Search

Bearing fault is a process of gradual development and deepening. In the early stage of the fault, if it can be found out in time and taken reasonable prevention and elimination measures, we can avoid serious losses and safety accidents. Therefore, the feature extraction and analysis of early weak fault has important practical significance. In this paper, an improved multiscale permutation entropy (IMPE) method was proposed to overcome the shortcomings in the coarse-grained process. In order to solve the problem that only considering a single coarse-grained sequence under a certain scale may lead to the loss of feature information, this paper proposed to calculate the time series with equal overlapping segments, that was to consider all coarse-grained sequences under the same scale to reflect the feature information of fault signals more comprehensively. In order to solve the problem that feature extraction is not refined enough when using the first-order moment mean value calculation in traditional MPE calculation, a calculation method based on the skewness of the third-order moment was proposed. The calculation method is more sensitive to the complexity and fluctuation of signals and can better describe the feature details and extract the fault features effectively. IMPE was applied to feature extraction of early weak fault of rolling bearing and input into Support Vector Machines (SVMs) for faults classification. Aiming at SVM parameter optimization problem, an improved chaos firefly optimization algorithm was proposed. Experimental results show that the new method of early weak fault identification based on IMPE-SVM was effective in detecting rolling bearing faults with different severity. [ABSTRACT FROM AUTHOR]

This paper introduces an innovative approach, Deep Multiscale Soft-Threshold Support Vector Data Description (DMS-SVDD), designed for the detection of anomalies and prediction of faults in heavy-duty gas turbine generator sets (GENSETs). The model combines a support vector data description (SVDD) with a deep autoencoder backbone network framework, integrating a multiscale convolutional neural network (M) and soft-threshold activation network (S) into the Deep-SVDD framework. In comparison with conventional methods, such as One-Class Support Vector Machine (OCSVM) and autoencoder (AE), DMS-SVDD demonstrates improvements in accuracy (by 22.94%), recall (by 32%), F1 score (by 12.02%), and smoothness (by 39.15%). The model excels particularly in feature extraction, denoising, and early fault detection, offering a proactive strategy for maintenance. Furthermore, the DMS-SVDD demonstrated enhanced training efficiency with a reduction in the convergence rounds by 66% and overall training times by 34.13%. The study concludes that DMS-SVDD presents a robust and efficient solution for gas turbine anomaly detection, with practical advantages for decision support in turbine maintenance. Future research could explore additional refinements and applications of the DMS-SVDD model across diverse industrial contexts. [ABSTRACT FROM AUTHOR]

The stable operation of the process industrial system, which is integrated with various complex equipment, is the premise of production, which requires the condition monitoring and diagnosis of the system. Recently, the continuous development of deep learning (DL) has promoted the research of intelligent diagnosis in process industry systems, and the sensor system layout has provided sufficient data foundation for this task. However, these DL-driven approaches have had some shortcomings: (1) the output signals of heterogeneous sensing systems existing in process industry systems are often high-dimensional coupled and (2) the fault diagnosis model built from pure data lacks systematic process knowledge, resulting in inaccurate fitting. To solve these problems, a graph feature fusion-driven fault diagnosis of complex process industry systems is proposed in this paper. First, according to the system's prior knowledge and data characteristics, the original multisource heterogeneous data are divided into two categories. On this basis, the two kinds of data are converted to physical space graphs (PSG) and process knowledge graphs (PKG), respectively, according to the physical space layout and reaction mechanism of the system. Second, the node features and system spatial features of the subgraphs are extracted by the graph convolutional neural network at the same time, and the fault representation information of the subgraph is mined. Finally, the attention mechanism is used to fuse the learned subgraph features getting the global-graph representation for fault diagnosis. Two publicly available process chemistry datasets validate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Deep learning has recently received extensive attention in the field of rolling-bearing fault diagnosis owing to its powerful feature expression capability. With the help of deep learning, we can fully extract the deep features hidden in the data, significantly improving the accuracy and efficiency of fault diagnosis. Despite this progress, deep learning still faces two outstanding problems. (1) Each layer uses the same convolution kernel to extract features, making it difficult to adaptively select convolution kernels based on the features of the input image, which limits the network's adaptability to different input features and leads to weak feature extraction. (2) Large number of parameters and long training time. To solve the above problems, this paper proposes an integrated deep neural network that combines an improved selective kernel network (SKNet) with an enhanced Inception-ResNet-v2, named SIR-CNN. First, based on the SKNet, a new three-branch SKNet was designed. Second, the new SKNet is embedded into a depthwise separable convolution network such that the model can adaptively select convolution kernels of different sizes during training. Furthermore, the convolution structure in the Inception-ResNet-v2 network was replaced by the improved depthwise separable convolution network to achieve effective feature extraction. Finally, the time-frequency maps of the raw vibration signals are obtained through short-time Fourier transform (STFT) and then sent to the proposed SIR-CNN network for experiments. The experimental results show that the proposed SIR-CNN achieves superior performance compared to other methods. [ABSTRACT FROM AUTHOR]

