Bearing Fault Conditions Research Articles

Smart systems for real-time bearing faults diagnosis by using vibro-acoustics sensor fusion with Bayesian optimised 1-D CNNs

ABSTRACT Diagnosis of bearing faults in real-time is challenging when healthy bearing conditions are mixed with faulty ones, affecting the overall system of rotating machinery. Deep Learning provides an effective approach for condition-based maintenance of bearing faults, bypassing traditional signal processing complexity. Convolutional Neural Networks (CNNs) excel in real-time fault detection in bearings, leveraging their feature extraction capabilities from heterogeneous sensors. Usually, selecting optimal hyperparameters for CNNs is time-consuming and impacts model performance. Recent literature demonstrates that CNNs-based models for detecting bearing faults typically undergo trial searches to select optimal hyperparameters, leading to time-intensive procedures. To fill this research gap, our study proposes a Bayesian optimised 1-D CNNs method to address hyperparameter tuning challenges. Using Machinery Fault Simulator®, we demonstrate the effectiveness of our approach in identifying various bearing fault conditions through vibro-acoustics sensors. Bayesian optimisation efficiently partitions datasets for parallel computation, optimises hyperparameters, and minimises loss functions to enhance validation accuracy. The proposed method achieved a test accuracy of 99.62%, surpassing the benchmark’s 99.27%. Its effectiveness for bearing fault diagnosis is evident, compared to 96.76% without optimisation. Therefore, this study presents technical innovations, showcasing the diagnosis of diverse bearing faults with limited data through the integration of vibro-acoustics sensors.

Bearing Fault Diagnosis under Transient Conditions: Using Variational Mode Decomposition and the Symmetrized Dot Pattern‐Based Convolutional Neural Network Model

An effective bearing fault diagnosis method for gearbox applications under variable operating conditions is proposed, utilizing variational mode decomposition (VMD) for feature extraction, symmetrized dot pattern (SDP) for visual representation, and convolutional neural network (CNN) for deep feature extraction and classification. First, three operating conditions of normal bearing, bearing outer ring fault, and bearing inner ring fault were simulated. The vibration and speed pulse signals under variable conditions were collected synchronously, and the nonstationary time‐domain vibration signals were converted into stationary angular‐domain vibration signals by the full‐period angular‐domain resampling technique. To eliminate the noise components, VMD was used to decompose the vibration signal into different frequency bands to obtain the intrinsic mode functions (IMFs), which were further represented by SDP images, and feature images were extracted according to the noncorrelation coefficient. Finally, based on the designed CNN model, the depth features of the SDP image were automatically extracted to form the feature vector, and the formed feature vector was used as the input to the softmax classifier to identify the bearing fault condition. Experimental results showed that the proposed method achieved the diagnostic accuracy of 100%. Compared with other SDP‐based methods, this method has good feature extraction and high fault diagnosis accuracy.

A Time–Frequency Residual Convolution Neural Network for the Fault Diagnosis of Rolling Bearings

A time–frequency residual convolution neural network (TFRCNN) was proposed to identify various rolling bearing fault types more efficiently. Three novel points about TFRCNN are presented as follows: First, by constructing a double-branch convolution network in the time domain and the frequency domain, the respective features in the time domain and the frequency domain were extracted to ensure the rich and complete feature representation of raw data sources. Second, specific residual structures were designed to prevent learning degradation of the deep network, and global average pooling was adopted to improve the network’s sparsity. Third, TFRCNN was better than the other models in terms of prediction accuracy, robustness, generalization ability, and convergence. The experimental results demonstrate that the prediction accuracy rate of TFRCNN, trained using mixing load data, reached 98.88 to 99.92% after optimizing the initial learning rate and choosing the optimizer and loss function. It was verified that TFRCNN can adaptively learn to extract deep fault features, accurately identify bearing fault conditions, and overcome the limitations of classical shallow feature extraction and classification methods, as well as common convolution neural networks. Hence, this investigation revealed TFRCNN’s potential for bearing fault diagnosis in practical engineering applications.

