Gear Fault Diagnosis Research Articles

A hybrid approach for fault diagnosis of spur gears using Hu invariant moments and artificial neural networks

A hybrid approach for fault diagnosis of spur gears using Hu invariant moments and artificial neural networks

Mass Laplacian Discriminant Analysis and Its Application in Gear Fault Diagnosis

Fault diagnosis is essentially the identification of multiple fault modes. How to extract sensitive features and improve diagnostic accuracy is the key to fault diagnosis. In this paper, a new manifold learning method (Mass Laplacian Discriminant Analysis, MLDA) is proposed. Firstly, it is assumed that each data point in the space is a point with mass, and the mass is defined as the number of data points in a certain area. Then, the idea of universal gravitation is introduced to calculate the virtual universal gravitation between data points. Based on the Laplace eigenmaps algorithm, the gravitational Laplacian matrix between the same kind of data and the heterogeneous data is obtained, and the discriminant function is constructed by the ratio of the virtual gravitation between the heterogeneous data and the virtual gravitation between the similar data; the projection function with the largest discriminant function value is the optimal feature mapping function. Finally, based on the mapping function, the eigenvalues of the training data and the test data are calculated, and the softmax algorithm is used to classify the test data. Experiments on gear fault diagnosis show that this method has higher diagnostic accuracy than other manifold learning algorithms.

Vibration Signature of Gear Pump of Missing One and Two Teeth

An unexpected failure of the gear pump may cause significant effect on itsperformance and structure, consequently there will be economic losses. For thisreason, fault diagnosis in meshed gears has been the subject of serious research.Vibration analysis is one of the pumps on condition monitoring techniques thatdetermines a suitable maintenance action. Generally, an increased vibration levelcan be considered as a warning form before failure or breakdown. By measuring,processing, and analyzing the frequencies of vibration signal, it is possible todetermine both the type and severity of the defect, and hence predict the machine’sfailure. The vibration signal of a gear pump contains the signature of the defects inthe gears, and early fault detection of the gear pump is possible by analyzing thevibration signal using different signal processing techniques (Frequency Spectrum,Envelop and Power Spectrum Density). This research experimentally, shows theexternal gear pump signature for normal and faulty gear at different rotationalspeeds (1080, 1200 and 1439 RPM). The considered faults herein are one missingtooth and two missing teeth in a single gear of the pump. The Faulty gear signaturedue to missing one tooth and two missing teeth and at different rotational speedsare identified. The amplitude of vibration increases by increasing the rotationalspeed. When the two missing teeth are facing each other results in decreasing thevibration amplitude than one missing teeth size.

Domain Adaptation-Based Transfer Learning for Gear Fault Diagnosis Under Varying Working Conditions

The nonidentical and nonindependent distribution problem, caused by various working conditions of gears, inevitably makes the effectiveness of gear fault diagnosis (GFD) degrade because of insufficient or low-quality data. Although some domain adaptation learning (DAL) models exist, most of them only aim at minimizing the global distributional mean discrepancy between the source domain (SD) and the target domain (TD), while the contribution of individual data is neglected. To address this issue, a new DAL, aiming to reduce the discrepancy on extracted features under a least square support vector machine (LSSVM) framework, is studied to exploit SD signals from another working conditions or adjacent mechanical parts to assist and boost target gear fault diagnostic performance in this article. In addition, the white cosine similarity criterion is adopted to quantize the distributional weights of SD and TD feature data, and then the weights are added into a regularization function that measures the projected distributional discrepancy in the LSSVM framework. Related experiments prove that the proposed method has higher diagnostic accuracies than other models. So, this strategy is expected to be a useful tool to transfer from SD to TD, and to boost the GFD performance under various working conditions.

Study on the Fault Diagnosis Method of Scraper Conveyor Gear under Time-Varying Load Condition

Vibration signal is often used in traditional gear fault diagnosis techniques. However, the working face of the scraper conveyor is narrow, harsh and easily explosive, so it is inconvenient to obtain vibration signals by installing sensors. Motor current signature analysis (MCSA) is a fault-diagnosis method without sensor installation, which is easier to realize in the mine. Therefore, a fault diagnosis method for local gear fault, which is based on bispectral analysis (BA) of analytical signal envelope obtained by processing a stator current under time-varying load condition, is proposed in our paper. In this method, the fault frequency component is enhanced by eliminating the interference of fundamental frequency and coal flow impact. Then, the enhanced fault frequency component is extracted by BA, and a quantitative analysis of the fault strength under time-varying load is carried out from the perspective of energy. Finally, the proposed method is verified on the number HB-kpl-75 scraper conveyor reducer, and the results show that this method can successfully diagnose the failure of the scraper conveyor gear under time-varying load conditions.