Railway axle box bearing fault signal contains high Q-factor resonance and low Q-factor transient shock components with periodic transient shock features that can characterize bearing faults. However, extracting fault features is usually difficult due to noise, transmission paths, and high-amplitude accidental shocks. Therefore, to address the abovementioned problems, the multilevel feature extraction method for adaptive fault diagnosis of railway axle box bearing was proposed. This paper used the maximum second-order cyclostationary blind convolution (CYCBD) to weaken the influence of disturbances, enhancing weak transient shock components. Resonant sparse decomposition(RSSD) is used for the selection of quality factor Q due to its excellent fault feature separation property. Considering the strict fault frequency component periodicity in the envelope spectrum, the envelope spectrum multipoint kurtosis (ESMK) is proposed as a metric. The Q factor is optimized using the gray wolf optimization algorithm (GWO) to obtain an adaptive sparse decomposition method (GWO-RSSD) for extracting bearing transient fault shocks to eliminate the signal effects of high-amplitude disturbance shocks and background noise. The simulation results and measured signal analysis of railway axle box bearing show that ESMK can effectively measure periodic transient shocks under strong shock interference compared with the MED-RSSD method. Thus, GWO-RSSD can adaptively separate the optimal low-Q resonance components, verifying the methods' effectiveness and superiority. [ABSTRACT FROM AUTHOR]

Transformer model is being gradually studied and applied in bearing fault diagnosis tasks, which can overcome the feature extraction defects caused by long-term dependencies in convolution neural network (CNN) and recurrent neural network (RNN). To optimize the structure of existing transformer-like methods and improve the diagnostic accuracy, we proposed a novel method based on the multiscale time-frequency sparse transformer (MTFST) in this paper. First, a novel tokenizer based on shot-time Fourier transform (STFT) is designed, which processes the 1D format raw signals into 2D format discrete time-frequency sequences in the embedding space. Second, a sparse self-attention mechanism is designed to eliminate the feature mapping defect in naive self-attention mechanism. Then, the novel encoder-decoder structure is presented, the multiple encoders are employed to extract the hidden feature of different time-frequency sequences obtained by STFT with different window widths, and the decoder is used to remap the deep information and connect to the classifier for discriminating fault types. The proposed method is tested in the XJTU-SY bearing dataset and self-made experiment rig dataset, and the following work is conducted. The influences of hyperparameters on diagnosis accuracy and number of parameters are analysed in detail. The weights of the attention mechanism (AM) are visualized and analysed to study the interpretability, which explains the partly working pattern of the network. In the comparison test with other existing CNN, RNN, and transformer models, the diagnosis accuracy of different methods is statistically analysed, feature vectors are presented via the t-distributed stochastic neighbor embedding (t-SNE) method, and the proposed MTFST obtains the best accuracy and feature distribution form. The results demonstrate the effectiveness and superiority of the proposed method in bearing fault diagnosis. [ABSTRACT FROM AUTHOR]