Vibration Analysis of a Roller Bearing Condition Used in a Tangential Threshing Drum of a Combine Harvester for the Smooth and Continuous Performance of Agricultural Crop Harvesting

Testing the reliability of the threshing unit is difficult and thus often neglected before the harvesting season, which can result in breakdown maintenance during peak harvesting time in difficult-to-access areas for sensor mounting. In this paper, the vibration analysis of the threshing condition of the combine harvester was performed by introducing the bracket for inaccessible locations. The Adash A4900 Vibrio M analyzer (Adash spol. s.r.o., Ostrava, Czech Republic) was used for a vibration signal measurement and the DDS Adash software was used for signal processing. The demodulated fast Fourier transform (FFT) root mean square (RMS) (500 Hz–16 kHz) method was used to evaluate the bearing condition and DDS Adash Fault Source Identification Tool (FASIT) technology was used to evaluate other mechanical conditions such as the looseness, misalignment, and unbalance of the threshing unit of the Massey Ferguson series of combine harvesters. Modal and random vibration analyses were performed on the bracket and components and compared to prevent the resonance phenomenon using the Ansys Software (Ansys, Inc., Canonsburg, PA, USA). The constrained modal analysis of the threshing drum was performed to observe the deformation. Decent results were obtained from the method used. The method was again validated by the tangential threshing test bench and successfully determined the bearing fault condition. The method used is an uncomplicated and effective way of performing the bearing analysis of the tangential unit of the combine harvester.

Application of Improved Robust Local Mean Decomposition and Multiple Disturbance Multi-Verse Optimizer-Based MCKD in the Diagnosis of Multiple Rolling Element Bearing Faults

Rolling element bearings are an important joint in mechanical equipment and have a high engineering application value. To solve the problem of the difficulty in extracting periodic fault pulses due to complex noise interference and the interference of transmission paths in rolling element bearing fault characteristic signals, a novel hybrid fault diagnosis method based on complementary complete ensemble robust local mean decomposition with adaptive noise (CCERLMDAN) combined with multiple disturbance multi-verse optimizer (MDMVO)-based Maximum correlated Kurtosis deconvolution (MCKD) is proposed in this paper, and applied in different rolling element bearing fault conditions. Firstly, the CCERLMDAN method adaptively decomposes the fault vibration signal into multiple product functions (PF), and then selects the PF with the most fault information through the sensitive index (SI). Finally, the MDMVO method adaptively selects the best parameter combination of the MCKD method and then uses MCKD to perform a deconvolution operation on the selected PF, highlighting the periodic fault pulse excited by the bearing fault. The field-measured vibration signals of rolling element bearing faults are applied to verify the proposed method. The final results show that the method effectively improves the fault diagnosis accuracy of rolling element bearings, and both CCERLMDAN and MDMVO methods achieve a better performance than the original method.

A self-adaptive multiple-fault diagnosis system for rolling element bearings

The inevitable simultaneous formation of multiple-faults in bearings generates severe vibrations, causing premature component failure and unnecessary downtime. For accurate diagnosis of multiple-faults, machine learning (ML) models need to be trained with the signature of different multiple-faults, which increases the data acquisition time and expense. This paper proposes a self-adaptive vibration signature-based fault diagnostic method for detecting multiple bearing faults using various single-fault vibration signatures. A time-frequency-based hybrid signal processing technique, which involves discrete wavelet transform and Hilbert transform, was adopted for signal decomposition, followed by the implementation of a sliding window-based feature extraction process. Seven optimized metaheuristic algorithms were used to find the best feature sets, which were further used for the training of three ML models. The results show that the proposed methodology has tremendous potential to detect multiple bearing fault conditions in any possible combination using single-fault data. This will be helpful where accessibility to large amounts of data is limited for multiple-fault diagnosis.

An efficient diagnosis approach for bearing faults using sound quality metrics

The perception of sound quality is an important source of information and can be used as a promising indicator to diagnose various faults in rotating machines. This work presented a methodology involving the detection of bearing faults using sound quality metrics. Head and Torso Simulator (HATS) was used to acquire the sound signals of different bearing fault conditions with a minimal background noise using a semi-anechoic chamber. After that, various sound quality features were extracted from the acquired signals, and descriptive analysis for five bearing conditions was presented based on these features. Moreover, these features were used to develop a self-adaptive fault diagnosis system using a Support Vector Machine (SVM). Experimental validation of the proposed method was also done with different background conditions and with different microphones. The results showed that the proposed methodology is reliable and effective for accurate fault diagnosis of bearings in a non-invasive manner.

Fault severity classification of ball bearing using SinGAN and deep convolutional neural network

Condition monitoring and diagnosis of a bearing are very important for any rotating machine as it governs the safety while the machine is in operating condition. To construct a feature vector selection of suitable signal processing techniques is a challenge for vibration-based condition monitoring techniques. In the methodology proposed, Short Time Fourier Transform (STFT), Walsh Hadamard Transform (WHT) and Variable Mode Decomposition (VMD) are used to generate 2-D time-frequency spectrograms from the various fault conditions of bearing. When Deep learning techniques apply for fault diagnosis, a large amount of dataset is required for training of machine learning model. To overcome this issue single image Generative Adversarial Network (SinGAN) as a data augmentation technique, utilized for generating additional 2-D time-frequency spectrograms from various fault conditions of ball bearing. To detect fault severity, four deep learning algorithms, ResNet 34, ResNet50, VGG16, and MobileNetV2 are used as a classifier. Experiments are conducted on a rolling bearing dataset provided by the bearing data center of Case Western Reserve University (CWRU) for validating the utility of methodology proposed. Results show that the proposed methodology enables to detect fault severity level with high classification accuracy.

