Feature Fusion

Rolling bearing is one of the most critical components in rotating machinery, so in order to efficiently select features, reduce feature dimensions and improve the correctness of fault diagnosis, a feature selection and fusion method based on weighted multi-dimensional feature fusion is proposed. Firstly, features are extracted from different domains to constitute the original high-dimensional feature set. Considering the large number of invalid and redundant features contained in such original feature set, a feature selection process that combines with support vector machine (SVM) single feature evaluation, correlation analysis and principal component analysis-weighted load evaluation (PCA-WLE) is put forward in this paper for selecting sensitive features. The selected features are weighted and fused according to their sensitivity so as to further weaken the interference of low important features. Finally, this process is applied to the data provided by the Case Western Reserve University Bearing Data Center and Xi'an Jiaotong University School of Mechanical Engineering, respectively, and the fault is diagnosed by using the particle swarm optimization-support vector machine (PSO-SVM). The results show that this method can accurately identify different fault categories and degrees of bearing, which is superior and practical than single-domain fault diagnosis with higher recognition ability.

Rolling Bearing Fault Diagnosis Based on the Coherent Demodulation Model

The rolling bearing is an important part of rotary machineries and the health state monitoring is currently a hot research topic. There are four typical failure modes of a rolling bearing: the inner race failure, the outer race failure, the rolling element failure and the support frame failure. Based on the principle of the coherent demodulation, a new method for rolling bearing fault diagnosis is proposed in this work. This method establishes a coherent demodulator simulation model which is usually adopted in the field of radio communications. The proposed method first demodulates the vibration signal with coherent demodulation principle. Then, the characteristic frequencies related the different failures modes are extracted to obtain the fault characteristic amplitudes, to diagnose the ongoing failure mode. The bearing fault data from the Case Western Reserve University Bearing Test Center is taken as an example to verify and analyze the effectiveness of the proposed method. The results show that this method can accurately diagnose the corresponding failure mode with different locations and sizes.

Rolling Element Bearing Fault Diagnosis using Empirical Mode Decomposition and Hjorth Parameters

The condition of rolling element bearing can be monitored by analyzing vibration signatures of machines at regular intervals. Many complex and computationally intensive bearing fault diagnosis schemes have been proposed in the past, extracting various features of vibration signals in time, frequency and time-frequency domains. This paper presents a time efficient and easy to implement scheme to diagnose bearing faults using Hjorth Parameters in time domain. The proposed scheme is experimentally validated on Case Western Reserve University Bearing Data Set, that comprises of normal faultless vibration signals and vibration signals with inner race, outer race and ball bearing faults. The feature vector is computed by calculating Hjorth parameters of representative Intrinsic Mode Functions obtained through Empirical mode decomposition of raw vibration signals. The data is then classified using four Rule-based classifiers. It is observed that PART classifier exhibits the best performance with an accuracy of 96.14% and 93.82 % for training and test datasets respectively. This paper proposes and validates that Hjorth Parameters, when used with Rule based machine learning algorithms, can be effectively used as fault sensitive features for rolling element bearing fault diagnosis.

Application of a New Enhanced Deconvolution Method in Gearbox Fault Diagnosis

When the mechanical transmission mechanism fails, such as gears and bearings in the gearbox, its vibration signal often appears as a periodic impact. Considering the influence of noise, however, the fault signal is often submerged in the noise, so it is necessary to propose a feasible and effective fault extraction method. MOMEDA (multipoint optimal minimum entropy deconvolution adjusted) overcomes the tedious iterative process of MED (minimum entropy deconvolution) and overcomes the resampling trouble in MCKD (maximum correlated kurtosis deconvolution). It is suitable for dealing with periodic impact signal. Besides, aiming at the poor ability of MOMEDA to capture the deconvolution result of target function in a strong noise environment, this paper proposes an improved MOMEDA gearbox fault feature extraction method. Considering that MOMEDA has poor anti-noise performance and can easily cause misdiagnosis in a strong noisy environment, this paper constructs an autoregressive mean sliding model to improve the noise immunity of MOMEDA. Firstly, the stability of the test signal is judged by the autocorrelation coefficient (ACF) and the partial correlation coefficient (PACF). Secondly, the ARMA (autoregressive moving average) model is constructed and a set of optimal model coefficients are obtained to filter the signal, which greatly improves MOMEDA’s ability to capture fault features. Thirdly, the fault feature is extracted by MOMEDA, and the fault information is extracted accurately under a strong noise environment. Finally, compared with AR-MED, ARMAMED, and other methods, the advantages of ARMAMOMEDA are verified. Moreover, the effectiveness and superiority of the proposed method are verified by simulation signals and experimental data from the Case Western Reserve University Bearing Data Center.