The transient impact component of early bearing faults is not obvious, and the traditional basis function expansion method is poor in feature extraction under strong noise conditions. In this paper, a transient feature extraction technique is proposed based on Laplace wavelet and orthogonal matching pursuit algorithm and combined with sparse representation theory. First, the overcomplete and redundant Laplace wavelet dictionary is adopted to represent vibration signals in a sparse way. Then the Hilbert resonance demodulation method is employed to obtain the envelope spectrum of sparse representation signal. Finally, the coefficient calculation problem of sparse representation is solved by orthogonal matching pursuit (OMP) algorithm. Simulation examples and experimental example are used to examine the performance of the proposed method. The results show that the weak-bearing faults feature extraction can be effectively realized since the transient shock component can be identified. Furthermore, the effectiveness of the proposed method is verified by the cyclic multishock simulated signals as well as the practical rolling-bearing vibration signals. Moreover, the comparison studies are also carried out to show that the proposed method outperforms the traditional basis function expansion methods in weak fault feature extraction of bearing. [ABSTRACT FROM AUTHOR]

The feature extraction of high-precision microseismic signals is an important prerequisite for multicategory recognition of microseismic signals, and it is also an important basis for intelligent sensing modules in smart mines. Aiming at the problem of unobvious feature extraction of multiclass mine microseismic signals, this paper is based on the unsupervised learning method in the deep learning method, combined with wavelet packet energy ratio and empirical modulus singular value decomposition, and proposes a method based on wavelet packet energy and empirical modulus singular value decomposition and proposes a method (M-W&E) based on wavelet packet energy and empirical modulus singular value decomposition. This method firstly performs empirical modulus singular value decomposition and wavelet packet energy ratio on the microseismic signal to construct the basic feature vector and then uses the unsupervised learning algorithm to perform the unsupervised learning method feature fusion of the basic feature vector to construct the fused feature vector. After visualization by t-SNE, various distinctions in the fusion feature vector are more obvious. After testing the fusion feature classification using SVM, it is found that the recognition rate of the new feature after feature fusion is better than that of a single wavelet packet empirical energy component and singular value of empirical modulus, which basically meets the engineering needs and is a mine microseism. The signal extraction and feature enhancement fusion of multiclass samples provide a new idea. [ABSTRACT FROM AUTHOR]

This paper aims at the shortcomings of the current conventional processing methods of bearing fault vibration signals in improving signal-to-noise ratio, fine feature extraction, and recognition. A feature extraction and recognition method of abnormal vibration signals based on Ensemble Empirical Mode Decomposition (EEMD) superresolution sparse decomposition is designed. First of all, the superresolution sparse decomposition method is used to refine the set of IMF components of vibration signals after EEMD decomposition. Secondly, the features of the set are extracted and their corresponding energy entropy is calculated. Thirdly, the classification and recognition are carried out. Finally, the effectiveness and feasibility of the method are verified by experiments. It has been proved that this method can better realize the denoising and fine processing aimed at abnormal vibration signals. It has certain theoretical significance and applied value. [ABSTRACT FROM AUTHOR]

Acoustic fault diagnosis technology equipment is non-contact, and the acoustic signal is easy to access. However, it is difficult to extract the feature information of the acoustic signal with low signal-to-noise ratio (SNR). In this paper, a fault diagnosis model (FDM) of axle box bearing based on Chirplet transform (CT) and support vector machine (SVM) is established to diagnose bearing fault based on acoustic signal. The availability of the model is verified by comparing with the vibration acceleration signal bearing fault diagnosis results, and the correctness of the model is verified by utilizing the open database of Western Reserve University. The acoustic-vibration comprehensive bearing fault diagnosis experiment platform (AVEP) is established to investigate the acoustic signal and acceleration signal diagnosis accuracy. The results suggest that, based on the FDM, the diagnosis accuracy and stability of acoustic signal are not as good as acceleration signal when the number of samples is small under the single condition; the diagnosis accuracy of acoustic signal is similar to that of acceleration signal when the number of samples is enough under the multiple condition, which provides a reference for the application of acoustic fault diagnosis technology in engineering in the future. [ABSTRACT FROM AUTHOR]