Automatic Feature Extraction and Construction Using Genetic Programming for Rotating Machinery Fault Diagnosis.

Feature extraction is an essential process in the intelligent fault diagnosis of rotating machinery. Although existing feature extraction methods can obtain representative features from the original signal, domain knowledge and expert experience are often required. In this article, a novel diagnosis approach based on evolutionary learning, namely, automatic feature extraction and construction using genetic programming (AFECGP), is proposed to automatically generate informative and discriminative features from original vibration signals for identifying different fault types of rotating machinery. To achieve this, a new program structure, a new function set, and a new terminal set are developed in AFECGP to allow it to detect important subband signals and extract and construct informative features, automatically and simultaneously. More important, AFECGP can produce a flexible number of features for classification. Having the generated features, k -Nearest Neighbors is employed to perform fault diagnosis. The performance of the AFECGP-based fault diagnosis approach is evaluated on four fault diagnosis datasets of varying difficulty and compared with 14 baseline methods. The results show that the proposed approach achieves better fault diagnosis accuracy on all the datasets than the competitive methods and can effectively identify different fault conditions of rolling bearing, gear, and rotor.

Efficient Fault Diagnosis of Rolling Bearings Using Neural Network Architecture Search and Sharing Weights

Bearing is one of the most vital components of industrial machinery. The failure of bearing causes severe problems in the machinery. Therefore, continuous monitoring for the bearings is essential rather than regular manual checking, with the requirement for accuracy of prediction and efficiency. This paper proposes a novel intelligent bearing fault condition monitoring and diagnosis method focusing on computation efficiency, which is an important aspect of a continuous monitoring and embedded-based diagnosis device. In the proposed method, acoustic emission signals containing bearing health information are converted into 2-D spectrograms by Constant Q-Transform (CQT) before using a convolutional neural network to infer the bearing state. To reduce the latency while maintaining high accuracy, we propose an efficient search space for neural network architecture search, i.e., a channel distribution search, that automatically obtain the best performing network. Moreover, we present a separation between two processes of condition monitoring and fault diagnosis to save overall computing resources, with a policy of sharing weights in the training process and sharing features in the testing process. The experimental results show that the proposed method reduces about 50% inference time compared to previous methods while achieving an accuracy of 99.82% for eight types of single and compound fault diagnosis for variable rotational speeds.

Comparison of Vibration, Sound and Motor Current Signature Analysis for Detection of Gear Box Faults

Gear box is used in automobiles and industries for power transmission under different working conditions and applications. Failure in a gear box at unexpected time leads to increase in machine downtime and maintenance cost. In order to overcome these losses, the most effective condition monitoring technique has to be used for early detection of faults. Vibration and sound signal analysis have been used for monitoring the condition of rotating machineries. Motor Current Signature Analysis (MCSA) has rarely been used in gearbox condition monitoring. This work presents a methodology based on vibration, sound and motor current signal analysis for diagnosis of gearbox faults under various simulated gear and bearing fault conditions. Statistical features were extracted from the raw data of these three transducer signals and the best features were selected from the extracted features. Then the selected features were given as an input to Artificial Neural Network (ANN) and Support Vector Machine (SVM) classifiers and their performances were compared. In recent years, Hybrid Electric Vehicles (HEV) are gaining more interest for their advances and this work had a scope in monitoring the power loss in hybrid electric vehicle gearbox using MCSA.

Bearing Fault Classification of Induction Motors Using Discrete Wavelet Transform and Ensemble Machine Learning Algorithms

Bearing fault diagnosis at early stage is very significant to ensure seamless operation of induction motors in industrial environment. The identification and classification of faults helps to undertook maintenance operation in an efficient manner. This paper presents an ensemble machine learning-based fault classification scheme for induction motors (IMs) utilizing the motor current signal that uses the discrete wavelet transform (DWT) for feature extraction. Three wavelets (db4, sym4, and Haar) are used to decompose the current signal, and several features are extracted from the decomposed coefficients. In the pre-processing stage, notch filtering is used to remove the line frequency component to improve classification performance. Finally, the two ensemble machine learning (ML) classifiers random forest (RF) and extreme gradient boosting (XGBoost) are trained and tested using the extracted feature set to classify the bearing fault condition. Both classifier models demonstrate very promising results in terms of accuracy and other accepted performance indicators. Our proposed method achieves an accuracy slightly greater than 99%, which is better than other models examined for the same dataset.