Classification of stages of wear in spur gears based on wear debris morphology

This work investigates the morphological changes in the wear debris, generated during the different stages of gear wear. Wear debris are generated at the mating load-bearing tooth surfaces having relative motion. The number and size of collected wear debris provide useful information for the gear fault diagnosis.
 In the present work, both online and offline analyses of wear debris are carried out for gear fault diagnosis. In the online analysis, the oil from the gear sump is passed through online wear debris counter (to estimate the number of wear particles per minute) and particle size bin. Along with the online process, the oil samples are collected periodically, and wear debris particle images are captured using a scanning electron microscope (SEM). These images are subsequently processed to determine the parameters related to the shape, size and boundary features of the particles. The results of the modified texture in different stages of gear wear are reported. The average wear mass calculated as the actual area of the wear particles is combined with particle per minute is used as the stage’s classification.
 The combined online and offline study provides a better prediction of mild wear progression along with the information on the wear mechanisms at different stages of wear. The presence of different type of particles (ferrous and non-ferrous) points to degradation of specific components.

Extensible and displaceable hyperdisk based classifier for gear fault intelligent diagnosis

The support vector machine (SVM) method has been widely used in gear fault diagnosis. SVM is essentially a convex hull (CH) based classifier. However, using the CH to approximate the class region underestimates the true region, which leads to unsatisfactory classification performance of SVM. The use of hyperdisk (HD) based classifiers for classification can effectively avoid the above problem. But the compactness or looseness of the HD model is not adjustable and the HD based classifier has poor robustness to outliers. Therefore, to achieve better classification results in gear fault diagnosis, a classifier based on extensible and displaceable hyperdisk (EDHD) is proposed. By introducing the extensibility factor, the EDHD based classifier allows the compactness or looseness of the HD model to be adjusted to better approximate the class region. Besides, by introducing the displaceability factor, the EDHD based classifier can achieve higher robustness to outliers. Meanwhile, to extract features from unstable and nonlinear signals, a new entropy called multiscale fuzzy distribution entropy (MFDE) is proposed. Moreover, empirical mode decomposition (EMD) is used to preprocess the original signal before feature extraction, and Laplacian Score (LS) is used for feature selection after feature extraction. Finally, the EDHD and MFDE based gear fault diagnosis method is proposed. The results of gear fault diagnosis experiments fully prove the effectiveness of the proposed method.

Coupling fault transfer characteristics of fixed-axis gear crack and planetary gear missing tooth

The test signal of multistage gear transmission system is complex. When coupling faults occur in the system, it is often difficult to accurately extract all kinds of fault features. In the past studies, we found that the fault signals have modulation and coupling effects. The failure of a gear is often reflected in the power spectrum in the form of associated changes in the meshing frequency of other stages. The separation of these features will facilitate the analysis of the coupling effects of multiple faults. The transfer characteristic method provides an effective method for decomposing such signal features. In this paper, through the analysis of transfer characteristics, the coupling fault transfer characteristics and the associated influence relationship of fixed-axis gear crack and planetary gear missing tooth are revealed. The contact force signals of each pair of gears with the coupling fault are obtained by dynamic simulation. The test signals of the fixed-axis gearbox and planetary gearbox are obtained under normal and coupling fault states. According to the method of system identification, the corresponding transfer function model of each path is established. The transfer process is revealed by the analysis of transfer characteristics. The association between two kinds of faults is found, which does not exist in a single fault and is difficult to obtain by signal analysis. This study reveals the association and vibration mechanism of the coupled fault, which provides a theoretical basis for fault diagnosis of multistage gears.

New method for gear fault diagnosis using empirical wavelet transform, Hilbert transform, and cosine similarity metric

In this article, a new feature extraction method is proposed for gear fault diagnosis by combining the empirical wavelet transform, Hilbert transform, and cosine similarity metric. In the first place, a number of empirical mode components acquisitions are done, using empirical wavelet transform. Since different empirical modes have different sensitivities to fault, not all of them are needed for further analysis. Therefore, the most sensitive empirical modes are selected using the cosine similarity metric method. Hilbert transform was then used to obtain the envelope for amplitude modulation. Finally, spectral analysis using fast Fourier transform is applied on the obtained envelope. Gear test rig with gears under different fault states has revealed an effective outcome and a solid stability of this new approach. The obtained results show that our approach is efficiently able to detect and expose the gear faults signatures, that is, it highlights their frequencies and the corresponding harmonics with respect to the rotary frequency. Furthermore, this proposed method demonstrates more satisfactory and advantageous performances compared to those of fast kurtogram, or the autogram.