We aim to address the issues of difficult acquisition of bearing fault data, few feature data sets, and low efficiency of intelligent diagnosis. In this paper, an orthogonal wavelet transform K-nearest neighbor (OWTKNN) diagnosis method has been proposed. The (OWT) method extracts the peaks of each detail signal as training samples and uses the K-Nearest Neighbor (KNN) method for fault classification. The classification results of the multiple fault test data obtained through rolling bearing tests show that the method can reach a fault recognition rate of 100%, and compared with KNN without extracted eigenvalues, it significantly improves the classification effects from various unknown fault data of the bearing inner ring and ball, shortens classification time, and improves the intelligent diagnosis efficiency. In addition, it achieves an overall recognition rate exceeding 95%, Comparing OWT, EMD, and VMD feature extraction methods, both the OWTKNN and k-center point clustering algorithm do not exceed 80% (KCA), also bearing testimony of the effectiveness of this method. [ABSTRACT FROM AUTHOR]

To extract the time-series characteristics of the original bearing signals and predict the remaining useful life (RUL) more effectively, a parallel multichannel recurrent convolutional neural network (PMCRCNN) is proposed for the prediction of RUL. Firstly, the time domain, frequency domain, and time-frequency domain features are extracted from the original signal. Then, the PMCRCNN model is constructed. The front of the model is the parallel multichannel convolution unit to learn and integrate the global and local features from the time-series data. The back of the model is the recurrent convolution layer to model the temporal dependence relationship under different degradation features. Normalized life values are used as labels to train the prediction model. Finally, the RUL was predicted by the trained neural network. The proposed method is verified by full life tests of bearing. The comparison with the existing prognostics approaches of convolutional neural network (CNN) and the recurrent convolutional neural network (RCNN) models proves that the proposed method (PMCRCNN) is effective and superior in improving the accuracy of RUL prediction. [ABSTRACT FROM AUTHOR]

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. [ABSTRACT FROM AUTHOR]

Condition monitoring and fault diagnosis of the bearing are essential for the smooth operation of rotating machinery. In this paper, an end-to-end intelligent fault diagnosis method for bearing combining one-dimensional convolutional neural network with long short-term memory network (1DCNN-LSTM) is proposed for the deficiencies of existing fault diagnosis methods. First, the proposed method takes one-dimensional fault data directly as input. Second, one-dimensional convolutional neural network (1DCNN) is used for self-adaptively extracting robust features from the original bearing signal, and more features are extracted while ensuring the validity and saliency of the extracted features by combining maximum pooling and average pooling layers to downsample features. Then, long short-term memory network (LSTM) is used to learn the temporal dependencies among features. At last, fault identification is achieved. 1DCNN-LSTM does not require any manual feature extraction, and the errors caused by reliance on expert experience and incomplete information in traditional feature extraction methods are avoided. The results show that the proposed classifier with good generalization performance not only diagnoses the category of fault quickly and accurately under different load conditions but also achieves an average fault identification accuracy of 99.95%. For its powerful learning abilities, this method can also be applied to the bearing fault diagnosis of rotating machinery in many fields. [ABSTRACT FROM AUTHOR]

To address the degradation of diagnostic performance due to data distribution differences and the scarcity of labeled fault data, this study has focused on transfer learning-based cross-domain fault diagnosis, which attracts considerable attention. However, deep transfer learning-based methods often present a challenge due to their time-consuming and costly nature, particularly in tuning hyperparameters. For this issue, on the basis of classical features-based transfer learning method, this study introduces a new framework for bearing fault diagnosis based on supervised joint distribution adaptation and feature refinement. It first utilizes ensemble empirical mode decomposition to process raw signals, and statistical features extraction is implemented. Then, a new feature refinement module is designed to refine domain adaptation features from high-dimensional feature set by evaluating the fault distinguishability and working-condition invariance of feature data. Next, it proposes a supervised joint distribution adaptation method to conduct improved joint distribution alignment that preserves neighborhood relationships within a manifold subspace. Finally, an adaptive classifier is trained to predict fault labels of feature data across varying working conditions. To prove the cross-domain fault diagnosis performance and superiority of the proposed methods, two bearing datasets are applied for experiments, and the experimental results verify that the model built by the proposed framework can achieve desirable diagnosis performance under different working conditions and that it apparently outperforms comparative models. [ABSTRACT FROM AUTHOR]