Hybrid Fault Diagnosis of Bearings: Adaptive Fuzzy Orthonormal-ARX Robust Feedback Observer

Rolling-element bearings (REBs) make up a class of non-linear rotating machines that can be applied in several activities. Conceding a bearing has complicated and uncertain behavior that exhibits substantial challenges to fault diagnosis. Thus, the offered anomaly-diagnosis algorithm, based on a fuzzy orthonormal-ARX adaptive fuzzy logic-structure feedback observer, is developed. A fuzzy orthonormal-ARX algorithm is presented for non-stationary signal modeling. Next, a robust, stable, reliable, and accurate observer is developed for signal estimation. Therefore, firstly, a classical feedback observer is implemented. To address the robustness drawback found in feedback observers, a multi-structure technique is developed. Furthermore, to generate signal estimation performance and reliability, the fuzzy logic technique is applied to the structure feedback observer. Also, to improve the stability, reliability, and robustness of the fuzzy orthonormal-ARX fuzzy logic-structure feedback observer, an adaptive algorithm is developed. After generating the residual signals, a support vector machine (SVM) is developed for the detection and classification of the bearing fault conditions. The effectiveness of the proposed procedure is validated using two different datasets for single-type fault diagnosis based on the Case Western Reverse University (CWRU) vibration dataset and multi-type fault diagnosis of bearing using the Smart Health Safety Environment (SHSE) Lab acoustic emission dataset. The proposed algorithm increases the classification accuracy from 86% in the SVM-based fuzzy orthonormal-ARX feedback observer to 97.5% in single-type fault and from 80% to 98.3% in the multi-type faults.

Three-phase induction motor fault detection based on thermal image segmentation

Induction motors are widely used in many industrial applications. Hence, it is very important to monitor and detect any faults during their operation in order to alert the operators so that potential problems could be avoided before they occur. In general, a fault in the induction motor causes it to get hot during its operation. Therefore, in this paper, thermal condition monitoring has been applied for detecting and identifying the faults. The main contribution of this study is to apply new colour model identification namely Hue, Saturation and Value (HSV), rather than using the conventional grayscale model. Using this new model the thermal image was first converted into HSV. Then, five image segmentation methods namely Sobel, Prewitt, Roberts, Canny and Otsu was used for segmenting the Hue region, as it represents the hottest area in the thermal image. Later, different image matrices containing the best fault information extracted from the image were used in order to discriminate between the motor faults. The values which were extracted are Mean, Mean Square Error and Peak Signal to Noise Ratio, Variance, Standard Deviation, Skewness and Kurtosis. All the above features were applied in three different motor bearing fault conditions such as outer race, inner race and ball bearing defects with different load conditions namely No load, 50% load and 100% load. The results showed that the proposed HSV colour model based on image segmentation was able to detect and identify the motor faults correctly. In addition, the method described here could be adapted for further processing of the thermal images.

Intelligent Fault Diagnosis of Synchromesh Gearbox Using Fusion of Vibration and Acoustic Emission Signals for Performance Enhancement

Condition monitoring system monitors the system degradation and it identifies common failure modes. Several sensor signals are available for monitoring the changes in system components. Vibration signal is one of the most extensively used technique for monitoring rotating components as it identifies faults before the system fails. Early fault detection is the significant factor for condition monitoring, where Acoustic Emission ( AE ) sensor signals have been applied for early fault detection due to their high sensitivity and high frequency. In this paper, vibration and acoustic emission signals are acquired under various simulated gear and bearing fault conditions from the synchromesh gearbox. Then the statistical features are extracted from vibration and AE signals and then the prominent features are selected using J48 decision tree algorithm respectively. The best features from the vibration and AE signals are then fused using feature-level fusion strategy and it is classified using Support Vector Machine ( SVM ) and Proximal Support Vector Machine ( PSVM ) classifiers and it is compared with individual signals for fault diagnosis of the synchromesh gearbox. From the experiments, it is observed that the performance of the fault diagnosis system has been improved for the proposed feature level fusion technique compared to the performance of unfused vibration and AE feature sets.