Fault Diagnosis Method for Wind Turbine Gearbox Based on Image Characteristics Extraction and Actual Value Negative Selection Algorithm

With the development of information theory and image analysis theory, the studies on fault diagnosis methods based on image processing have become a hot spot in the recent years in the field of fault diagnosis. The gearbox of wind turbine generator is a fault-prone subassembly. Its time frequency of vibration signals contains abundant status information, so this paper proposes a fault diagnosis method based on time-frequency image characteristic extraction and artificial immune algorithm. Firstly, obtain the time-frequency image using wavelet transform based on threshold denoising. Secondly, acquire time-frequency image characteristics by means of Hu invariant moment and correlation fusion gray-level co-occurrence matrix of characteristic value, thus, to extract the fault information of the gearing of wind turbine generator. Lastly, diagnose the fault type using the improved actual-value negative selection algorithm. The application of this method in the gear fault diagnosis on the test bed of wind turbine step-up gearbox proves that it is effective in the improvement of diagnosis accuracy.

Gear Fault Diagnosis through Vibration and Acoustic Signal Combination Based on Convolutional Neural Network

Equipment condition monitoring and diagnosis is an important means to detect and eliminate mechanical faults in real time, thereby ensuring safe and reliable operation of equipment. This traditional method uses contact measurement vibration signals to perform fault diagnosis. However, a special environment of high temperature and high corrosion in the industrial field exists. Industrial needs cannot be met through measurement. Mechanical equipment with complex working conditions has various types of faults and different fault characterizations. The sound signal of the microphone non-contact measuring device can effectively adapt to the complex environment and also reflect the operating state of the device. For the same workpiece, if it can simultaneously collect its vibration and sound signals, the two complement each other, which is beneficial for fault diagnosis. One of the limitations of the signal source and sensor is the difficulty in assessing the gear state under different working conditions. This study proposes a method based on improved evidence theory method (IDS theory), which uses convolutional neural network to combine vibration and sound signals to realize gear fault diagnosis. Experimental results show that our fusion method based on IDS theory obtains a more accurate and reliable diagnostic rate than the other fusion methods.

New intelligent gear fault diagnosis method based on Autogram and radial basis function neural network

Nowadays, fault detection, identification, and classification seem to be the most difficult challenge for gear systems. It is a complex procedure because the defects affecting gears have the same frequency signature. Thus, the variation in load and speed of the rotating machine will, inevitably, lead to erroneous detection results. Moreover, it is important to discern the nature of the anomaly because each gear defect has several consequences on the mechanism’s performance. In this article, a new intelligent fault diagnosis approach consisting of Autogram combined with radial basis function neural network is proposed. Autogram is a new sophisticated enhancement of the conventional Kurtogram, while radial basis function is used for classification purposes of the gear state. According to this approach, the data signal is decomposed by maximal overlap discrete wavelet packet transform into frequency bands and central frequencies called nodes. Thereafter, the unbiased autocorrelation of the squared envelope for each node is computed in order to calculate the kurtosis for each one at every decomposition level. Finally, the feature matrix obtained from the previous step will be the input of the radial basis function neural network to provide a new automatic gear fault diagnosis technique. Experimental results from the gearbox with healthy state and five different types of gear defects under variable speeds and loads indicate that the proposed method can successfully detect, identify, and classify the gear faults in all cases.

Exploring Sample/Feature Hybrid Transfer for Gear Fault Diagnosis Under Varying Working Conditions

AbstractUnknown environmental noise and varying operation conditions negatively affect gear fault diagnosis (GFD) performance. In this paper, the sample/feature hybrid transfer learning (TL) strategies are adopted for GFD under varying working conditions, where source working conditions are considered to help the learning of target working conditions. Here, a multiple domains-feature vector is extracted where certain insensitive features offset the adverse effects of varying working conditions on sensitive features, including time domain, frequency domain, noise domain, and torque domain. Before TL, the signed-rank and chi-square test-based similarity estimation frame is adopted to select source data sets, aiming to reduce the possibility of negative transfer. Then, the hybrid transfer model, including the fast TrAdaBoost and partial model-based transfer (PMT) algorithm, is carried out, whose weights are allocated in sample and feature, respectively. Related experiments were conducted on the drivetrain dynamics simulator, which proves that feature transfer is more suitable for low-quality source domains while sample transfer is more suitable for high-quality source domains. Compared with non-transfer strategy, transfer learning is a useful tool to solve a practical GFD problem when facing with multiple working conditions, thus enhancing the universality and application value in fault diagnosis field.