The vibration signal of rolling bearing fault is nonlinear and nonstationary under the interference of background noise, and it is difficult to extract fault features from it. When feature mode decomposition is used to analyze signals, prior parameter settings can easily affect the decomposition results. Therefore, a fault feature extraction method based on improved whale optimization algorithm is proposed to optimize feature modal decomposition parameters. The improved WOA integrates Lévy flight and adaptive weight, and envelope entropy is used as fitness function to optimize feature modal decomposition parameters. The feature mode decomposition of the original signal is performed using the optimal combination of parameters to obtain multiple IMF components. The optimal IMF component envelope demodulation analysis is selected according to the kurtosis value, and the fault feature is extracted through the envelope spectrum. Comparing the LMWOA method with PSO and WOA methods by simulated and experimental signals, the results show that the optimization speed of LMWOA is faster than that of other methods. Compared with CEEMD, VMD, and FMD methods, the improved WOA-FMD method has higher fault feature ratio and can accurately extract fault features under noise interference. This method can effectively solve the parameter adaptive ability and improve the accuracy of fault diagnosis, which has practical significance. [ABSTRACT FROM AUTHOR]

Traditional fault diagnosis methods require complex signal processing and expert experience, and the accuracy of fault identification is low. To solve these problems, a fault diagnosis method based on an improved convolutional neural network (CNN) is proposed. Based on the traditional CNN model, a variety of convolution stride modes were added to extract features of different scales of signals and expand the feature dimension. Firstly, the vibration signals were collected and grouped. Then, the data were divided into a training set and a test set and input into improved CNN for feature extraction and model training to realize fault identification. The proposed model achieved a classification accuracy of 99.3% when testing the vibration data of the armored vehicle. Finally, the proposed model was used to classify different fault types of planetary gearboxes. The gradient-weighted class activation mapping (Grad-CAM) method was used to visualize the classification weight of samples. The results showed that the classification accuracy reaches 98% under various working conditions of the planetary gearbox. [ABSTRACT FROM AUTHOR]

An automatic recognition algorithm based on the feature extraction of working parameters to recognize each state in the loading cycle process of an electric mining shovel was proposed. The working parameters were collected using the electric shovel's online monitoring system. The swing angle of the shovel boom and motor operating signals were used as key recognition objects; the waveform features of each stage were extracted as recognition marks in the loading cycle of an electric shovel using the time domain characteristic analysis method. The algorithm was developed to recognize the loading cycles in real time. Moreover, the dynamic time warping (DTW) algorithm was used to detect and classify the preliminary recognition results by optimizing its distance threshold parameters, reducing the error rate of the model. The method was validated by comparing synchronous video-recordings with the results of the algorithm. Results showed that the proposed recognition method of the shovel loading cycle process exhibited real-time performance and high accuracy in understanding the different work tasks, providing effective data support for mining and the analysis of shovel working parameters, helping to improve the energy efficiency of electric mining shovel. [ABSTRACT FROM AUTHOR]

As the key bearing part of the crane, the low-speed hub bearing of the crane exists in special working conditions of low-speed and alternating heavy load. It is difficult to extract its fault characteristics accurately by existing analysis methods. The main idea of the broadband mode decomposition (BMD) method previously proposed is to search in the association dictionary library containing broadband and narrowband signals. However, when it is applied to the broadband signals interfered by strong noise, the decomposition is easy to produce modal confusion, so the modulated broadband mode decomposition (MBMD) method is proposed. The fault signal just can be analyzed by MBMD, so it is applied to the fault diagnosis of low-speed hub bearing of the crane. To realize the fault identification of low-speed hub bearing of the crane, firstly, the original signal is decomposed by MBMD. Secondly, the eigenvalues of the first three-component signals are calculated, the eigenvalue matrix is constructed, and the marked features are selected by the distance evaluation technique (DET). Finally, the marked features are input into BP neural network for training and testing to identify the types of bearing fault. Compared with EEMD, VMD, and BMD, the MBMD method combined with BP neural network has good performance in feature extraction and fault identification. [ABSTRACT FROM AUTHOR]