A sparse stacked denoising autoencoder with optimized transfer learning applied to the fault diagnosis of rolling bearings

Fault diagnosis is an important technology in the development of modern industrial safety. Vibration information is commonly used to determine the state of bearings. Driven by big data, deep learning brings new opportunities to fault diagnosis. As an unsupervised deep learning algorithm, a stacked autoencoder (SAE) can relieve the pressure of labelling data. Due to the diversity and variability of the actual fault diagnosis distribution, an optimized transfer learning (TL) algorithm is proposed to solve the domain adaptation. By directly inheriting features obtained from the pre-training process in the source domain and changing only the fine-tuning process, the complexity of the algorithm is reduced. Considering the data reconstruction ability and robustness, a sparse stacked denoising autoencoder (SSDAE) is proposed for feature extraction, which can indirectly improve the diagnostic accuracy in the target domain. The results for data from the Case Western Reserve University Bearing Data Center show that the proposed SSDAE-TL algorithm is feasible and easy to implement for the fault diagnosis of bearings.

An Intelligent Detection System Development for Local Faults in a Ball Bearing

Local faults can be produced in ball bearings during their manufacturing process. An efficient, fast and accurate local fault detection method can help improve the quality of ball bearings. To overcome this problem, an intelligent detection system for a ball bearing with the local faults is developed based on the NI LabVIEW software. This system includes the determination of bearing fault parameters, signal acquisition, envelope analysis, time-domain parameter analysis and bearing fault status modules. In this system, the frequency-domain feature method is based on the envelope demodulation analysis, the effective statistical indexes, and the Pearson correlation coefficient. The frequency-domain feature method is used to determine the threshold range for each fault level in the system. This system can in turn be used to determine the fault location and sizes for the ball bearings. A case study for the calculation and analysis for the frequency and time-domain acceleration is presented to predict the location and size of the local faults in a ball bearing. The test data from the Case Western Reserve University Bearing Data Center is used to verify the developed intelligent detection system for local faults in the ball bearing. The results show that the proposed detection system can be used to detect the local fault in the ball bearings.

Adaptive Multiclass Mahalanobis Taguchi System for Bearing Fault Diagnosis under Variable Conditions.

Bearings are vital components in industrial machines. Diagnosing the fault of rolling element bearings and ensuring normal operation is essential. However, the faults of rolling element bearings under variable conditions and the adaptive feature selection has rarely been discussed until now. Thus, it is essential to develop a practicable method to put forward the disposal of the fault under variable conditions. Considering these issues, this paper uses the method based on the Mahalanobis Taguchi System (MTS), and overcomes two shortcomings of MTS: (1) MTS is an effective tool to classify faults and has strong robustness to operating conditions, but it can only handle binary classification problems, and this paper constructs the multiclass measurement scale to deal with multi-classification problems. (2) MTS can determine important features, but uses the hard threshold to select the features, and this paper selects the proper feature sequence instead of the threshold to overcome the lesser adaptivity of the threshold configuration for signal-to-noise gain. Hence, this method proposes a novel method named adaptive Multiclass Mahalanobis Taguchi system (aMMTS), in conjunction with variational mode decomposition (VMD) and singular value decomposition (SVD), and is employed to diagnose the faults under the variable conditions. Finally, this method is verified by using the signal data collected from Case Western Reserve University Bearing Data Center. The result shows that it is accurate for bearings fault diagnosis under variable conditions.