A bearing fault and severity diagnostic technique using adaptive deep belief networks and Dempster–Shafer theory

An artificial intelligent bearing fault and hierarchical severity diagnosis framework is proposed in this study. The framework utilizes a combined deep belief networks (DBNs) and Dempster–Shafer (D-S) theory fault diagnosis scheme and adopts a two-stage approach in classifying (1) bearing fault conditions and (2) fault severities. The combined fault diagnostic scheme first employs two parameter-optimized DBNs to process the horizontal and vertical vibration data acquired from the bearing house of a test rig, where the parameters of the DBNs are optimized using a hybrid genetic algorithm and particle swarm optimization algorithm proposed in this study. The classification results from the two DBNs are fused further using the D-S theory to improve the diagnostic accuracy. The fault diagnosis scheme is used first to classify the bearing fault conditions in Stage 1 from a bulk dataset containing all bearing operation conditions under study. The same diagnosis scheme is applied once more to classify the hierarchical fault severities for each fault condition in Stage 2 using the pre-classified data from Stage 1. The effectiveness of the framework is then evaluated on a set of bearing condition monitoring data. A comparison study between the results obtained using the current method and those from existing published work is also presented in the article. It is shown that the accuracy for bearing fault and severity diagnosis can be substantially improved by using the current framework.

Fault diagnosis of rolling element bearing based on artificial neural network

This paper proposes the expert system for accurate fault detection of bearing. The study is based upon advanced signal processing method as wavelet transform and artificial intelligence technique as artificial neural network (ANN) and K-nearest neighbor (KNN), for fault classification of bearing. An adaptive algorithm based on wavelet transform is used to extract the fault classifying features of the bearing from time domain signal. These features have been used as inputs to proposed ANN models and the same features have also been used for KNN. Dedicated experimental setup was used to perform the test upon the bearing. Single data set for four fault conditions of bearing is collected to train ANN and KNN. The processed and normalized data was trained by using backpropagation multilayer perceptron neural network. The results obtained from ANN are compared with KNN, ANN results proved to be highly effective for classification of multiple faults.

Sound Reconstruction of an Overdetermined Mixture in a Mini Plant: An Experimental Study

In this study, we propose a sound-based signal analysis technique emitted by rotating machines. Three water pump machines are set in a row to reflect a mini plant. We raise an overdetermined case with four sensors that are used to record three machines’ signal sounds with unbalanced, normal and bearing fault conditions. We use non-negative matrix factorization (NMF) approach to reconstruct the signal sounds of each machine which had previously been mixed. Based on the results, the type of engine damages can be determined based on the fundamental frequency of each machine. The first reconstructed signal Ŝ1 is similar to the baseline signal of normal machine with LSD of 1 and MSE of 5.84 × 10−9 The second reconstructed signal Ŝ2 has a similarity to the baseline signal of the machine with bearing fault with LSD of 1 and MSE of 1.10 × 10−7, while the third reconstruction signal Ŝ3 is similar to the machine’s baseline signal with unbalanced condition,that is, LSD of 1.02 and MSE of 1.38 × 10−8.

Fault Diagnosis of Rolling Element Bearing Using Multi-Scale Lempel-Ziv Complexity and Mahalanobis Distance Criterion

A new fault diagnosis technique for rolling element bearing using multi-scale Lempel-Ziv complexity (LZC) and Mahalanobis distance (MD) criterion is proposed in this study. A multi-scale coarse-graining process is used to extract fault features for various bearing fault conditions to overcome the limitation of the single stage coarse-graining process in the LZC algorithm. This is followed by the application of MD criterion to calculate the accuracy rate of LZC at different scales, and the best scale corresponding to the maximum accuracy rate is identified for fault pattern recognition. A comparison analysis with Euclidean distance (ED) criterion is also presented to verify the superiority of the proposed method. The result confirms that the fault diagnosis technique using a multi-scale LZC and MD criterion is more effective in distinguishing various fault conditions of rolling element bearings.

Rolling Bearing Fault Diagnosis Using Modified LFDA and EMD With Sensitive Feature Selection

In order to improve the accuracy of bearings fault diagnosis, one of the most crucial components of rotating machinery, a novel features extraction procedure incorporating an improved features dimensionality reduction method is proposed. In the first step, using the empirical mode decomposition method, the original statistical characteristics were calculated from intrinsic mode functions of the vibration signal. Due to information redundancy of the original statistical characteristics, this paper presents a novel features extraction method that combines K-means method and standard deviation to select the most sensitive characteristics. Furthermore, a modified features dimensionality reduction method is proposed, to realize the low-dimensional representations for high-dimensional feature space. Finally, the performance of the fault diagnosis model is evaluated by vibration signals with 12 bearing fault conditions, which are provided by Bearing Data Center of Case Western Reserve University. Experiment results show that the proposed fault diagnosis model can serve as an effective and adaptive bearing fault diagnosis system.