As an entropy representing the complexity of sequence, slope entropy (SloE) is applied to feature extraction of bearing signal for the first time. With the advantage of slope entropy in feature extraction, the effectiveness of bearing fault signal diagnosis can be verified. Five different kinds of entropy are selected to be comparative methods for experiments, and they are permutation entropy (PE), dispersion entropy (DE), a version of entropy adapted by PE, which is weighted permutation entropy (WPE), and two versions of entropy adapted by DE, which are fluctuating dispersion entropy (FDE) and reverse dispersion entropy (RDE). A method of extracting a single feature of bearing fault signals based on SloE is carried out. Firstly, the features of the bearing signals are extracted by the six kinds of entropy. Then, some relevant data are computed, and the identification ratios are calculated by the K-nearest neighbor (KNN) algorithm. The experimental result indicated that the identification ratio of SloE is the highest at 97.71% by comparing with the identification ratios of the other five kinds of entropy, which is higher by at least 13.54% than the others and 27.5% higher than the lowest one. [ABSTRACT FROM AUTHOR]

The classification and recognition of ship-radiated noise (SRN) is of great significance to the processing of underwater acoustic signals. In order to improve the stability of recognition and more accurately identify SRN, single feature extraction and dual feature extraction based on hierarchical dispersion entropy (HDE) are proposed. For single feature extraction, HDE of the best node among the eight nodes of the third layer decomposition is extracted. For dual feature extraction, HDE of the best two nodes among the 14 nodes of the first-, second-, and third-layer decompositions are required. The results show that the recognition rate of single and dual feature extraction originated from the method based on HDE reaches 85% and 100%, respectively, better than the method of hierarchical reverse dispersion entropy (HRDE) and hierarchical permutation entropy (HPE). [ABSTRACT FROM AUTHOR]

Early diagnosis of failures can prevent financial losses and industry downtime. In this article, the author proposes an early fault diagnosis technique for rotor-bearing faults. The proposed technique is based on the recognition of sound signals. The author measured and analyzed the three states of the rotor-bearing system: the rotor-bearing system under normal operating conditions, the rotor-bearing system with faulty bearings, and the rotor-bearing system with rotor friction. In this article, an original feature extraction method is described, namely, the 1/3 doubling method (a method of selecting the amplitude of the frequency ratio that is a multiple of 30% of the maximum amplitude). This method is used to form feature vectors. A classification of the obtained vectors was performed by the KNN (K-nearest neighbor classifier), the SVM (support vector machine), and the decision tree. The method is also compared with the Fourier synchrosqueezed transform. The experimental results show that the method can diagnose early faults of rotor-bearing systems simply and quickly and can be used to protect the safe operation of mechanical equipment. [ABSTRACT FROM AUTHOR]

Mechanical vibration constitutes a valuable cue for performing fault diagnosis as it is directly related to the transient regime of rolling machinery. This study establishes a multidomain feature fusion network (MFFN) to extract and fuse multidomain features through a novel multistream architecture. Three primary features are simultaneously extracted from the time, frequency, and time-frequency domains. Then, highly representative features are extracted via three convolutional branches in one- or two-dimensional spaces. A novel squeeze-connection-excitation (SCE) module is proposed to adaptively fuse features in the three domains. The advantage offered by the proposed method is that it can leverage cues from the raw vibration signal, resulting in accurate fault diagnosis. Experimental results comprehensively demonstrate and analyze the high accuracy and generalization achieved by this MFFN-based fault diagnosis method. [ABSTRACT FROM AUTHOR]