Visibility Graph Feature Model of Vibration Signals: A Novel Bearing Fault Diagnosis Approach.

Reliable fault diagnosis of rolling bearings is an important issue for the normal operation of many rotating machines. Information about the structure dynamics is always hidden in the vibration response of the bearings, and it is often very difficult to extract them correctly due to the nonlinear/chaotic nature of the vibration signal. This paper proposes a new feature extraction model of vibration signals for bearing fault diagnosis by employing a recently-developed concept in graph theory, the visibility graph (VG). The VG approach is used to convert the vibration signals into a binary matrix. We extract 15 VG features from the binary matrix by using the network analysis and image processing methods. The three global VG features are proposed based on the complex network theory to describe the global characteristics of the binary matrix. The 12 local VG features are proposed based on the texture analysis method of images, Gaussian Markov random fields, to describe the local characteristics of the binary matrix. The feature selection algorithm is applied to select the VG feature subsets with the best performance. Experimental results are shown for the Case Western Reserve University Bearing Data. The efficiency of the visibility graph feature model is verified by the higher diagnosis accuracy compared to the statistical and wavelet package feature model. The VG features can be used to recognize the fault of rolling bearings under variable working conditions.

Condition monitoring of critical mechanical elements through Graphical Representation of State Configurations and Chromogram of Bands of Frequency

Fault detection is a crucial aspect to avoid catastrophic failures on mechanical systems, as well as to save money for companies. Currently, a number of non-destructing tests, signal processing analysis and artificial intelligence techniques are used for processing larger and larger amounts of maintenance data in all industry fields, either independently or combined. This manuscript presents a novel methodology for the condition monitoring of machinery, based on vibration analysis. The methodology is supported on two novel signal processing techniques: Graphical Representation of State Configurations (GRSC) and Chromogram of Bands of Frequency (CBF). These two new techniques apply basic concepts of the machine deterioration theory to the frequency spectrum. In order to prove the successful of the work presented, the methodology is tested against two real examples: vibration signals from the Case Western Reserve University (CWRU) Bearing Data Centre, and vibration signals from a high-speed train in normal operation. The results show that these new techniques can process large amounts of data without using artificial intelligence, identify adequately the operating condition of the tested systems and give precise information about that operating system by means of simple graphs and colours.

Bearing fault detection in varying operational conditions using wavelet packet decomposition and support vector machine

Roller bearing is one of the most important elements in machine components. In generally, the bearing can’t work in a steady condition. In this paper, we propose a method based on wavelet packet transform (WPT) and auto regressive (AR) model spectrum to extract fault features in varying operational conditions, and use support vector machine (SVM) to set an effective pattern recognition model. Our bearing experiment data comes from Case Western Reserve University Bearing Data Center. According to the result, WPT, AR model spectrum and SVM can solve the problem of fault diagnosis in varying operational conditions.

Rolling Bearing Fault Diagnosis Based on an Improved Denoising Method Using the Complete Ensemble Empirical Mode Decomposition and the Optimized Thresholding Operation

Vibration signals are widely used in monitoring and diagnosing of rolling bearing faults. These signals are usually noisy and masked by other sources, which may therefore result in loss of information about the faults. This paper proposes an improved denoising method in order to enhance the sensitivity of kurtosis and the envelope spectrum for early detection of rolling bearing faults. The proposed method is based on a complete ensemble empirical mode decomposition with an adaptive noise (CEEMDAN) associated with an optimized thresholding operation. First, the CEEMDAN is applied to the vibration signals to obtain a series of functions called the intrinsic mode functions (IMFs). Second, an approach based on the energy content of each mode and the white noise characteristic is proposed to determine the trip point in order to select the relevant modes. By comparing the average energy of all the unselected IMFs with the energy of each selected IMFs, the singular IMFs are selected. Third, an optimized thresholding operation is applied to the singular IMFs. Finally, the kurtosis and the envelope spectrum are used to test the effectiveness of the proposed method. Different experimental data of the Case Western Reserve University Bearing Data Center are used to validate the effectiveness of the proposed method. The obtained experimental results illustrate well the merits of the proposed method for the diagnosis and detection of rolling bearing faults compared to those of the conventional method.