Research on Fault Diagnosis of Gearbox Based on Acoustic Emission Signal Monitoring

For the problem of large noise interference in gearbox fault diagnosis and inaccurate time-frequency resolution of analysis signals, a method for analyzing acoustic emission signals using Hilbert-Huang transform method was proposed. Firstly, use the acoustic emission method to collect the fault signal of the gearbox, which reduces the interference of environmental noise. Then perform domain of time analysis on the single, and the time-frequency analysis was performed by short-time Fourier transform, wavelet transform and Hilbert-Huang transform. The results show that the Hilbert-Huang transform method is the best to analyze the acoustic emission signal, and the frequency of the fault can be analyzed accurately, and the time-frequency aggregation is accurate, which verifies the superiority of the method in gear fault diagnosis.

Gear pitting fault diagnosis with mixed operating conditions based on adaptive 1D separable convolution with residual connection

Gear pitting fault diagnosis has always been an important subject to industry and research community. In the past, the diagnosis of early gear pitting faults has usually been carried out under single gear health state. In order to diagnose the early gear pitting faults with mixed operating conditions and reduce the number of training parameters, a new method is proposed in this paper. The proposed method uses an adaptive 1D separable convolution with residual connection network to classify gear pitting faults with mixed operating conditions. Compared to the traditional convolutional neural network, the separable convolution with residual connection network can carry out the channel convolution with point-by-point convolution to effectively reduce the number of network parameters. The residual connection can solve the representational bottleneck problem of the features in the model. Moreover, the method proposed in this paper applies the search algorithm to select better hyperparameters of the model. The raw vibration signals of the gear pitting faults at different speeds collected in a gear test rig are used to validate the effectiveness of the proposed method. The results show that the proposed method can accurately diagnose the early gear pitting faults with mixed speeds. In comparison with other machine learning models, the proposed method has provided a better diagnostic accuracy with fewer model parameters.

A Fast Multi-tasking Solution: NMF-Theoretic Co-clustering for Gear Fault Diagnosis under Variable Working Conditions

Most gear fault diagnosis (GFD) approaches suffer from inefficiency when facing with multiple varying working conditions at the same time. In this paper, a non-negative matrix factorization (NMF)-theoretic co-clustering strategy is proposed specially to classify more than one task at the same time using the high dimension matrix, aiming to offer a fast multi-tasking solution. The short-time Fourier transform (STFT) is first used to obtain the time-frequency features from the gear vibration signal. Then, the optimal clustering numbers are estimated using the Bayesian information criterion (BIC) theory, which possesses the simultaneous assessment capability, compared with traditional validity indexes. Subsequently, the classical/modified NMF-based co-clustering methods are carried out to obtain the classification results in both row and column tasks. Finally, the parameters involved in BIC and NMF algorithms are determined using the gradient ascent (GA) strategy in order to achieve reliable diagnostic results. The Spectra Quest’s Drivetrain Dynamics Simulator gear data sets were analyzed to verify the effectiveness of the proposed approach.

Learning Attention Representation with a Multi-Scale CNN for Gear Fault Diagnosis under Different Working Conditions.

The gear fault signal under different working conditions is non-linear and non-stationary, which makes it difficult to distinguish faulty signals from normal signals. Currently, gear fault diagnosis under different working conditions is mainly based on vibration signals. However, vibration signal acquisition is limited by its requirement for contact measurement, while vibration signal analysis methods relies heavily on diagnostic expertise and prior knowledge of signal processing technology. To solve this problem, a novel acoustic-based diagnosis (ABD) method for gear fault diagnosis under different working conditions based on a multi-scale convolutional learning structure and attention mechanism is proposed in this paper. The multi-scale convolutional learning structure was designed to automatically mine multiple scale features using different filter banks from raw acoustic signals. Subsequently, the novel attention mechanism, which was based on a multi-scale convolutional learning structure, was established to adaptively allow the multi-scale network to focus on relevant fault pattern information under different working conditions. Finally, a stacked convolutional neural network (CNN) model was proposed to detect the fault mode of gears. The experimental results show that our method achieved much better performance in acoustic based gear fault diagnosis under different working conditions compared with a standard CNN model (without an attention mechanism), an end-to-end CNN model based on time and frequency domain signals, and other traditional fault diagnosis methods involving feature engineering.