An intelligent fault severity detection method based on variational mode decomposition- (VMD-) Wigner-Ville distribution (WVD) and sparrow search algorithm- (SSA-) optimized deep belief network (DBN) is suggested to address the problem that typical fault diagnostic algorithms are inappropriate for extremely comparable vibration signals when the samples are insufficient. VMD is used to process the original vibration signal to obtain the band intrinsic mode functions (BIMFs) with different frequencies. WVD produces the two-dimensional spectrum of the key BIMF with the highest variance contribution rate. The input sample of DBN is composed of a characteristic matrix formed by the two-dimensional spectrum of multiple fault signals. DBN's learning rate and batch size are both tuned globally by SSA, which has a significant influence on network error. The fitness function in the parameter optimization process is the network's root mean square error (RMSE). Finally, the input samples are loaded into a DBN that has the best structure for detecting severity. Experiments show that, based on VMD-WVD and SSA-DBN, accuracy of the fault severity detection model for rotating machines, which has good generalization ability and robustness, can reach 98%. Compared with BPNN, the traditional DBN, VMD-DBN, VMD-PSO-DBN, and other methods, the proposed algorithm has strong adaptive feature extraction ability and generalization of application. [ABSTRACT FROM AUTHOR]

In the defect diagnosis of the gear-shaft-bearing system with compound defects, the generated vibration signals are complicated. In addition, the information acquired by a single sensor is easily affected by uncertain factors, and low diagnostic accuracy is caused when traditional defect diagnosis methods are used, which cannot meet the high-precision diagnosis requirements. Therefore, a method is developed to identify the defect types and defect degrees of the gear-shaft-bearing system efficiently. In this method, the vibration signals are collected using multiple sensors, the dual-tree complex wavelet and the optimal weighting factor (OWF) methods are used for the data layer fusion, and the preprocessing is realized through wavelet transform and FFT. A learning model based on two-stream CNN composed of 1D-CNN and 2D-CNN is established, and the obtained wavelet time-frequency map and FFT spectrum are used as the input. Then, the trained features from the output of the connected layer are classified by the SVM. Compared with the OWF-1DCNN and OWF-2DCNN models, the time consumption of the OWF-TSCNN model is increased by 14.5%–26.6%, and the convergence speed of the network is decreased. However, its accuracy reaches 100% and 99.83% in the training set and test set, and the loss entropy and over-fitting rate are also greatly reduced. The feature extraction ability and generalization ability of the OWF-TSCNN model are increased, reaching 100% diagnosis accuracy on different defect types and defect degrees, which is more suitable for defect diagnosis of the gear-shaft-bearing system. [ABSTRACT FROM AUTHOR]

The fault vibration signal of a bearing has nonstationary and nonlinear characteristics and can be regarded as the combination of multiple amplitude- and frequency-modulation components. The envelope of a single component contains the fault characteristics of a bearing. Local characteristic-scale decomposition (LCD) can decompose the vibration signal into a series of multiple intrinsic scale components. Some components can clearly reflect the running state of a bearing, and fault diagnosis is conducted according to the envelope spectrum. However, the conventional LCD takes a single-channel signal as the research object, which cannot fully reflect the characteristic information of the rotor, and the analysis results based on different channel signals of the same section will be inconsistent. To solve this problem, based on full vector spectrum technology, the homologous dual-channel information is fused. A vector LCD method based on cross-correlation coefficient component selection is given, and a simulation analysis is completed. The effectiveness of the proposed method is verified by simulated signals and experimental signals of a bearing, which provides a method for bearing feature extraction and fault diagnosis. [ABSTRACT FROM AUTHOR]

Detection of cavitation in centrifugal pumps is critical in their condition monitoring. In order to detect cavitation more accurately and confidently, more advanced signal processing techniques are needed. For the classification of a pump conditions based on the outputs of these techniques, advanced machine learning techniques are needed. In this research, an automatic system for cavitation detection is proposed based on machine learning. Bispectral analysis is used for analyzing the vibration signals. The resulting bispectrum images are given to convolutional neural networks (CNNs) as inputs. The CNNs are a pretrained AlexNet and a pretrained GoogleNet, which are used in this application through transfer learning. On the contrary, a laboratory test setup is used for generating controlled cavitation in a centrifugal pump. The suggested algorithm is implemented on the vibration dataset acquired from the laboratory pump test setup. The results show that the cavitation state of the pump can be detected accurately using this system without any need to image processing or feature extraction. [ABSTRACT FROM AUTHOR]