Fault detection of rolling element bearings using optimal segmentation of vibrating signals

Abstract Change detection and diagnosis are important research directions and activities in the field of system engineering and conditional maintenance of equipments and industrial processes. The paper promotes a new method for change detection and optimal segmentation of vibrating data obtained during operation of rolling element bearings (REB). After a description of the bearing faults and dynamic simulation of REB, the paper makes a review of the change detection and segmentation approaches, that could be used in REB fault detection and diagnosis. A new approach for change detection and optimal segmentation of vibrating signals, aiming to determine the change points in signals generated by the faults, produced during REB operating, is presented; the efficiency of the segmentation method is proven using Monte Carlo simulations for different signal models, including models with changes in the mean, in FIR, and AR model parameters, frequently used in processing vibrating signals. In the final part, the paper analyses some experimental results obtained using this approach and data from the Case Western Reserve University Bearing Data Center.

A Semi-Supervised Approach to Bearing Fault Diagnosis under Variable Conditions towards Imbalanced Unlabeled Data.

Fault diagnosis of rolling element bearings is an effective technology to ensure the steadiness of rotating machineries. Most of the existing fault diagnosis algorithms are supervised methods and generally require sufficient labeled data for training. However, the acquisition of labeled samples is often laborious and costly in practice, whereas there are abundant unlabeled samples which also imply health information of bearings. Thus, it is worthwhile to develop semi-supervised methods of fault diagnosis to make effective use of the plentiful unlabeled samples. Nevertheless, considering the normal data are much more than the faulty ones, the problem of imbalanced data exists among unlabeled samples for fault diagnosis. Besides, in practice, bearings often work under uncertain and variable operation conditions, which would also have negative influence on fault diagnosis. To solve these issues, a novel hybrid method for bearing fault diagnosis is proposed in this paper: (1) Inspired by visibility graph, a novel fault feature extraction method named visibility graph feature (VGF) is proposed. The obtained features by VGF are natively insensitive to variable conditions, which has been validated by a simulation experiment in this paper; (2) On basis of VGF, to deal with imbalanced unlabeled data, graph-based rebalance semi-supervised learning (GRSSL) for fault diagnosis is proposed. In GRSSL, a graph based on a weighted sparse adjacency matrix is constructed by the k-nearest neighbors and Gaussian Kernel weighting algorithm by means of the samples. Then, a bivariate cost function over classification and normalized label variable is built up to rebalance the importance of labels. Finally, the proposed VGF-GRSSL method was verified by data collected from Case Western Reserve University Bearing Data Center. The experiment results show that the proposed method of bearing fault diagnosis performs effectively to deal with the imbalanced unlabeled data under variable conditions.

A novel classification method combining adaptive local iterative filtering with singular value decomposition for fault diagnosis

As a novel time-frequency analysis method, adaptive local iterative filtering (ALIF) can decompose the time series into several stable components which contain the main fault information. In addition, the amplitude of singular value obtained by singular value decomposition (SVD) can reflect the energy distribution. Naturally, there are certain differences in the energy produced by different faults such as the broken tooth, wearing and normal. Thus, a novel method of mechanical fault classification method based on adaptive local iterative filtering and singular value decomposition is proposed in this paper. Firstly, ALIF method decomposed the original vibration signal into a number of stable components to establish an initial feature vector matrix. Then, the singular values energy corresponding to the feature matrix is employed as a criterion to identify various faults. Compared with the conventional EMD method by simulation experiments, ALIF method has obvious superiority in solving modal aliasing, which is more conducive to the advanced analysis. In this paper, the proposed method is employed to extract the fault information of rolling bearing fault signals from Case Western Reserve University Bearing Data Center. To further verify the effectiveness of the method, the case study is conducted at Drivetrain Diagnostics Simulator. To further illustrate the effectiveness of the method, the results obtained by this method are compared with EMD and EEMD. The results indicated the proposed method performs better in the classification of different mechanical faulty modes.