A fast filtering method based on adaptive impulsive wavelet for the gear fault diagnosis

The useful fault features applied for the fault diagnosis are usually overwhelmed by noise and other interference factors in rotation machinery. The impulses masked in vibration signals can represent the faults of gears or bearings in a gearbox. The key to finding impulsive components is to identify the modeling parameters (such as damping ratio, central frequency) of a transient (Morlet wavelet, Laplace wavelet), which can be used as an adaptive filter to denoise the vibration signal. However, its engineering application is limited by the time-consuming computation. In order to tackle this issue, a fast algorithm based on an adaptive impulsive wavelet is proposed to filter the fault signal so that the fault characteristic frequency can be identified. Firstly, a correlation coefficient maximum criterion is employed to find one of the optimal parameters of the impulsive wavelet. Then, the other parameter is optimized by the minimum Shannon wavelet entropy criterion. Finally, the impulsive wavelet filter with optimal parameters is applied to extract the fault characteristic frequency. Simulation signals are applied to verify the efficiency of the proposed approach, and comparison analysis is conducted as well. Further, the proposed method is applied to detect the gear fault of a gearbox. The experimental results show that the proposed method is effective with high efficiency.

Critical evaluation and comparison of psychoacoustics, acoustics and vibration features for gear fault correlation and classification

Gear fault diagnosis has gained importance in the last few decades with the focus of fault diagnosis function for maintenance purpose. This paper investigates the ability of various vibration, acoustics and psychoacoustic features to correlate and classify faults in the gearbox. This study is motivated by the process of gearbox fault diagnosis at the end of the assembly line inspection of gearbox. It is observed that during the end of assembly line inspection, the gearbox is operated on a test bench and operator takes a decision about the presence of fault by listening to the sound emitted by the gearbox. This decision is based on an operator’s judgement and past experience which involves subjectivity. Efforts are made to address this issue of end of the line inspection, by applying a scientific, objective psychoacoustic based technique which works on similar principles of listening. Experiments are performed in a laboratory by simulating four types of gear faults and vibration and acoustic signals are acquired to extract statistical features of vibration and acoustic signal along with psychoacoustic features. It is observed that different features respond to faults in different ways and changes in feature values are also dependent on loading condition. Therefore, correlating these features with faults is not simple and need intelligent techniques for correlating faults to features. This correlation of faults to features is attempted with different techniques like ANN and discriminant classifier. It is found that the psychoacoustic features have better ability to classify faults compared to acoustics and vibration statistical features which are found to be 93.02% using quadratic discriminant classifier and 95.93% using multilayer feed-forward back-propagation neural network. It is shown that the psychoacoustic based fault identification technique can be applied for the end of the assembly line inspection of gearbox.

Gear Fault Diagnosis Based on Genetic Mutation Particle Swarm Optimization VMD and Probabilistic Neural Network Algorithm

The decomposition number $K$ and penalty factor $\alpha $ in the variational mode decomposition (VMD) algorithm have a great influence on the decomposition effect and the accuracy of subsequent fault diagnosis. Therefore, a gear fault diagnosis method based on genetic mutation particle swarm optimization VMD and probabilistic neural network (GMPSO-VMD-PNN) algorithm is proposed in this paper. Firstly, the GMPSO algorithm is used to optimize the $[K,\alpha]$ parameter combination in the VMD algorithm, and the optimal $[K,\alpha]$ parameter combination of each gear fault vibration signal to be decomposed is selected. Then, the gear fault vibration signal is decomposed into several intrinsic mode functions (IMFs) by VMD, and the sample entropy value of each IMFs is extracted to form the feature vector of subsequent fault diagnosis. Finally, the characteristic vector of gear fault vibration signal is input into PNN model, and gear fault is accurately classified. By comparing with fixed parameter VMD algorithm, empirical mode decomposition (EMD) and complete ensemble empirical mode decomposition adaptive noise (CEEMDAN) algorithm, the superiority of this method in gear fault diagnosis is verified. Therefore, the GMPSO-VMD-PNN algorithm proposed in this paper has certain application value for gear fault diagnosis.