Transform-domain sparse representation based classification for machinery vibration signals

The working state of machinery can be reflected by vibration signals. Accurate classification of these vibration signals is helpful for the machinery fault diagnosis. A novel classification method for vibration signals, named Transform Domain Sparse Representation-based Classification (TDSRC), is proposed. The method achieves high classification accuracy by three steps. Firstly, time-domain vibration signals, including training samples and test samples, are transformed to another domain, e.g. frequency-domain, wavelet-domain etc. Then, the transform coefficients of the training samples are combined as a dictionary and the transform coefficients of the test samples are sparsely coded on the dictionary. Finally, the class label of the test samples is identified by their minimal reconstruction errors. Although the proposed method is very similar to the Sparse Representation-based Classification (SRC), experimental results illustrates its performance is far superior to SRC in the classification of vibration signals. These experiments include: frequency-domain classification of bearing vibration data from the Case Western Reserve University (CWRU) Bearing Data Center and wavelet-domain classification of six fault-types gearbox vibration data from our rotating machinery experimental platform.

Rolling‐Element Bearing Fault Data Automatic Clustering Based on Wavelet and Deep Neural Network

A method based on wavelet and deep neural network for rolling‐element bearing fault data automatic clustering is proposed. The method can achieve intelligent signal classification without human knowledge. The time‐domain vibration signals are decomposed by wavelet packet transform (WPT) to obtain eigenvectors that characterize fault types. By using the eigenvectors, a dataset in which samples are labeled randomly is configured. The dataset is roughly classified by the distance‐based clustering method. A fine classification process based on deep neural network is followed to achieve accurate classification. The entire process is automatically completed, which can effectively overcome the shortcomings such as low work efficiency, high implementation cost, and large classification error caused by individual participation. The proposed method is tested with the bearing data provided by the Case Western Reserve University (CWRU) Bearing Data Center. The testing results show that the proposed method has good performance in automatic clustering of rolling‐element bearings fault data.

Bearing Fault Automatic Classification Based on Deep Learning

An automatic classification method based on deep learning for bearing fault diagnosis is proposed. The method is designed with the ability of faulty signal automatic clustering without human knowledge. A dataset in which each sample is given a random label is configured after extracting the features of vibration signals from the frequency domain. The dataset is used to train a deep neural network (DNN) to obtain the initial classification. The classification results are assessed by testing the subsignals extracted from the raw data, and the sample labels are modified according to the evaluation result. The modified dataset is used to train the DNN a second time. Samples with characteristic faults are clustered in various classes after iterating the DNN training and testing. The proposed method is tested with the bearing data provided by the Case Western Reserve University (CWRU) Bearing Data Center, which is a standard reference to test fault detection algorithms. The 12k drive end, 48k drive end, and 12k fan end CWRU bearing data are classified into 7, 6, and 4 groups, respectively. The testing results show that the proposed method can achieve automatic clustering for vibration signals with a variety of faults.

An improved wrapper-based feature selection method for machinery fault diagnosis.

A major issue of machinery fault diagnosis using vibration signals is that it is over-reliant on personnel knowledge and experience in interpreting the signal. Thus, machine learning has been adapted for machinery fault diagnosis. The quantity and quality of the input features, however, influence the fault classification performance. Feature selection plays a vital role in selecting the most representative feature subset for the machine learning algorithm. In contrast, the trade-off relationship between capability when selecting the best feature subset and computational effort is inevitable in the wrapper-based feature selection (WFS) method. This paper proposes an improved WFS technique before integration with a support vector machine (SVM) model classifier as a complete fault diagnosis system for a rolling element bearing case study. The bearing vibration dataset made available by the Case Western Reserve University Bearing Data Centre was executed using the proposed WFS and its performance has been analysed and discussed. The results reveal that the proposed WFS secures the best feature subset with a lower computational effort by eliminating the redundancy of re-evaluation. The proposed WFS has therefore been found to be capable and efficient to carry out feature selection tasks.